The New Educational Imperative...The New Educational Imperative: Improving High School Computer Science Education Final Report of the CSTA Curriculum Improvement Task Force February

The New Educational Imperative:Improving High School Computer Science EducationUsing worldwide research and professional experience to improve U. S. SchoolsBy CSTA Curriculum Improvement Task Force

The New Educational Imperative:Improving High School Computer Science Education

Using worldwide research and professional experience to improve U. S. Schools

By The CSTA Curriculum Improvement Task Force

The New Educational Imperative:Improving High School

Computer Science Education

Final Report of the CSTACurriculum Improvement Task Force

February 2005

Task Force Chair

Chris StephensonCSTA Executive Director

Committee Members

Judith Gal-EzerOpen University of Israel

Bruria HabermanHolon Academic Institute of Technologyand The Weizmann Institute of Science

Anita VernoBergen Community College

Computer Science Teachers AssociationAssociation for Computing Machinery

1515 Broadway, 17th FloorNew York, New York 10036

Copyright © 2006 by the Association for Computing Machinery, Inc (ACM).Permission to make digital or hard copies of portions of this work for personal orclassroom use is granted without fee provided that the copies are not made ordistributed for profit or commercial advantage and that copies bear this notice and thefull citation on the first page. Copyrights for components of this work owned by othersthan ACM must be honored. Abstracting with credit is permitted.

To copy otherwise, to republish, to post on servers or to redistribute to lists, requiresprior specific permission and/or a fee. Request permission to republish from:Publications Dept. ACM, Inc. Fax +1-212-869-0481 or E-mail [email protected] other copying of articles that carry a code at the bottom of the first or last page,copying is permitted provided that the per-copy fee indicated in the code is paidthrough the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923.

ACM ISBN: # 1-59593-335-2ACM Order Number: # 999064

Additional copies may be ordered prepaid from:

ACM Order Department Phone: 1-800-342-6626P.O. Box 11405 (U.S.A. and Canada)Church Street Station +1-212-626-0500New York, NY 10286-1405 (All other countries)

Fax: +1-212-944-1318E-mail: [email protected]

This material is based upon work supported by the National Science

Foundation under Grant No. 0455403. Any opinions, findings, and conclusions

or recommendations expressed in this material are those of the author(s) and

do not necessarily reflect the views of the National Science Foundation.

Letter from the CSTA President

In the face of a global economy that rewards innovation and technological supremacy,many countries, including the United States, are looking to education to ensure theircontinuing competitiveness.

With the specific goal of keeping the United States technologically competitive infields such as nanotechnology, supercomputing, and bioinformatics, the United Stateshas committed to training thousands of high school teachers to lead advancedplacement math and science courses.

Computer science as an academic discipline provides the knowledge and skillfoundation for all of these technological advances.

Computer science education in the United States is at a critical juncture. Thenumber of students pursuing computer science as a career choice or course of studyhas dropped precipitously at all levels. The number of high school students taking theAdvanced Placement computer science exam, for example, has declined almost 20%over the past three years. In addition, the lack of national curriculum standards andconsistent and rational teacher certification requirements continue to hamper ourability to ensure that our students are adequately prepared to compete in thisincreasingly technological world.

In short, policy makers at the local, state, and national levels must prepare for a21st century technological workforce by giving those workers the computer scienceknowledge base necessary for success in these computing-intensive fields.

The warning signs are unmistakable. The solution is clear. Unless we act quicklyand decisively to remedy the disconnect between our national technological goals andcomputer science education at the high school level, the United States will soon facean educational, competitive, and economic crisis.

I urge you to read closely and carefully this final report of the CSTA CurriculumImprovement Task Force and heed its sobering message. Improving high schoolcomputer science education is indeed a national educational imperative.

Regards,

Robb CutlerChairperson

Computer Science Teachers Association

Acknowledgments

The CSTA Curriculum and Improvement Task Force would like to thankthe following organizations and individuals. First, we would like to thankthe National Science Foundation for their generous support of this projectand for their understanding of the importance of improving computerscience education at the high school level.

We would also like to thank Allen Tucker and the ACM K–12 CurriculumCommittee for their ground-breaking work on the ACM Model Curriculum

for K–12 Computer Science and Jane Margolis and Joanna Goode for sharingthe important work that they are doing as part of the Computer ScienceEquity Collaborative.

We are especially grateful to the wonderful international educators whotook part in our panel on creating and implementing an effectivecurriculum, specifically Michael Chiles, Judith Gal-Ezer, and Jackie Martin.

Thanks are also due to our wonderful designer Bob Vizzini, and to ourexacting copy editor, Kate Conley.

Finally, to all of the staff and volunteer leadership of ACM, who broughtCSTA into being and continue to support us every day.

c o n t e n t sEXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

1.0 CHAPTER 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.1 What is Computer Science Anyway?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.2 What is Happening in Computer Science Education? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

1.2.1 The Curriculum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

1.2.2 Students and Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

1.2.3 The Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.3 Is Real Change Possible? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.4 How Can This Document Help? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

1.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23


2.1 Computer Science Education in High Schools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2.1.1 United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2.1.2 Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

2.1.3 Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

2.2 Who Are We, and How Should We Portray Ourselves? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.2.1 An Inconsistent Self-image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.2.2 An Incorrect Public Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.2.2.1 Computer Science Equals Programming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2.2.2.2 Computer Science Equals Computer Literacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2.2.2.3 Computer Science is a Tool for Studies in Other Disciplines . . . . . . . . . . . . . . . . . . . . . . 28

2.2.2.4 Computer Science is Not a Scientific Discipline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.2.2.5 Computer Science is a Male Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.2.3 Challenging Student Preconceptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2.3 High School Computing Studies and Academic Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.4 What Should We Teach and How? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.4.1 Guiding Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.4.2 Teaching the Key Concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

2.4.2.1 Basic Computing Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

2.4.2.2 Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.4.2.3 Recursion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.4.2.4 Abstraction and Abstract Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.4.2.5 Non-determinism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

2.4.2.6 Correctness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

2.4.2.7 Algorithm Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

2.4.2.8 Program Testing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

2.4.3 Pedagogical Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

2.4.3.1 Problem-Solving Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

2.4.3.2 Learning Different Programming Paradigms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

2.4.3.3 Software Design and Project Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

2.5 Whom Should We Teach? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

2.6 Exemplary Teachers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

2.6.1 Pre-service Teacher Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.6.1.1 Technical Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.6.1.2 Pedagogical Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.6.1.3 Socio-cultural Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

2.6.2 On-going Professional Development for Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

2.6.3 The Role of Professional Associations in Supporting Teachers. . . . . . . . . . . . . . . . . . . . . . . . 43

2.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

2.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44


3.1 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3.1.1 Method of Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3.1.2 Employing a Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

3.1.3 Understanding Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

3.1.3.1 Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

3.1.3.2 Israel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

3.1.3.3 Scotland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3.1.3.4 South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

3.1.3.5 United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

3.2 Exploring Curriculum Development and Implementation Using Frank’s Framework . . . . . . . . . . . . . . . 57

3.2.1 Policy Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

3.2.2 Theoretical Adequacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

3.2.3 Empirical Validity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

3.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

3.4 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63


4.1 How Do We Design A Better Curriculum?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

4.1.1 Ten Core Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

4.1.2 Key Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

4.1.3 Example of a Comprehensive Curriculum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

4.2 How Do We Ensure Successful Implementation? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.2.1 Policy Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.2.2 Theoretical Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

4.2.3 Empirical Validity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

4.2.4 Example of a Comprehensive Implementation Plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

4.3 How Do We Improve How Teachers Teach? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.3.1 Five Qualities of Exemplary Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.3.2 Seven Systemic Changes Required to Improve Teaching. . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

4.4 Call To Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

4.4.1 Federal Government Policy Makers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

4.4.2 State Government Policy Makers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

4.4.3 School District Policy Makers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

4.4.4 School Principals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

4.4.5 Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

4.4.6 University and College Faculty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

4.4.7 Business and Industry Leaders. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

4.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

WHAT IS THE COMPUTER SCIENCE TEACHERS ASSOCIATION? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

THE ECONOMIC AND EDUCATIONIMPERATIVE

The United States, a nation once proud of itsleadership in education, is sitting quietly on thesidelines while other countries make improvementsto ensure their high school graduates will be ready tomeet the demands of tomorrow’s high-tech society.Countries around the world that once looked to theUnited States for guidance when planning new andinnovative curricula are now taking the lead incomputer science education. While other countries arerequiring a computer science course for high schoolstudents just as they require math or biology, highschool computer science education in the United Statesis disappearing. In light of the anticipated shortage ofqualified candidates for the 1.5 million computer andinformation technology jobs by 2012, this lack ofengagement with issues relating to computer scienceeducation is shortsighted and potentially disastrous.

DEFINING COMPUTER SCIENCE

Unlike other more static disciplines, computer science isconstantly being reshaped. New thinking and newtechnologies continue to expand our understandingof what computer scientists can and need to know.This has resulted in considerable debate about a singledefinition of computer science. The ACM Model

Curriculum for K–12 Computer Science, however, providesthe most appropriate definition of computer science forhigh school educators. Computer science, it argues, is

neither programming nor computer literacy. Rather, itis “the study of computers and algorithmic processesincluding their principles, their hardware and softwaredesign, their applications, and their impact on society”(Tucker, 2003). Computer science therefore includes:

• programming, • hardware design, • networks, • graphics, • databases and information retrieval, • computer security, • software design, • programming languages, • logic, • programming paradigms, • translation between levels of abstraction,

artificial intelligence,• the limits of computations (what computers

can’t do), • applications in information technology and

information systems, and• social issues (Internet security, privacy,

intellectual property, etc.).

HIGH SCHOOL COMPUTER SCIENCEEDUCATION IN THE UNITED STATES

Computers have infiltrated all areas of society, andthere is now a clear link between technology,innovation, and economic survival. In light of this,one would expect a move within our society tosupport and standardize computer science education.

The New Educational Imperative:Improving High School Computer Science Education

Final Report of the CSTA Curriculum Improvement Task Force

February 2005

Executive Summary

13

Yet, no national K–12 computer science curriculumexists. Lack of leadership on high school computerscience education at the highest legislative and policylevels has resulted in insufficient funding forclassroom instruction, resources, and professionaldevelopment for computer science teachers. Inaddition, complex and contradictory teachercertification requirements as well as salaries thatcannot possibly compete with industry make itexceedingly difficult to ensure the availability ofexemplary computer science teachers.

In the face of confusing definitions of computerliteracy, information fluency, and the various sub-branches of computer science itself, many schoolshave lost sight of the fact that computer science is ascientific discipline and not a “technology” thatsimply supports learning in other curriculum areas.Computer science is not about point and click skills.It is a discipline with a core set of scientific principlesthat can be applied to solve complex, real-worldproblems and promote higher-order thinking. Inshort, knowledge of computer science is now asessential to today’s educated student as any of thetraditional sciences.

WHAT THE RESEARCH REVEALS

The body of research from around the world relatingto high school computer science education indicatesthat learning computer science provides direct benefitsto students. While there are distinct differencesbetween how various countries implement their highschool computer science programs, a growing number of countries already require computer scienceeducation of all high school students.

The primary limitation of the current body ofresearch is that it is too narrowly focused onprogramming and therefore does not give anappropriately broad view of the discipline. Despitethis limitation, however, the research provides vitalinformation regarding students’ misconceptionsabout computer science (for example, that it is justabout programming or that it is a “male” field). It

also provides critically important insights about theunderlying principles of computer science education.

• Students should acquire a broad overview ofthe field to construct a comprehensive picture of computer science as a discipline.

• Students must understand not only thetheoretical underpinnings of the discipline butalso how that theory influences practice.

• Computer science instruction should focus onproblem solving and algorithmic thinking.

• Concepts should be taught independent ofspecific applications and programminglanguages.

• Students should be taught what will beexpected from them in the “real world,”specifically, what is actually required to writeand maintain computer programs and largesoftware systems. Computer Science should betaught using real-world applications rather than specialized educational tools.

• Computer science instruction should includeintegrative and interdisciplinary knowledge.

• High school students should be exposed toadvanced topics of computer science (e.g.computational models, modeling, and parallelcomputing) to enable them to become familiarwith some of the theoretical aspects of computerscience.

• Students must recognize recurring concepts and principles such as abstraction, complexity,modularity, and reusability.

• Programming should be taught in a broadsense, covering not only the coding act itself, but also the design of the algorithmsunderlying the programs and, to some extent,considerations of correctness and efficiency.

• The curriculum should be designed to addressthe under-representation of women andminority students in computer science.

• Teachers should motivate students to endurethe rigors of traditional computer scienceprograms by engaging students with exciting,accessible, and leading-edge courses.

• Teaching and learning activities should be

14

15

designed to treat common misconceptions ofthe essence of computer science.

The research also indicates that student success incomputer science is predicated on the teacher’sknowledge of the discipline and ability to actively engagestudents in their own learning, and identifies several key factors that help to ensure exemplary teaching.

• High school computer science teachers musthave a thorough formal background incomputer science.

• Pre-service teacher education must prepareteachers to better employ general pedagogicalprinciples as well as teaching methods in thecontext of computer science education.

• To be best qualified, computer science highschool teachers should be certified and offeredcourses in computer science education inaddition to regular computer science courses.

• Classroom teachers require on-going access to appropriate and relevant professionaldevelopment opportunities that allow them to master new technologies, implement newcurricula, and constantly improve their teaching.

• Teachers should become part of a collaborativecomputer science teachers' community ofpractice by joining local and national associationsthat provide useful resources and support theirongoing learning and leadership development.

WHAT WE CAN LEARN FROM OTHERCOUNTRIES

Looking at the experiences of other countries thathave designed and implemented a high schoolcomputer science curriculum can provide valuablelessons as to the factors that must be in place toensure that our efforts to improve computer scienceeducation are successful. An examination of theexperiences of an international panel of computereducators from Canada, Israel, Scotland, South Africa,and the United States support the argument that suchcurriculum initiatives will be successful only to the

extent to which they meet the following criteria:• There is a link between the outcome required

and the strategies used.• Change is driven by real learning needs and not

politically manufactured needs. • Educational change must be seen in the context

of larger social and economic forces.• All of the stakeholders must agree to the need

for change and on the strategies put in place to achieve it.

• Change requires the commitment of adequateresources through all phases of the design,implementation, and testing of the newcurriculum.

• Change is a long-term process, not a short-termintervention.

ACTIONS TO IMPROVE HIGHSCHOOL COMPUTER SCIENCEEDUCATION

Education in the United States is at a crossroads. Wecan either commit to ensuring that our students havethe skills to be participants and innovators in thistechnological world or resign ourselves to adecreasing international presence in the globaleconomy. We know that the world continues tochange and that many of the changes we face arerelated to the burgeoning possibilities andconsequences of our growing dependence upon andinterrelation with computing technology. Maintainingour ability to meet present and future challengesrequires us to acknowledge computer science as acore element of all STEM (science, technology,engineering, and mathematics) initiatives. Sustainingour technological and innovative edge in thisincreasingly global economy requires a multi-level,nationwide commitment to improving computerscience education at the high school level. We mustmake this commitment if our schools are to continueto provide relevant education and our society is tocontinue to solve problems on the cutting edge ofinnovation. Here is what we need to do:

• We must begin with a definition of computerscience education that recognizes computing as a scientific discipline and encompasses abreadth of knowledge and skills.

• We must implement a national computerscience curriculum for high schools. Thiscurriculum must be principle-based, mustaddress core content and key skills, and mustincorporate appropriate strategies to reach andteach our students.

• We must support the implementation of thisnew curriculum with a plan that includes arealistic timeframe and the provision of theresources required to achieve it.

• We must ensure that computer science is taughtby exemplary teachers who have the requisiteknowledge to teach the curriculum and who continue to upgrade their technical andteaching skills throughout their careers.

Schools alone cannot achieve these outcomes. Thisissue requires a national vision, supportive action,and commitment at all levels of the political andeducational systems. The final chapter of thisdocument therefore provides practical, achievablesuggestions for ways in which

• federal government policy makers, • state government policy makers, • school district policy makers, • school principals, • teachers, • university and college faculty, and • business and industry leaders

can work together to achieve long-term, systemicimprovements to high school computer scienceeducation.

ORGANIZATION OF THIS DOCUMENT

This document has been designed to address anumber of different aspects of high school computerscience education. Chapter One (High School Computer

Science in the United States) provides an overview of thecurrent state of high school computer scienceeducation. This chapter discusses the ways in whichthe United States has failed to support and standardizecomputer science education despite the infiltration of computing technology into all areas of society.

Chapter Two (Challenges and Perspectives in High

School Computer Science Education: A Retrospective

Literature Review) provides an extensive review ofinternational research literature in the area of high schoolcomputer science education, particularly as it relates tocourse content and pedagogy. This chapter highlights thecontinuing focus on programming despite more recenteducational models that suggest that students need abroader theoretical and skills foundation to ensure thatthey are prepared for college and the workplace.

Chapter Three (Using Frank’s Framework to Explore

the Development and Implementation of High School

Computer Science Curricula in Five Countries) looks atthe development and implementation of high schoolcomputer science curricula in five countries. Using aspecific framework for critiquing educational policyreform and implementation, this chapter examinesthe experiences of curriculum developmentspecialists from Canada, Israel, Scotland, SouthAfrica, and the United States as presented at a panelsession at the 2005 National Educational ComputingConference. This chapter provides valuable insight onkey implementation issues such as funding,professional development, and time lines.

Finally, Chapter Four (Strategies for the Successful

Design and Implementation of a National High School

Computer Science Curriculum) proposes a set of practical, achievable strategies andrecommendations that could profoundly improvehigh school computer science education in the UnitedStates. These changes, requiring the support andcommitment of federal and state policy-makers,school district leaders, classroom teachers, universityand college faculty, and business and industry, would help provide students with the scientificknowledge base required for success in the 21st

century and would ensure that the United Statesremains competitive in the global economy.

16

1.0 INTRODUCTION

Although an increasing number of labor andeconomic specialists are beginning to expressconcern about the direct and pressing link betweentechnology and innovation and national economicsurvival, very few school administrators andeducational policy leaders understand the profoundneed for computer science education at the highschool level. Despite the need for students toincorporate the foundational computer science skillsthat foster an understanding of the essentialtechnologies found in almost every industry today,in most high schools in the United States, there islittle recognition of computer science as a scientificdiscipline distinct from mathematics or fromtechnology training (National Research Council,1999; Tucker, 2003)

In today’s scientific and economic globalcommunities, computer science is helping to push the boundaries of what we know and what we can do. In areas such as nanotechnology andbioinformatics, computer scientists are addressingkey questions such as:

• Why do people become ill?• How can we do better at feeding the people of

the world?• How can we ensure our own national security

and the safety of our people?

As a result of these kinds of scientific breakthroughs,the U.S. economy is expected to add 1.5 millioncomputer- and information-related jobs by 2012.

The problem, however, is that our educationsystem is not producing enough highly skilled peopleto fill these critical positions. Current projections

indicate that this country will have only half thatmany qualified graduates. As a result, if we do notaddress these issues immediately and vigorously atthe national level, we face long-term skills shortagesthat will cripple both academic computing andindustry (International Technology Association ofAmerica, 2002; Sargent, 2004). As a result, the UnitedStates will be seriously compromised in its ability touphold its leadership position in computing,communications, information science, andengineering. In addition, we will be unable tocapitalize on current and future innovations that arealready providing untapped economic and socialopportunities.

1.1 WHAT IS COMPUTER SCIENCEANYWAY?

One of the challenges we face when discussingcomputer science education is that the field ofcomputer science seems to evolve so quickly that it isdifficult even for computer scientists to clearly defineits contents and delimit its boundaries. AsShackelford (2005) noted in his presentation Why can’t

smart people figure out what to do about computing

education?, the landscape of computing continues toevolve and trying to figure out what students need tolearn is like trying to hit a moving target. While wedo know that computing now provides theinfrastructure for how we work and communicateand that it has redefined science, engineering, andbusiness, it is still poorly understood.

Even at the university level, where computerscience has long been recognized as a key educationaland career field, it has resisted a single definition or

CHAPTER ONE

HIGH SCHOOL COMPUTER SCIENCE EDUCATIONIN THE UNITED STATES

1717

application. As Shackelford indicated, prior to 1990the discipline was defined as Computer Science iftaught as part of the traditional Arts and Sciencecurriculum and as Information Systems if taught aspart of the Business curriculum. Since that time,however, it has burgeoned into a number of distinctareas, each of which involves its own approach toboth theory and application. These areas includeComputer Engineering, Software Engineering, andInformation Technology. In terms of university degreeprograms, the post-1990s are typified by intersectingsub-disciplines including:

• Electrical Engineering,• Computer Engineering,• Software Engineering,• Computer Science,• Information Technology, and• Information Systems

that focus to varying degrees on hardware, software,and organizational needs.

For high school educators, though, the mostprofound confusion arises when trying to distinguishbetween the three most common kinds of computingeducation typical of the 9–12 grade levels. While eachof these areas has been known by various names, forthe purposes of this discussion we define them as:

• Educational Technology• Information Technology, and• Computer Science.

In general, Educational Technology can be defined asusing computers across the curriculum, or morespecifically, using computer technology (hardware andsoftware) to learn about other disciplines. For example,the science teacher may use a computer-basedsimulation program to provide students with a betterunderstanding of specific physics principles, or anEnglish teacher may use word-processing software tohelp students improve their editing and revision skills.

Tucker, Deek, Jones, McCowan, Stephenson, andVerno (2003) defined Information Technology as “theproper use of technologies by which peoplemanipulate and share information in its various

forms” (p. 6). While Information Technology involveslearning about computers, it emphasizes thetechnology itself. As Shackleford (2005) noted,Information Technology specialists “assumeresponsibility for selecting appropriate hardware andsoftware products, integrating those products withorganizational needs and infrastructure, andinstalling, customizing and maintaining thoseresources” (p. 22). Information Technology courses,therefore, focus on:

• installing and administering computernetworks,

• installing, maintaining, and customizingsoftware,

• managing e-mail systems, • designing web pages, and• developing multimedia resources and other

digital media.

Computer Science, on the other hand, spans a widerange of computing endeavors, from theoreticalfoundations to robotics, computer vision, intelligentsystems, and bioinformatics. According toShackelford, for example, the work of computerscientists is concentrated in three areas:

• designing and implementing software, • developing effective ways to solve computing

problems, and• devising new ways to use computers.

Tucker et al. (2003) has also offered a definition ofcomputer science that has direct relevance to highschool computer science education. They defined thediscipline as follows: “Computer science (CS) is thestudy of computers and algorithmic processes,including their principles, their hardware andsoftware designs, their applications, and their impacton society” (p. 6). They argue that in order to fulfillthis definition, K–12 computer science curricula mustinclude the following elements:

• programming, • hardware design, • networks, • graphics,

18

• databases and information retrieval, • computer security, • software design, • programming languages, • logic, • programming paradigms, • translation between levels of abstraction, • artificial intelligence, • the limits of computation (what computers

can’t do), • applications in information technology and

information systems, and • social issues (Internet security, privacy,

intellectual property, etc.).

Citing the conclusion of the National ResearchCouncil (1999) that a basic understanding of all thesetopics is now an essential ingredient to preparinghigh school graduates for life in the 21st century,Tucker et al. (2003) further argued that the goals of aK–12 computer science curriculum are to:

• introduce the fundamental concepts ofcomputer science to all students, beginning atthe elementary school level,

• present computer science at the secondaryschool level in a way that would be bothaccessible and worthy of a curriculum credit(e.g., math or science),

• offer additional secondary-level computerscience courses that will allow interestedstudents to study it in depth and prepare themfor entry into the work force or college, and

• increase the knowledge of computer science forall students, especially those who are membersof underrepresented groups.

Two other terms that often appear in discussions ofcomputing education are Information Technology Literacy

and Information Technology Fluency (National ResearchCouncil, 1999). As Tucker et al. (2003), indicate:

Whereas IT literacy is the capability to use today’s

technology in one’s own field, the notion of ITfluency adds the capability to independently learn

and use new technology as it evolves …

throughout one’s professional lifetime. Moreover,IT fluency also includes the active use ofalgorithmic thinking (including programming) tosolve problems, whereas IT literacy is morelimited in scope (p. 6).

1.2 WHAT IS HAPPENING INCOMPUTER SCIENCEEDUCATION?

At present, all evidence points to a crisis in computerscience education at the high school level. This crisisis most clearly manifested in the decreasing numberof computer science courses being offered to studentsand the resulting drop in enrollments in computerscience programs. Course enrollments are droppingat both the secondary and post-secondary levels(Taulbee, 2003), and the number of young womenand minority students studying computing is at anall-time low (The College Board, 2005). TheComputer Science Teachers Association (CSTA,2005a) has identified a number of factors thatcontribute to this situation, including:

• the lack of a national high school curriculum for computer science education,

• the chronic under-funding of computer scienceprograms in high schools,

• the absence of standards for certification ofcomputer science teachers,

• a shortage of professional developmentopportunities that would allow teachers todevelop and keep their technical andpedagogical skills current,

• the inability of school districts to attract ormaintain highly qualified teachers in the face of salary and benefit competition fromindustry, and

• the lack of understanding on the part ofstudents, parents, guidance counselors, andteachers about computer science in general,how it differs from other areas of computerstudy, and its newly developing careeropportunities.

19

1.2.1 The Curriculum

In recent decades, computers have come to occupy apivotal place in our work, personal, and home lives.Accordingly, the field of computer science hasexpanded to encompass the technical, innovative,conceptual, and psychological ramifications of thepresence of the computer at work and at home. AsTucker et al. (2003) noted, however, very littleattention has been paid to meeting the need for awell-structured, conceptually based high schoolcomputer science curriculum at the time when moststudents are likely to be building the conceptualframeworks of their intellectual development andexploring the options for their life’s work.

To date, efforts to standardize the study ofcomputer science in U.S. high schools have beenidiosyncratic and uncoordinated. While othercountries have designed and implemented nationalcomputer science education programs in order tobetter prepare their students for the increasinglycompetitive global economy, attempts to bringcoherent computer science education into U.S. highschools failed to address the need to instill a fluentunderstanding of algorithmic thinking and problemsolving, and to provide exposure to softwaredevelopment using programming skills in U.S.students. While various researchers, institutions, andorganizations have attempted to define anddisseminate new computer science curricula, theseefforts have been typified by:

• a persistent shortage of support resources, • a lack of initial and sustaining funding,• an absence of on-going, classroom-relevant

teacher training, and• the relentless “projectitis” (Fullan, 2002) that

prevents any one educational initiative fromreceiving anyone’s attention for very long.

In addition, the extreme localization of educational(especially curriculum) policy and decision making(which is often relegated to individual schools) and acomplete lack of committed engagement by thefederal government (despite the growing awareness

of the impending skills shortage crisis) have resultedin an abject failure to grapple with this pressingeducational issue.

1.2.2 Students and Courses

In 2004, CSTA (2005) surveyed 14,000 high schoolteachers who defined themselves as computer science,computer programming, or AP computer scienceteachers. The study’s purpose was to create a snapshotof the current state of computer education in theUnited States, and it revealed some disturbing trends.Despite the growing importance of computerknowledge and skills, only 26% of responding schoolsrequired students to take a computer science course,and only 40% of responding schools even offered anAdvanced Placement (AP) computer science course.Among students who took an introductory levelcomputer science course, on average only 32% werefemale. Among students taking the AP computerscience course, the number of females dropped to 23%.

CSTA’s research is further supported by TheCollege Board (2005), which reported that from2002–2004 for example, while AP exam taking inother disciplines rose overall by 19%, the number ofstudents taking the computer science A examdropped by 8%, and the number taking the computerscience AB exam decreased by 19%. In addition, in2004 while 56% of the AP test takers overall werefemale, among CS AB test takers, only 11% werefemale (a drop from the 1999 data of 17%), and only6% were from under-represented minorities.

When asked to explain these increasinglyproblematic enrollment results, teachers noted that thegreatest impediment to students taking high schoolcomputer science courses was not the perceiveddifficulty of the subject matter or even the perceptionthat computer science was a “geeky” course, butrather the lack of time in the students’ schedules. Asthe number of mandated courses high school studentsare required to take and competition for acceptance toprestigious university programs increase, studentshave fewer and fewer opportunities to study

20

computer science because these courses are electiverather than mandatory courses.

Considerable anecdotal evidence also suggeststhat perceptions concerning career opportunities areplaying a role in diminishing student interest incomputer science courses. Teachers are reporting thatan increasing number of students and their parentsbelieve (based upon media stories rather thanmarketplace realities) that computer science is nolonger a source of rewarding and varied careeropportunities. In the absence of solid educational andcareer information, students are therefore beingactively dissuaded from computer science courses andfrom considering careers in the high tech industry.

1.2.3 The Teachers

Students are also not the only ones affected by thecurrent crisis in computer science education. A 2002study of 4,000 U.S. high school computer scienceteachers conducted by the Association for ComputingMachinery (ACM, 2005) revealed that 89% of highschool computer science teachers indicated that theyexperience a sense of isolation and a lack of collegialsupport within schools and school districts. Notingthat rapid changes to both technology and teachingprovide significant challenges, the teachers alsoindicated that their greatest professionaldevelopment need was actually finding time for theirown on-going learning (CSTA, 2005). They alsoindicated that the on-going battle for adequateresources, the lack of acceptance and understandingof computer science as a scientific discipline distinctfrom technology training, and increasing budget cutsin these times of fiscal restraint deterred manyinterested and qualified teachers from teachingcomputer science.

One additional persistent and often ignoredproblem is that there is little motivation for thosewith the requisite skills to pursue a career teachinghigh school computer science. In most jurisdictions,teachers’ salaries are so low and the workingconditions so unpleasant when compared to other

career fields, that it is impossible for education toattract individuals with the appropriate skills. Evenfor those who are considering a second career incomputer science education and for whom salaryissues may not be a primary factor, the lack ofconsistent and readily available informationconcerning certification requirements make it almostimpossible to determine how one should go aboutpreparing for such a career change.

1.3 IS REAL CHANGE POSSIBLE?

Education is a highly complex and multifacetedenvironment within which an almost unimaginablenumber of policies, methodologies, strategies,ideologies, and bureaucracies interact (Bauch &Goldring, 2000). And while at its most basic level, thetransition of knowledge might appear as a simpleprocess, effective teaching depends significantly uponthe contexts within which teachers work—departmentand social organization and culture, professionalassociations and networks, community educationalvalues and norms, and secondary and higher educationpolicies (McLaughlin, 1992). These contexts determinenot only what is taught and how it is taught, but alsothe value placed upon a given academic discipline’splace within the larger school curriculum and theresources that are made available to instructors.

At present, many of the changes that occur withineducation are the result of small-scale bottom-upinnovation at the individual school level. Usually,such change is structural and superficial and rarelytranslates into large-scale innovation at the system ornational levels (Ballantyne, McLean, & Macpherson,2003). In effect, such changes tinker with, but do nottransform, the educational culture. They do not leadto lasting change because they do not change whatpeople in the organization value and how they workto accomplish it (Fullan, 2002).

To date many initiatives have been launched andmuch money has been spent on proposedinnovations that have resulted in very little changeover time, primarily because these initiatives have

21

failed to grapple with the multiplicity of demandsand expectations placed upon schools and teachers.The reasons such undertakings fail aremultitudinous, but such failures most commonlyresult from some or all of the following factors.

1. Too many messages. The education system isbombarded with demands for changes frommultiple external stakeholders in allcurriculum areas, resulting in too muchcompetition for attention and resources.Everyone has an answer, but no one really seeshis/her own suggestions within the holisticframework of education.

2. Poor communication of Return on Investment

(ROI). Outside interests urge and expectschools to change but fail to consider oreffectively communicate what the ROI will befor students, teachers, and administrators.

3. The barrier of vocabulary. The field of educationhas its own language where seemingly simplephrases (such as assessment-based learning)can be a minefield for unsuspecting outsiders.

4. A lack of respect for K–12 educators. Pre-collegeeducators are often treated with disdain bytheir university colleagues, by industry, and bypoliticians. Even those most interested inbuilding partnerships with K–12 frequentlyundermine their own good intentions withtheir profound lack of understanding aboutwhat it means to teach in a K–12 environment.

5. Frustration with past failures. Over the years,industry representatives have becomeincreasingly frustrated by the frequency withwhich they have contributed to initiatives withlittle or no lasting impact or return. Theproblem is that those promising to ensure thesechanges do not have a foundationalunderstanding of issues of change or theefficacy of change agents in the K–12 arena.

Systemic educational change demands a profoundunderstanding of and engagement with the culture inwhich the change must take place. While achieving thelevel of change required to address the current issues

in high school computer science education in U.S.schools may seem like an impossible challenge, timehas shown that when sufficient recognition, resources,and long-term commitment are brought to bear on anational challenge, the unthinkable can be achieved.

1.4 HOW CAN THIS DOCUMENTHELP?

The purpose of this document is to set forth some ofthe important factors that must be considered inorder to achieve a substantial improvement to highschool computer science education, an improvementthat must occur if we are to meet the shiftingknowledge needs of our citizens and so remain acompetitive force within the global economiccommunity.

To achieve this, we must begin by acknowledgingthat other nations may have knowledge they canshare with us, that will help us ensure that we useour resources wisely and do not repeat mistakes thatothers have already made. The second chapter of thisdocument, therefore, provides a comprehensivereview of the international body of research relatingto high school computer science education. Itexamines the issues that have been raised, thechallenges that have been identified, and thesolutions that have been proposed.

The third chapter provides an analysis of fivecountries, including the United States, in whichefforts to design and implement a comprehensivehigh school computer science curriculum have beenundertaken. This chapter demonstrates that definingthe curriculum is really just a first step and that thefactors that determine the extent to which thatcurriculum is actually implemented may in fact bemore important than the curriculum itself.

The last section draws from all of the previoussections of the document to provide a comprehensiveset of recommendations and strategies that willsignificantly increase the likelihood that our nationalefforts to improve high school computer scienceeducation will achieve long-term success.

22

1.5 REFERENCES

ACM K–12 Task Force. (2005). ACM K–12 Task Force

Curriculum Survey Results. Retrieved December 22,2005, from http://csta.acm.org/Research/sub/TuckerSurveyResults.html

Ballantyne, R., McLean, S. V., & Macpherson, I.(2003). Knowledge and skills required for creating a

culture of innovation: Supporting innovative teaching

and learning practices. Brisbane: Faculty ofEducation, Queensland University of Technology.

Bauch, P. A., & Goldring, E. B. (2000). Teacher workcontext and parent involvement in urban highschools of choice. Educational Research and

Evaluation, 6(1), 1–23.

Computer Science Teachers Association (2005a).Strategic Plan. Retrieved December 22, 2005, fromhttp://csta.acm.org/About/sub/CSTAStrategicPlan.html/.

Computer Science Teachers Association. (2005).Results of the national secondary computer science

survey. Retrieved October 16, 2005 fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html

Fullan, M. (2002). The change leader. Educational

Leadership, 59(8), 16–21.

Information Technology Association of America.(2002). Bouncing Back: Job skills and the continuing

demand for IT workers. Arlington, VA: InformationTechnology Association of America.

McLaughlin, M. W. (1992). How district communitiesdo and do not foster teacher pride. Educational


National Research Council Committee onInformation Technology Literacy. (1999). Being

fluent with information technology. Washington, DC:National Academy Press.

Sargent, J. (2004). An overview of past and projectedemployment changes in the professional IToccupations. Computing Research News, 16(3), 1–22.

Shackelford, R. (2005). Why can’t smart people figure out

what to do about computing education. Presentation atthe CSTA/ISTE Computer Science and InformationTechnology Symposium, Philadelphia, PA.

Taulbee, O. E. (2003). 2002–2003 Taulbee study:

Undergraduate enrollments drop; Department growth

expectations moderate. Retrieved December 22,2005, from http://www.cra.org/statistics/

The College Board. (2005). AP report to the nation.

Retrieved December 22, 2005, fromhttp://www.collegeboard.com/about/news_info/ap/2005/index.html

Tucker, A., Deek, F., Jones, J., McCowan, D.,Stephenson, C. and Verno, A. (2003). A model

curriculum for K–12 computer science: Report of the

ACM K–12 Education Task Force Computer Science

Curriculum Committee. New York, NY: Associationfor Computing Machinery.

23

CHAPTER TWO

CHALLENGES AND PERSPECTIVES IN HIGH SCHOOLCOMPUTER SCIENCE EDUCATION:

A RETROSOPECTIVE LITERATURE REVIEW

24

2.0 INTRODUCTION

Computer science is considered a young and arapidly developing discipline, and its “extremelyshort history has led to a diversity of opinions aboutits very substance” (Gal-Ezer & Harel, 1998, p. 79).While many educators believe that computer scienceshould be taught to prospective students as anindependent subject (Gurbiel, Hardt-Olejniczak, &Kolczyk, 2005) and as a full-fledged scientific subjecton a par with other scientific subjects (Gal-Ezer, Beeri,Harel, & Yehudai, 1995), others contend that it shouldbe offered as a set of integrated courses that linkcomputing to other subjects (Dagienë, 2005). Thisdiscussion about how computer science should betaught is further complicated by the fact that manypeople have difficulty identifying exactly whatcomputer science is. As a result, educators have beendeliberating how to portray the field to others(especially to newcomers) in a compact, compelling,and coherent way (Denning, 2004), and how toexpose others to a bird’s-eye view of the field (Gal-Ezer & Harel, 1998).

In recent years computer science educationresearch has received increasing interest andrecognition in the computer science educationcommunity (Holmboe, McIver, & George, 2001; Dale,2002; Almstrum, Hazzan, & Ginat, 2004; Fincher &Petre, 2004; Goldweber, Fincher, Clark, & Pears, 2004).Holmboe, McIver, & George (2001), for example,provided the following general categorization ofcomputer science education research to date:

• New, untested ideas and methods for teachingand learning.

• Reports from the trenches—sharing ideas and

techniques for teaching in particular courses. • Discussion of theory—with regard to

epistemological theories such as constructivism. • Computer-aided learning and intelligent

systems—aimed at learning about the students’ways of thinking and giving feedback.

• Expert/novice differences as a means of settingup benchmarks for novice achievement.

• Empirical studies, which focus on realprogramming, as a basis for creating effectivetools for teaching programming.

Despite the increased interest in computer scienceeducation research by the educational researchcommunity, there are still many open questionsrelating to computer science education at the highschool level for which answers are still very muchneeded. These include:

• Should we really teach computer science in high school, or is it best left to be taught at the post-secondary level?

• What are the goals of high school computerscience education, which topics should betaught, and how should the curriculum beorganized?

• How can we successfully promote a massiveimplementation of a new computer sciencecurriculum?

• What pedagogic strategies and methods ofinstruction should be applied to reduce thecomplexity and instability that characterizes a dynamic discipline?

• How should computer science teachers best beprepared to cope with the increasing complexityand rapid change?

• How should we cope with students’preconceptions and expectations, and howshould we better address gender differencesand minorities’ special needs?

All of these questions have become more pressing inlight of current concerns regarding droppingenrollments in computer science courses at alleducational levels and predictions of high techworker shortages worldwide. For many countries, therelationship between an educational system thatproduces workers capable of building the technologytools that the rest of the world will use and long-termeconomic prosperity is becoming increasing clear.This has led to efforts to improve computer scienceeducation, especially at the high school level, wheremany students have their first opportunity to testtheir interest and ability in this field. In the UnitedStates, however, the situation is far more tenuous asthe lack of national standards in both curriculum andteacher certification have resulted in significantdiscrepancies in both quality and quantity, not justfrom state to state but from school to school. Inaddition, lack of administrator understanding of thenature and importance of this complex and evolvingdiscipline may actually be diminishing support forcomputer science as part of the high schoolcurriculum at exactly the time that it should bereceiving more attention and support.

The aim of this review of computer scienceeducation research is to illuminate various aspects ofcomputer science education at the high school level,and to summarize reported studies and fieldexperiences, as well as educators’ recommendationsfor addressing the questions outlined above.

2.1 COMPUTER SCIENCEEDUCATION IN HIGH SCHOOLS

The research on high school computer scienceeducation is complex, not just because there is a lot ofit and it focuses on many different aspects ofeducation, but also because it is truly international in

scope, with researchers from many countriescontributing to the growing body of knowledge. This section provides a brief overview of high school computer science education in the UnitedStates, Israel, and European countries with the goal of providing a snapshot of its current focus andstatus internationally.

2.1.1 United States

In fall 2004, the Computer Science TeachersAssociation (CSTA, 2005) surveyed 14,000 high schoolcomputer science, computer programming, andcomputer applications teachers in an attempt toprovide a baseline understanding of high schoolcomputer science education in the United States. Thesurvey respondents (1,047 teachers) provided a richbody of data on the current state of computer scienceeducation including the following (Roberts &Halopoff, 2005).

• Many institutions teach computing courses at alower level than the Advanced Placement (AP)curriculum suggests; programming is the mostcommonly covered topic, followed byhardware, ethics, graphics, and webdevelopment (in that order). Courses that focuson applications are less likely to coverprogramming and vice versa.

• Schools offer more pre-AP computer sciencecourses than AP courses, and the AP coursestend to have significantly fewer students.

• The percentage of women declines from 32% inthe pre-AP courses to 23% in the AP program.

• Teachers believe the primary reason thatstudents are not taking computer sciencecourses is that they cannot fit them into theirtightly prescribed timetables.

• Teachers perceive the “rapidly changingtechnology” as a serious problem and thegreatest challenge in teaching computer science.

• Teachers indicated “time for training” as animportant professional development need andpreferred short workshops as the most effective

25

method for delivering professional development.• Teachers are surprisingly poorly informed about

the rules concerning curriculum content andteacher certification and the administrativestructures within their own states.

2.1.2 Israel

In Israel, computer science has existed as anautonomous subject in the high school curriculumsince the mid-1970s. As in many countries, however,despite its successful implementation and theinvolvement of many highly respected educators andresearchers in the development and evolution of itscurriculum, it still suffers from the perception that itis not a full-fledged scientific subject like physics,biology, or chemistry.

Originally, the curriculum focused onprogramming and included elective InformationTechnologies modules such as electronic spreadsheetlanguage. Since 1991, however, a new curriculum thatplaces significantly more emphasis on principles andtheoretical aspects has been gradually implementedin Israeli high schools. The new curriculum“emphasizes the foundations of algorithmic thinking,and teaches programming as a way to get thecomputer to carry out an algorithm” (Gal-Ezer, Beeri,Harel & Yehudai, 1995, p. 73). The program ismodular and comes in two versions—basic andadvanced. It includes mandatory Fundamentals andSoftware Design modules as well as elective modules,including Second Paradigm, Theory, andApplications.

The success of the high school computer sciencecurriculum in Israel is largely due to the care withwhich the government planned the implementationprocess and the resources that were put in place tosupport that implementation. This support includedthe development of course materials (learningmaterials for students and corresponding teacherguides) and an intensive in-service teacher trainingprogram (Gal-Ezer, et al. 1995; Haberman & Ginat,1999).

2.1.3 Europe

A substantial body of research on computing educationin Europe has supported the argument that educationalstandards, continuous teacher training, and enthusiasticteachers who permanently engage themselves in newtools and concepts are critical factors for the successfulestablishment and maintenance of high school computerscience education, more commonly called informatics, inEuropean schools (Reiter, 2005; Micheuz, 2005; Dagienë,2005; Dorninger, 2005; Kuznetsov & Beshenkov, 2005).According to Mittermeir (2005), however,

the penetration of personal computing and the(almost) ubiquitous presence of certain types ofapplication software have had substantial impact on computer science, more commonly calledinformatics, education across Europe. Over time, theprinciples of abstraction and algorithmization gaveway to intellectually less rewarding topics. (p. 2)

As a result, a number of countries have driftedtoward a more Information Technology (IT) focusthat was less rigorous and less embedded in science.

In 2004, a study of twenty European EconomicUnion member states as well as Bulgaria, Iceland,Liechtenstein, Norway, and Romania (Eurydice, 2004)revealed that studies of information and communicationtechnology form a part of the compulsory curriculum inthe upper secondary level in all countries. The ways inwhich that curriculum is implemented, however, differconsiderably from country to country. In the majority ofEuropean countries, information and communicationtechnology (or informatics) is included in the nationalcore curriculum as an independent subject. In Ireland,however, students do not study programming insecondary schools (Bergin & Reilly, 2005), and in Finlandand Italy, information and communication technology isno longer taught as a separate subject but rather as a toolto support learning in other curriculum areas (Kavander& Salkoski, 2004; Grandell, 2005; Cartelli, 2002).

In Poland, all students study informatics inprimary schools and in middle school. High schoolstudents also take a mandatory Information Technology

course that included topics such as networks and

26

multimedia tools for managing information. Anelective course called Informatics is also available forstudents who are interested in computer science as anelement of their future education. This course focuseson the science of computing with the goal thatstudents understand how “computer science as adiscipline is connected with designing andimplementing new systems of information processing”(Gurbiel, Hardt-Olejniczak, & Kolczyk, 2005, p. 49).

Informatics has been part of the high schoolcurriculum in Austria since 1985 when it was firstintroduced for all students in Grade 9. The recentreform of the upper secondary level of the academicschools also added an elective Informatics course forstudents in Grades 10–12 that reinforced integrationof informatics methods in other subjects as well(Micheuz, 2005). The development of informatics inthe Austrian schools is characterized by constantchanges and improvements in hardware, software,and by corresponding pedagogical approaches as well.“Didactics in Informatics changed in many schoolsfrom a programming and algorithm orientation to anapplication oriented approach” (Micheuz, 2005, p. 27).

Informatics as a separate subject was also establishedin Lithuanian schools in the late 1970s and early1980s. The subject is compulsory for students in Grades9–12. Its main goal is to develop students’ informationculture in a broad sense. Students in Grades 11 and 12may also choose an optional advanced course thatfocuses on programming, databases, and multimedia.“During the lessons the integrative nature of thecourse is being stressed; students are prompted to seeparallels with other subjects” (Dagienë, 2005, p. 56).

2.2 WHO ARE WE, AND HOWSHOULD WE PORTRAYOURSELVES ?

2.2.1 An Inconsistent Self-image

As is apparent from the very brief description of highschool computer science education in a number ofcountries, the discipline suffers from something of a

crisis of definition. As Gal-Ezer and Harel (1998) andHolmboe et al. (2001) have noted, even amongcomputer scientists, there is no clear agreement onthe name of the field (computer science, informatics,

information science), and efforts to determine its exactcontents have been similarly challenging. Researchershave identified many factors to account for this lackof consistency. Gal-Ezer and Harel (1998) posited thatthe disagreements regarding the essence of computerscience are due to a dichotomy between themathematical and engineering facets of the field andto dichotomies within each facet. Denning, Comer,Gries, Mulder, Tucker, Turner, and Young (1989)similarly described computing as sitting “at thecrossroads among the central process of appliedmathematics, science and engineering” (p. 11).

In addition, constant changes both in thetechnology and in our understanding of computingas a whole have made it extremely difficult to create astatic representation of the core elements of thediscipline. In 1989, when the number of coretechnologies was much smaller, Denning et al. (1989)suggested that an intellectual framework for thediscipline based upon a matrix model of nine coretechnologies should replace the common definition ofcomputer science in terms of its three majorparadigms: theory, abstraction, and design. Morerecently, however, in an effort to more accuratelyportray the rapidly developing and now morecomplex field, Denning (2004) recommended that anew framework of great principles and computingpractices was required to replace the conception ofthe discipline as a set of core technologies.

2.2.2 An Incorrect Public Image

Computer science’s public image, like its self-image,is highly problematic. Research has indicated thatbeyond disagreements about the nature and contentof the discipline, computer science is plagued by anumber of powerful misconceptions that are sharednot only by “real” outsiders (members of otherdisciplines, policy makers, school principals, etc.) but

27

also by students, prospective teachers, in-serviceteachers, and others involved in computer scienceeducation. Here is a series of brief descriptions of themost common of these misconceptions.

2.2.2.1 Computer Science Equals Programming

The most common misconception about computerscience arises from the fact that it is still widelyperceived as a programmer’s field (Denning, 2004).Denning et al. (1989) stated that the view thatcomputer science equals programming is especiallystrong in most of the curricula because introductorycourses focus (sometimes exclusively) onprogramming and this focus limits the ability toreliably describe the intellectual substance of thediscipline. Greening (1998) and Gries (2002) alsodemonstrated that students’ most commonmisconceptions about the discipline are compatiblewith the computer science equals programming view.Schollmeyer (1996) further noted that this limitedview of the discipline could have a profoundlynegative effect on students’ expectations of furthercomputer science studies in college.

2.2.2.2 Computer Science Equals ComputerLiteracy

The tremendous effort undertaken to integratecomputer use across the curriculum in K–12 educationhas led many educators and administrators to confusecomputer science education with computer literacy,and as a result, in many schools today there is noadequate separation between teaching computerscience, computer literacy, and using computers toteach other subjects (Gal-Ezer & Harel, 1998; Grandell,2005). Mittermeir’s (2005) study of nationalperspectives on teaching Informatics during the lasttwenty years in secondary schools in Europe concludedthat intellectually challenging curriculum content (forexample, abstraction and algorithmic thinking) is oftensacrificed to the teaching of simple software

applications such as word processors and spreadsheets.Sabin, Higgs, Riabov, and Moreira (2005) also

criticized high school computing courses in theUnited States because they are sharply polarizedbetween two curricular approaches: (1) basic trainingin using Microsoft office applications and (2) strongemphasis on AP programming (as preparation for APexams). Their concern was that neither of theseapproaches reliably portray the essence of computerscience, and they both reinforce the computer science

equals computer literacy and the computer science equals

programming misconceptions. Sabin et al. (2005)concluded that high school students should beexposed to an extracurricular enrichment computerscience program that focuses on fundamental conceptsin computing and some of its relevant applicationsand interweaves attractive learning activities.

2.2.2.3 Computer Science is a Tool for Studiesin Other Disciplines

In many countries there has been considerableconfusion between teaching computer science using areal-world approach and simply using the tools ofcomputer science to support education in other(particularly scientific) disciplines. Today there is alarge body of research evidence supporting use ofcomputer science principles and problem-solvingmethods to solve “realistic” problems in variousdomains (Sims-Knight & Upchurch, 1993; Dagienë,2005). Stevenson (1993) for example, argued thatstudents need to acquire a scientific-engineeringinterdisciplinary approach to solve complex problemsin various domains; hence, computer science shouldnot be taught isolated from mathematics and othersciences. Scherz and Haberman (2005) alsorecommended that students should use the tools ofcomputer science to develop knowledge-based projectsrelated to descriptive topics of a qualitative nature.

While these calls for an interdisciplinary approachwere never intended to be construed as an argumentfor eliminating computer science as a distinct subjectof academic study, in some countries, such as Finland

28

(Kavander & Salakoski, 2004; Grandell, 2005), therehas been a profound shift toward teaching computerscience only as a tool for addressing problems inother academic disciplines (Gal-Ezer & Harel, 1998).It is ironic to note this misinterpretation of theessence of the interdisciplinary pedagogic approachmay have resulted precisely because computerscience educators attempted to integrate computersinto all high school courses as a means of promotinghigh school students’ choice of computer science as amajor. This approach is illustrated by O’Lander (1996)who stated, “This will make more studentscomfortable and expert at using computers for a widevariety of purposes, and therefore increase theprobability that more of them will choose computerscience as a major in college” (p. 29).

2.2.2.4 Computer Science is Not a ScientificDiscipline

Members of the computing community of practiceconsider computer science as a scientific discipline(Denning, 2004). For example, one of the underlyingprinciples of the new high school program incomputer science in Israel is “Computer science is afull-fledged scientific subject. It should be taught inhigh school on a par with other scientific subjects”(Gal-Ezer et al., 1995, p. 75). Computer sciencecourses in high schools, however, seem to generallyhave a second-class status when compared to otherscience courses such as biology, physics, andchemistry, and rarely count towards satisfying thescience requirements in the high school curricula(Morris & Lee, 2004; Tucker, Deek, Jones, McCowan,Stephenson, & Verno, 2004). In order to establish acorrect public image, Denning (2004) suggested thatcomputer science should portray itself in the samemanner that the mature sciences (e.g., physics,biology, and astronomy) portray themselves, namelywith a principle-based approach that “promotesunderstanding from the beginning and shows howthe science transcends particular technologies” (p.337). Specifically, he recommended that computer

science, like other sciences, use a set of interwovenstories about the structure and the behavior of thefield elements.

2.2.2.5 Computer Science is a Male Field

Both male and female students continue to seecomputer science as a primarily male field and tomake their career choices accordingly. As Moormanand Johnson (2003) note, “Women in computing stillperceive themselves as strangers in a strange, male-dominated land” (p. 193). Educators believe that thelack of interest in computer science among girlsmight be rooted in stereotypes of computer scienceformed early in their school experience. Girls thinkthat computer science is “a boring subject, devoid ofinteresting applications and stimulating only to‘geeks’.” (Graham & Latulipe, 2003, p. 322).Consequently, female enrollment in undergraduatecomputer science programs has been constantlydeclining, and women tend to be underrepresented incomputing courses (Stiller & LeBlanc, 2003; Graham& Latulipe, 2003).

2.2.3 Challenging StudentPreconceptions

Students might actually possess the misconceptionspresented above even before they begin to learn thesubject in school. Rountree, Vilner, Wilson, & Boyle(2004) argue that the youngsters’ increased access tohome computers may cause many students to form amisleading view of what computer science is. Manypotential students also incorrectly believe that mostsoftware professionals will spend a significantamount of their career programming games (Latulipe& Graham, 2005). Furthermore,

Young people who like playing computers andsurfing the Internet declare their interest in optingfor an elective subject [i.e., computer science].They do not know what it is really about, becausethere was no signal … that problem solving,

29

algorithms, and programming need some skills ofthinking, reasoning, and understandingmathematical concepts (Gurbiel, Hardt-Olejniczak, & Kolczyk, 2005, p. 51).

Numerous researchers have noted that commonmisconceptions about computer science “raiseobstacles in encouraging students to pursuecomputer science majors, minor in computer science,or simply choose computer science electives in theirprogram of studies” (Sabin et al., 2005, p. 177). Guptaand Houtz (2000), for example, argued that studentsmay hesitate to enroll in computer science programsbecause of the negative perceptions about technologycareers and a lack of understanding of the skillsneeded to succeed in these careers. Foley (2004) alsosupported this contention, noting that sinceprogramming is often considered as a “solitaryactivity practiced by so-called geeks,” students mightavoid studying computer science. The generalconclusion of the body of research is that educatorsmust be aware of common misleading preconceptions and expectations and make everyeffort to deal with prospective students and findways to motivate them to pursue computer scienceprograms (Sabin et al., 2005).

In his study of university students’ beliefs aboutcomputer science, Greening (1998) attempted todetermine whether student misconceptions wererelated to high early student dropout rates in first-year computer science courses or to students decidingnot to enroll in computer science courses in the firstplace. Greening’s study revealed the following.

Most students were unable to give anapproximate definition of computer science or offeredpoor or erroneous descriptions of the domain.

Many potential students referred to computerscience in terms of the study of the computer as amachine and indicated that computer science is aboutthe use of computers for running applications.

Most of the students shared the perception thatfirst-year computing courses should teachprogramming, general usage skills, and providefamiliarity with general computing skills.

A significant percentage of potential studentsexpected to learn about computers and society-centered issues.

These results led Greening to conclude thatstudents did not understand the discipline ofcomputer science or its goals, and that effort requiredto address student misconceptions should be“directed at potential students in order to educatethem about the actual nature of computer scienceeducation, and directed at computer scienceeducation in order to acknowledge changingexpectations of potential students” (p. 154).

Researchers have also attempted to identify thefactors affecting high school students’ choice ofcomputer science courses, as well as the factorsleading to the decline of computer science majors. Forexample, O’Lander’s (1996) study of high schoolstudents who take computer science courses focusedon students’ attitudes and demographic andbackground characteristics. The results of this studyclearly indicated gender differences in favor of theboys. As a result, O’Lander recommended thatfollowing changes be made to ensure that computerscience be perceived as more attractive to all students:

• a unified computing curriculum should beestablished to eliminate disparities,

• computing teachers’ certification should becomea requirement (ensuring that competentteachers implement a challenging, interestingcurriculum, and eliminating or minimizing thecurrent practice of teachers certified in otherfields teaching computer science courses in highschool, and

• computers should be integrated into all highschool courses.

Gupta and Houtz (2000) conducted a similar study ofhigh school students. The primary purpose of theirstudy, however, was to better understand theperceptions and attitudes of females and minoritiestowards computer use and technology careers. Thestudents in Gupta and Houtz’s study identifiedkeyboarding as the primary skill necessary to succeedin IT careers, followed by computer skills,

30

programming, and math (in that order). This studyalso showed that boys were significantly moreinterested in enrolling in Information Technologyprograms and pursuing IT careers than girls were.The researchers therefore recommended that schools:

• develop a high-standard, uniform, mandatoryhigh school computer science curriculum;

• use software that appeals to both genders; • make computers exciting and challenging for

students; • implement a vigorous computer career

counseling program; and • explore race differences and interest in

computer careers.

2.3 HIGH SCHOOL COMPUTINGSTUDIES AND ACADEMICSUCCESS

Over the years, many professional bodies have calledfor the creation of a standardized computer sciencecurriculum for high school students as an importanttool for preparing students for post-secondarystudies. The report of the ACM Task Force on the Core

of Computer Science, which laid the foundation for thecomputer science curriculum for the major colleges,for example, recommended that students entercomputer science degree programs with someprogramming skills, presumably learned in highschool (Denning, Comer, Gries, Mulder, Tucker,Turner, & Young, 1988). Yet despite theseorganizational recommendations, some computerscience professors have long maintained thatstudying computer science in high school is in factdetrimental to students (Taylor & Mounfield, 1989).Research, however, largely disproves this contention.

In general, the research findings indicate thattaking computer studies in high school is a positivefactor for readiness for the college curriculum andcollege success (Franklin, 1987; Ramberg, 1986; Taylor& Mounfield, 1989). Specifically, researchers foundevidence that prior exposure to computers, whetherin high school or college programming courses

(Ramberg, 1986; Taylor & Mounfield, 1989; Cantwell-Wilson & Shrock, 2001; Cantwell-Wilson, 2002), oreven in computer literacy courses (Ramberg, 1986), isa critical factor for success in academic computerscience studies. Moreover, a “prior programmingcourse builds a foundation of logic background”(Ramberg, 1986, p. 37). Gal-Ezer and Harel (1998)have noted, however, that it is important thatstudents’ programming experience be of a highquality, and cover not only coding but also the designof algorithms (underlying programs), and basicconsiderations of correctness and efficiency.

Campbell and McCabe (1984) also attempted todetermine the specific factors that influence successin the first year of a computer science major. Theirstudy indicated that students’ background in highschool mathematics and science is one importantfactor for success. Based upon these results, theyconcluded that student participation in high schoolcourses that focus on problem solving and scientificreasoning may contribute to the success of freshmenwith a computer science major.

These conclusions were further supported byFranklin (1987). Franklin’s study aimed atdetermining whether high school computer sciencecourses were a predictor of success in college entry-level courses. The results showed a highly significantrelationship between success in the entry-levelcollege course and the completion of one or morehigh school computer science courses. A more recentstudy, conducted by the Open University of Israel,also revealed that students with previousprogramming experience were more successful in auniversity introductory level computer science coursethan those who reported having no previousprogramming experience (Gal-Ezer, Vilner, & Zur,2003; Rountree et al., 2004). Sabin et al. (2005)concluded that an early start in studying computerscience is appropriate in high school for thosestudents whose professional future is related tocomputer science and emphasized the importance ofcomputing courses designed to motivate prospectivestudents to pursue computer science programs.

It is important to note, however, that most of the

31

research literature on predictors of success incomputer science studies refers to traditional(imperative-first) introductory courses. Recently, as aresult of the trend towards teaching introductorycomputer science courses using an objects-firstapproach, researchers are being challenged toinvestigate the relationship of the objects-firstapproach to the previously identified predictors forsuccess in imperative-first introductory courses. Twosuch studies were conducted with contradictoryresults. A study aimed at assessing the effects of priorprogramming experience on the success in anintroductory objects-first course revealed that priorprogramming experience (not general experience orexperience with object-oriented programminglanguages) had no significant influence on success.Hence, the researchers concluded that priorprogramming experience is not a predictor of successfor the objects-first introductory course (Ventura &Ramamurthy, 2004). In contrast, Hagan andMarkham (2000) found that students who hadexperience in at least one programming language atthe beginning of an introductory Java programmingcourse performed significantly better than those withnone, and that the more languages with which astudent has experience, the better her/hisperformance tends to be. Furthermore, the style ofthe programming language previously learned wasnot shown to be significant.

2.4 WHAT SHOULD WE TEACHAND HOW?

The main goals of high school computer scienceeducation are to ensure that students are comfortablewith everyday computing activities, to open awindow of opportunity for young students in thefield of computing, to educate them about the natureof the field, and to motivate students who have theability and interest to continue their academic studiesin this discipline. The aim is to expose the students toa fundamental scientific domain whose principles arecharacteristic of algorithmic thinking as well as

system-level perception (Denning et al., 1989; Gal-Ezer et al., 1995).

The rapid growth of the field of computing andthe continuous emergence of new computingtechnologies requires educators to give considerablethought to what and how they should teach. Itrequires them to provide a coherent view ofcomputer science in a comprehensive and appealingway and to consider students’ perceptions andexpectations and any factors that may affect students’further studies in the field. In addition, high schoolcomputer science educators must determine whattopics are required to best prepare students forcollege-level computer science courses (Schollmeyer,1996), and how the long-term needs of students canbe addressed so that they will aspire to become partof the dynamic technologies-based world (Pham,1997; Mitchell, 2002).

Educators and scientists also argue thatknowledge of fundamental concepts of computerscience is essential for understanding the technologythat facilitates the “real world” and that thesefundamental concepts should be the core of a highschool curriculum. For this reason, it is important toorganize the curriculum with a suitable balancebetween conceptual, experimental, and practicalissues (Merritt, 1995; Gal-Ezer et al., 1995). As ascientific discipline, computer science has lasting andfundamental values (essential characteristics), whichare independent of computing technology’s change(accidental characteristics). Hence, teaching computerscience should emphasize the scientific facets of thefield, along with the knowledge and skills that areindependent of specific computers and programminglanguages (Gal-Ezer et al., 1995). The currentapproaches to high school computer scienceeducation, however, often hide the vision and thegrand challenges of computer science from studentsbecause they focus exclusively on developingcomputer programming skills and ignore the basicprecepts of the scientific method that form thefoundation in natural sciences (Lee, 2004).

Also, computing is a dynamic field, and thediscipline of computer science has continued to evolve

32

since its inception. This has presented educators withtwo additional pedagogical challenges:

• complexity—an increase in the programmingdetails students must master, and

• instability—rapid changes in the programminglanguages and tools.

Distinguishing between and separating essential andaccidental characteristics of the subject being learnedwould help eliminate this complexity and instability. Inan effort to directly address both the complexity andthe instability of the software, for example, Roberts(2004) argued for the development of a simple andstable environment explicitly designed forpedagogical use. Specifically, he concluded thatprogramming language instruction should emphasizeessential principles and conceptual aspects of theparticular programming paradigm while decreasingthe complexity of accidental implementation featuresthat are technology-based.

Denning (2004) noted the need to also address theseissues at a curriculum level. He suggested a frameworkfor organizing a computing curriculum around greatprinciples and practices that “promote a greaterunderstanding of the science and engineering behindinformation technology” (p. 340). This frameworkoffered a balance between concepts and practice. It alsoprovided a stable context for the core technologies ofcomputing related to both mechanics and practices.

In ACM’s Model Curriculum for K–12 Computer

Science, however, Tucker et al. (2003) outlined acomprehensive model curriculum for K–12 thatbuilds upon the notion of IT fluency put forward bythe National Research Council (1999). In this context(education), IT fluency is defined not only as thecapability to use today’s technology (the typicaldefinition of IT literacy) but also the capability toindependently learn and use new technology as itevolves. In this way, IT fluency expands theconception of a successfully educated person toinclude the active use of algorithmic thinking to solveproblems. The ACM Model Curriculum is acomprehensive four-level model for K–12 computerscience education that focuses on fundamental

concepts. It incorporates and enhances the NationalEducational Technology Standards (InternationalSociety for Technology in Education, 2002) for gradesK–8 (Foundations of Computer Science) and thenproposes a set of three courses:

• Computer Science in the Modern World,

• Computer Science as Analysis and Design, and • a special Topics in Computer Science course that

could include AP courses and courses leading to industry certification.

2.4.1 Guiding Principles

Many researchers and educators have attempted todefine the underlying principles or concepts that arefoundational to all of high school computer science.This section provides a brief description of the mostcommonly agreed-upon principles.

• General capabilities and skills—Students shoulddevelop a wide range of cognitive capabilitiesand practical skills, independent of specifictechnologies, to enable them to continue tolearn and adapt quickly to new environmentsand work practices (Pham, 1997; Gal-Ezer et al.,1995). They must also acquire a breadth of skillsrequired for future employment such as writing,communication, and presentation, and otherskills that enable them to continue to learn andadapt quickly to new environments and workpractices (Pham, 1997; Mitchell, 2002).

• Breadth-first approach—Students should acquirea broad overview of the field to construct acomprehensive picture of computer science as adiscipline. They should have a sense of thehistory of the discipline so that they can betterappreciate how computers have been used toaddress real problems (Impagliazzo & Lee,2004; Hazzan, Impagliazzo, Lister, & Schocken,2005). Different programming paradigmsshould be taught to acquire alternative ways ofalgorithmic thinking as well as differentproblem-solving methods (Vandenberg &Wollowski, 2000; Gal-Ezer et al., 1995).

33

• Understanding the interplay between theory and

practice—Students must understand not onlythe theoretical underpinnings of the disciplinebut also how that theory influences practice.Accordingly, conceptual and experimental issuesshould be interwoven throughout the program(Gal-Ezer et al., 1995; Ben-Ari, 2002). However,since beginning computer science students donot have an effective model of the computer,“the seductive reality of the computer must notbe allowed to replace construction of models”(Ben-Ari, 2001); meaning that programmingexercises should therefore be delayed until classdiscussion has enabled students to construct agood model of the computer.

• Problem solving and algorithmic thinking—

“Students must be taught and coached in thealgorithmic way of thinking” (Gal-Ezer & Harel,1998, p. 83). To best prepare high school studentsfor computer science courses in college, theemphasis at the high school level should not belimited to the instruction of the syntax of aprogramming language. Rather, students’problem-solving skills should be addressed. Thisincludes emphasis on teaching problem-solvingmethodologies and critical thinking skills (Fadi& McHugh, 2000) as well as programspecification and design (Schollmeyer, 1996).Integrated learning environments should also bedesigned to support novices through allproblem-solving and programming stages (Fadi& McHugh, 2001).

• System-level perspective—Students should betaught what will be expected from them in the“real world,” specifically, what is actuallyrequired to write and maintain computerprograms and large software systems. Theymust therefore develop a high-levelunderstanding of systems as a whole: thestructure of computer systems and theprocesses involved in their construction andanalysis (Gal-Ezer & Zeldes, 2000; Schollmeyer,1996). Design principles should be taughtexplicitly, otherwise, the first programming

course might turn into a course on syntax nomatter how simple the underlyingprogramming language is (Felleisen, Findler,Flatt, and Krishmamurthi et al., 2002).

• Integrative and interdisciplinary knowledge—

Denning et al. (1989, p. 13) suggested that“interacting with other disciplines to support theirinterests in effective use of computing” shouldbe an integral part of a curriculum for computingmajors. Stevenson (1993) presented a possiblefoundation for computational science, which is amuch wider domain than computer science, andembodies a scientific-engineering interdisciplinaryapproach to solving very difficult problems. Heremarked that computer science students “caneasily see computer science as devoid of meaningand programming devoid of empirical import”(p. 9). He also noted that students are not wellprepared to cope with complex problems thatinvolve mathematical, scientific, and engineeringreasoning. Hence, the computer science studies inhigh school should be enriched with mathematicsand sciences, with emphasis on numericalanalysis applications, and scientific softwareengineering concepts, and students should beeducated to employ an interdisciplinary teamapproach (Stevenson, 1993).

• Theoretical knowledge—Educators believe that itis important to expose high school students toadvanced topics of computer science (e.g.computational models, modeling, and parallelcomputing) to enable them to become familiarwith some of the theoretical aspects of computerscience (Armoni & Gal-Ezer, 2003; Loidl,Muhlbacher, & Schauer, 2005; Ben-David Kolikant,2004). These concepts may be introduced in atraditional way; however, they can be made morecomprehensible when demonstrated by meansof new media such as applets (Loidl et al., 2005).

• Familiarity with common principles—studentsmust recognize recurring concepts andprinciples such as abstraction, complexity,modularity, and reusability (Denning et al.,1989; Gal-Ezer et al., 1995).

34

• Comprehensive programming—Programmingshould be taught as a means for getting thecomputer to execute an algorithm. It should betaught in a broad sense, “covering not only thecoding act itself, but also the design of thealgorithms underlying the programs and, tosome extent, considerations of correctness andefficiency” (Gal-Ezer & Harel, 1998, p. 82). Highschool programming classes should attempt toprepare the students to be successful incorresponding college-level classes by teachingmaterial that is universally applicable to anyprogramming paradigm (Schollmeyer, 1996).Fincher (1999), however, has questioned theargument that students must learnprogramming first and disputed the notion thatthere is a distinguished order to conceptacquisition. “Instead of accepting the view thatstudents need to learn to code and that fromthis experience they will learn complex,transferable skills…. leaving the students toabstract these for themselves,” (p. 12a4-5) sheargued, one should first identify the skills thatshould be acquired and support studentlearning to achieve these goals.

• Meaningful learning of computing principles in

attractive experimental environments—Teachersshould motivate students to endure the rigors oftraditional computer science programs byenticing students with exciting, accessible, andleading-edge courses. Guzdial and Solloway(2003) and Stiller and LeBlanc (2003), forexample, have suggested that introducingcomputing in an authentic environment, such asthe world of audio and visual media, wouldmotivate students and interest them in the restof computer science.

• Teaching from an historical perspective—

Böszörmenyi (2005) recommended that inaddition to teaching formal definitions andprogramming practice, teachers should providestudents with an understanding of the great ideasof the major thinkers in computer science and thecontext in which these ideas were developed.

• Diminishing common computer science

misconceptions—Teaching and learningactivities should be designed to treat commonmisconceptions of the very essence of computerscience. In particular, it is important to diminishthe computer science equals programmingmisconception by developing an appropriatefundamentals course that emphasizes thebreadth of the discipline beyond the interest incomputer programming (Gries, 2002; Gal-Ezer& Harel, 1998; Peckham, DiPippo, Reynolds,Paris, Monte, & Constantinidis, 2000;Vandenberg & Wollowski, 2000).

• Gender differences and minority issues—Thecurriculum should be designed to address theunder-representation of women and minoritystudents in computer science programs (Rich,Perry, and Guzdial, 2004; Guzdial & Soloway,2003; Graham & Latulipe, 2003; Payton, 2003;Stiller & Leblanc, 2003).

2.4.2 Teaching the Key Concepts

There is also a large body of research that examined theimportance of taking a conceptual approach to teachingcomputer science in high schools. In this section, we briefly summarize these research studies with the intention of highlighting the key findings andsuggestions for curriculum contents and enhancements.

2.4.2.1 Basic Computing Concepts

Several researchers (Perkins, Schwartz, & Simmons,1988; Sleeman, Putnam, Baxter, & Kuspa, 1989; duBoulay, O’Shea, & Monk, 1989) have noted thatstudents lack an adequate knowledge of programmingand also demonstrate many misconceptions aboutbasic computing concepts because they lack viablemental models of the computer. Ben-Ari (2001) arguedthat the common difficulties experienced by novicescould be explained by computer science educationbeing heavily weighted on the side of bricolage and

35

involving too much and/or too early use of thecomputer (e.g., try it and see what happens, endlessdebugging, and avoiding abstraction). As a result, heconcluded, high school students aspire to programdirectly on the computer, and they sometimescomplain that algorithm development and analysis,instead of just getting on with writing and debuggingprograms, is a waste of time.

2.4.2.2 Algorithms

A number of researchers have addressed the specialneeds of novice programmers. Haberman and Ben-David Kolikant (2001) concluded that teaching novicestudents basic concepts in computation is not asimple task. In fact, they reported that educators maybe inclined to believe that these concepts are trivialand therefore easy for students to understand, whenin fact, novice programming students may notunderstand them at all. They therefore recommendthat teachers carefully choose an instructionalmethod that clearly demonstrates and reinforces evenseemingly simple concepts.

Haberman, Averbuch, and Ginat (2005) alsoconcluded that novices tend to have limitedunderstanding of algorithms, and so often consideran algorithm as a stand-alone product, thus ignoringissues such as correctness and efficiency. Moreover,they concluded that students perceive algorithms asoperational process emulation elements, rather thanstructural concepts that yield a specified I/Orelationship. As a result of these findings, theresearchers suggested that students should be taughtthat explanations and justifications are inherentelements in the solution of an algorithmic problem andthat teachers should demonstrate the complete analysisand design processes instead of just final products.

2.4.2.3 Recursion

Levy and Lapidot (2002) conducted a study todetermine how novices see recursion as an

interdisciplinary concept and identified patterns ofstudents’ expressions and ways of thinking whencreating recursive descriptions using a naturallanguage. The study findings indicated that classdiscourse plays an important role in helping studentsto understand recursion and in helping educators tounderstand students’ conceptual schemes.

Haberman and Averbuch’s (2002) study focusedon students’ conceptions of the base case as anintegral component of a recursive algorithm. Theyfound that students have difficulties identifying basecases. They handle redundant base cases, ignoreboundary values and degenerated cases, avoid out-of-range values, and may not even define any base caseswhen formulating recursive algorithms. Theysuggested that teachers should discuss differentaspects of the base case concept, such as declarativeand procedural aspects of categorizing and handlingbase cases as part of formulating recursion. Inaddition, Haberman (2004b) found that students wholearned recursion in logic programming using adeclarative approach before learning it in a proceduralparadigm, constructed more adequate mental modelsof recursion and were therefore better able to userecursion as a tool for knowledge representation.

2.4.2.4 Abstraction and Abstract Data Types

Many studies have investigated high-school students’perceptions of abstraction and abstract data types.Gal-Ezer and Zeldes (2000) for example, found thatstudents who studied a software design unit,acquired technical skills that improved their handlingof problems requiring the definition and use ofabstract data types, even though many of them didnot fully understand the concept itself. Haberman,Shapiro, and Scherz’s (2002) study on the cognitiveaspects of using abstract data types as tools forknowledge representation and problem solving in thedeclarative logic programming paradigm revealedthat students’ perceptions of the abstract data typeconcept varied as did their strategies for programdevelopment. Perhaps surprisingly, Haberman

36

(2004a) also found that high school students felt morecomfortable than undergraduate students withalgorithms with low-level abstraction.

2.4.2.5 Non-determinism

Armoni and Gal-Ezer’s (2003) study of high schoolstudents’ perceptions of the advanced abstractconcept of non-determinism showed that studentshad a marked tendency to use the deterministicmodel independently of the given problem,indicating that they had only a partial understandingof the non-deterministic model. Based on the studyresults, the researchers suggested that the teachingprocess should emphasize both the theoretical andtechnical aspects of computer science.

2.4.2.6 Correctness

Haberman and Averbuch (2002) found significantevidence of students’ misconceptions about correctness.In their study, students indicated that recursivealgorithms were correct even though they did nothandle every possible instance of the problem while atthe same time they indicated that an algorithm wasincorrect when it handled imperceptible, thoughrelevant, base cases. This conclusion was furthersupported by Ben-David Kolikant (2005). She foundthat, in contrast to the professional dichotomousdefinition of correctness, students understoodcorrectness as a relative property of the program andtherefore tolerated errors. For example, students whoproduce incorrect programs might describe theprograms as mostly correct; or they might considerprograms as correct when they produce the expectedoutput, yet in addition, they produce more unexpectedoutput. Recent research by Haberman, Averbuch, andGinat (2005) also determined that students hold themisconception that a short algorithm is not satisfactoryif it neither represents the story described in theproblem nor the problem’s analysis process. Hence,students might not consider short concise algorithms as

satisfactory solutions, even though they correctly yieldthe desired I/O relationship.

2.4.2.7 Algorithm Efficiency

Ginat’s (1996) findings indicated that students’solutions to algorithmic problems varied considerablyin terms of efficiency, reflecting different levels ofinsight into the problems. He found that fruitful classdiscussions of different solutions to a given problemregarding efficiency widened students’ repertoire ofpossible solutions and made them realize theimportance of analysis and planning in developingefficient solutions. Gal-Ezer and Zur (2002) alsoinvestigated high school students’ perceptions of theconcept of algorithm efficiency. They found that mostbeginners (grade 10 students) had much moredifficulty internalizing the concept of algorithmefficiency than the 11th graders, because of relatedmisconceptions such as “the more variables, the moreexecution time.” Grade 11 students who opted forcomputer science, internalized the concept anddemonstrated the ability to write efficient programs tosolve simple algorithmic problems. When solvingcomplex algorithmic problems, however, thesestudents tended to write insufficient programs becauseof their inadequate ability to analyze the problem. Gal-Ezer and Zur therefore recommended that teachersshould invest more effort teaching studentsmathematical analysis before writing programs.

2.4.2.8 Program Testing

Ben-David Kolikant (2005) investigated high schoolstudents’ work habits and conceptions of testing. Thestudy findings indicated that the students usedinadequate methods of testing, and that theirdefinitions of systematic testing were inherentlydifferent from those of professionals. Ben-DavidKolikant and Pollack (2004a) found that the students’norm is to produce working, but not necessarilyerror-free, programs and to argue for their correctness

37

solely on the basis of a few executions. Theysuggested the integration of explanatory proofs as ameans of establishing norms of precision in thestudents’ practices and methods of communication.Haberman, Averbuch, and Ginat (2005) furthersupported this conclusion and argued that explicitrules of oral and written discourse should beelaborated in order to establish reliable and coherentstudent-teacher and student-peer communication.

2.4.3 Pedagogical Approaches

In addition to examinations of principles and conceptsin high school computer science instruction,considerable research effort has been spent attemptingto identify the most effective pedagogical approachesfor addressing the principles and concepts that mustbe taught. Much of the research focuses on the use ofspecific programming languages and/or integrateddevelopment environments. In this section, however,we will concentrate on a selection of key studies thatfocus on instructional methodologies.

2.4.3.1 Problem-Solving Strategies

Ginat (2000) implemented a problem-solvingapproach that used colorful and attractive challengesand games involving physical objects at the primarystage of problem analysis to prompt students to lookfor the problem’s characteristics “from variousangles, in different ways, and for diverse tasks” (p.81). They argued that this approach allowed highschool students to develop “a set of values andperspectives that enhanced their mathematical andcomputer science point of view” (Ginat, 2000, p.84).

Muller (2005) investigated the influence of pattern-oriented-instruction on the development of problem-solving skills. The preliminary results of the studyshowed that students who had acquired a repertoire ofpatterns achieved greater insight into a problem’sessence in a shorter time and were able to developsolution ideas more easily and with less distraction

from the details of a problem’s context. Moreover,pattern-oriented-instruction improved the written andverbal formulation of ideas of both teachers andstudents).

2.4.3.2 Learning Different ProgrammingParadigms

Haberman and Averbuch (2002) found negativetransfer among students transitioning from thedeclarative logic programming paradigm to aprocedural paradigm. Their findings indicated thatstudents who learn list processing in Prolog tendedto ignore boundary cases that control the terminationof recursive computation when developingprocedural algorithms. In contrast, Haberman(2004b) found that students who learned recursion inlogic programming according to a declarativeapproach, before learning it in a proceduralparadigm, constructed better mental models ofrecursion than students who were acquainted withrecursion in procedural programming only.

Paz and Lapidot (2004) investigated theperformance of high school students while they werefirst exposed to the functional programmingparadigm. The most prominent finding in the studyconcerned the students’ conception that a listprocessing function may change the value of its inputparameter.

Ragonis and Ben-Ari’s (2005) study focused onhow novice programmers learn object-orientedprogramming. Their investigation into the conceptsheld by novices revealed many difficulties anderroneous conceptions that the researcherscategorized into four primary categories:

• class versus object, • instantiation and constructors, • simple versus composed classes, and • program flow.

The study findings further indicated that when theinstructor provided a detailed description of thedifficulties and conceptions together with anintegrated and holistic view supported by specific

38

examples, authentic episodes, and interpretations, thestudents developed a much better understanding ofthe basic principles of object-oriented programming.

2.4.3.3 Software Design and Project Development

Sims-Knight and Upchurch’s (1993) study found thatwhen instructors taught software design before aspecific programming language, students grasped thebasic concepts necessary to create high-level designseven though they had no programming experience.Gal-Ezer and Zeldes (2000), however, found that highschool students who studied software designexhibited difficulties in designing general top-downsolutions for a given problem; and instead, theypreferred to deal with specific examples. The students,however, were able to reuse general structures todistinguish complex tools from basic tools.

Collofello’s (2002) study indicated that studentswho had access to a simulated environment in whichthey could learn and practice software developmentacquired their understanding of the softwaredevelopment process as a gradual process composedof many steps. Moreover, they were better atdocumenting the development requirements for theirprojects, employing use cases, and in usingsystematic testing and inspection techniques.

Pollack and Scherz (2005) investigated theinfluence of supportive learning materials on highschool students’ motivation, performance, and finalproducts as they developed projects in computerscience. The study findings indicated that studentswho tended to perceive projects as a school activitywere mainly motivated by outside rewards such asthe projects’ external assessment. Students who didnot use supporting materials for project-basedlearning and instruction used a bricolage approach todevelop their projects instead. They also tended tomodify the original problem according to their abilityto develop a proper project. In contrast, students whoused support materials for project-based learningshowed a greater commitment to the project andsubmitted a product that referred to the original

problem. Scherz and Haberman (2005) also foundstudents who used problem-solving organizing toolsin developing their projects were more likely to useabstract data types that resulted in a structured andwell-organized development process.

2.5 WHOM SHOULD WE TEACH?

The current structure of high school educationprovides many indicators as to the value placed upona foundational understanding of the traditionalsciences as part of the required knowledge of everyeducated citizen. Physics, biology, and chemistryhave long been required elements of high schoolstudent learning. For computer science, however, thediscussion of its increasing importance as part of therequired knowledge base of every citizen in ourincreasingly technological world, has been sadlyabsent, and hence is long overdue. Lee (2004), forexample, argued that academic computer sciencemust “re-educate the public-at-large (especiallychildren) on the scientific foundations andfundamental questions of the field.” Guzdial andSolloway (2003) also suggested, “Students shouldemerge from an introductory course with a senseabout what’s interesting in the field, and they shouldhave some practical knowledge that they can applyin their fields” (p. 5). Unfortunately, however,research has demonstrated computer science’scontinuing inability to attract young women andminority students to the discipline in representativenumbers and to keep them there beyond theintroductory level.

Even in countries where there is a commitment toproviding computer science education for all highschool students, research has identified two long-standing barriers to equity. Specifically, youngwomen and minority students continue to bedisproportionately under-represented in computerscience education. This pattern is most clearlyexemplified in the United States by the demographicbreakdown of students taking Advanced Placement(AP) computer science. The College Board (2005)

39

reported that in 2004 only 15% of the students takingthe AP computer science exams were female. Inaddition, fewer than 5% of students of either sex self-identified as Latino/Latina and fewer than 3% wereAfrican American.

To date there has been a considerable amount ofeducational research dedicated to determining whythis under-representation persists in computerscience and not in other sciences. Fisher and Margolis(2002), for example, found that the context ofcomputing is often very important for women andthat more women than men link their interest incomputer science to other arenas. They thereforerecommended that both the curriculum and theculture are changed to ensure that young women donot perceive that they will succeed only if they“model themselves after the stereotypical malecomputer science student” (p. 80). As alreadymentioned, research by Moorman and Johnson (2003)provided further support for Fisher and Margolis’conclusions. They found that “Women in computingstill perceive themselves as strangers in a strange,male-dominated land” (p. 193). Moorman andJohnson (2003) also found that both male and femalestudents continue to see computer science as aprimarily “male” field and to make their careerchoices accordingly. They suggested that single-sexsummer workshops for high school girls and mother-daughter computer science clubs for elementaryschool girls would help to combat this perception.

Moorman and Johnson’s call for single-sexeducational opportunities was consistent with anearlier study by Crombie, Abarbanel and Anderson(2000). Their study on girl-only learning settingsfound that female enrollment significantly increasedin all-female computer science classes. The girls intheir study also exhibited a higher level of confidencein their own skills than did girls in mixed genderclasses. These results were also supported by Grahamand Latulipe (2003). As a result of their investigationat the University of Waterloo, they developed an all-female computer science seminar for high school girlsto encourage female students to enter computerscience. This program focused on showing young

women a real-world perspective on computer scienceand providing them with positive role models todispel the negative stereotypes.

Research has also provided significant evidencethat minority students have been similarly disaffectedfrom computer science, both as a discipline and as acareer area. Although policymakers and industryleaders encourage minority participation in scienceand technology academic studies with the intentionof increasing the number of minority students whoenter, are retained in, and succeed in high-techcareers, the research indicates that a very lowpercentage of minority students intend to major inthis field. Payton (2003), for example, found thatAfrican-American teenagers tended to avoidmajoring in computer science, information science, orinformation technology.

2.6 EXEMPLARY TEACHERS

While it is clear that student attitudes are animportant factor in students choosing to studycomputer science in high school, it is also clear thatteachers are the corner stone of successful curriculumimplementation. As a result, there has beenconsiderable research relating to the development ofappropriate computer science teacher trainingprograms (Gal-Ezer et al., 1995; Lapidot & Hazzan,2003; Tucker et al., 2004; Dorninger, 2005). Thegeneral consensus of this research, as Gal-Ezer andHarel (1998) note, is that high school computerscience teachers should be trained to convey technicalknowledge correctly and reliably, to teach skills, toprovide perspective, and to infuse the students withinterest, curiosity, and enthusiasm. Recruiting andretraining competent high school computer scienceteachers, however, has proven extremely problematicdue to a lack of adequate teacher training programsand the relatively low pay compared to that ofindustrial computer scientists (Poirot, 1979; Poirot,Taylor, & Norris, 1988).

Researchers such as Poirot (1979) have thereforeidentified several levels of educational intervention

40

and support needed to ensure an adequate supply ofhighly trained computer science teachers. Theseinclude:

• conducting training institutes for in-servicecomputer science secondary teachers,

• implementing a service course within thecomputer science curriculum designedespecially for prospective computer sciencesecondary teacher training, and

• implementing teacher certification programs incomputer science.

The following sections examine the research relatingto both pre-service and in-service teacher training inmore detail.

2.6.1 Pre-service Teacher Training

The existing research highlighting the link betweenstudent success at the college/university level andthe methodology used to teach high school students(Taylor & Mounfield, 1991) has raised fundamentalquestions concerning pre-service computer scienceteacher education. According to Lapidot and Hazzan(2003) the pressing questions articulated by thisresearch include the following:

• Which appropriate frameworks would fit aMethods for Teaching Computer Science in High

School (MTCS) course? • Which topics should be included in such

courses? • Which kinds of activities would be best suited

for the preparation of future computer scienceteachers?

In an effort to address similar questions, Gal-Ezerand Harel (1998) recommended that computerscience teachers should acquire both technical andpedagogical knowledge: “Beyond the mastery of coreCS material, good CS educators should also befamiliar with a significant body of material that willexpand their perspectives of the field, andconsequently, enhance the quality of their teaching”

(p. 77). This conclusion was supported by Lapidotand Hazzan (2003) who noted that “a merger ofcomputer science with pedagogy into one amalgamof content and pedagogy” (p. 30) would prepareteachers to better employ general pedagogicalprinciples as well as teaching methods in the contextof computer science education.

2.6.1.1 Technical Knowledge

According to Gal-Ezer and Harel (1998) and Poirot(1979), high school computer science teachers shouldhave a thorough formal background in computerscience (on an academic level). Teachers must also havea broad overview of the pertinent areas of computerscience, and to have acquired a bird’s-eye view of thediscipline: “For those desiring a more thoroughcoverage of a particular topic, existing computer sciencecoursework would be required. The course content ofa computer education course must not only includecomputer related material, but also the motivation forcovering each topic” (Poirot, 1979, p. 102).

2.6.1.2 Pedagogical Knowledge

Many computer science educators lack formaltraining in education and are unfamiliar witheducational theories; hence, they rarely attempt toapply these theories in the computer scienceclassroom (Gal-Ezer & Harel, 1998; Ben-Ari, 2004). Tobe best qualified, computer science high schoolteachers should be certified and offered courses incomputer science education in addition to regularcomputer science courses (Schollmeyer, 1996). Gal-Ezer and Harel (1998) additionally recommend thatseveral topics that would expand teachers’perspectives of the discipline should be part of acomputer science teacher’s education. These wouldinclude:

• a history of computer science, • the nature of computer science and its

relationship to other disciplines,

41

• various computer science curricula, • pedagogical knowledge related to teaching

computer science, and • the use of tools and aids in teaching the subject.

They also suggest integrating these subjects intospecially tailored in-service training programs forhigh school teachers, especially those who lack aproper pre-service computer science and pedagogicaleducation. Based on their experience of teaching theMTCS course to prospective teachers, Lapidot andHazzan (2003) additionally recommended optionalcourse frameworks and correspondingimplementations arranged as clusters of optionalactivities. For example, they suggested an activelearning-based teaching model that would supportthe construction of computer science teachers’professional perception (Hazzan & Lapidot, 2004a).

Tucker et al. (2004) argued that as part of theirpre-service training, prospective computer scienceteachers must also practice the teaching of computerscience in schools before becoming actual computerscience teachers. This argument was furthersupported by Hazzan and Lapidot (2004b), whosuggested that teaching practice should be organizedin a way that supports bridging gaps betweentheoretical knowledge and actual performance:

no matter how much this topic is discussed in theMTCS course, the actual implementation ofdifferent teaching methods in real teachingsituations, together with a reflection process thatfollows, can improve the prospective teachers’understanding with respect to this topic. (p. 48)

They also noted that the illustration of situations thattake place during the practice may help bridge thegap between the theory (the MTCS course), reality(what actually happens in schools), and theuniversity mentor’s perspective (academia).

2.6.1.3 Socio-cultural Knowledge

There is also a small but interesting body of research

indicating that prospective computer science teachersshould be educated to become part of a collaborativecomputer science teachers’ community of practice(Haberman, Lev, & Langley, 2003; Ben-DavidKolikant & Pollack, 2004b). Specifically, they shouldbe motivated to learn from the experience of expertsand experienced teachers and to seek theircounseling. Moreover, they should be guided tocontact supportive groups such as local and nationalteacher associations and to attend their activities(Lapidot, 2002). In this way, it is suggested thatteachers will gradually proceed towards full-fledgedmembership in the computer science teachers’community of practice.

2.6.2 On-going ProfessionalDevelopment for Teachers

The rapid change of programming languages andtools for teaching computer science makes itexceedingly difficult for teachers to remain current inthis discipline. This challenge makes computerscience less attractive to qualified teachers and oftenleads to defections to other subject areas (Roberts,2004). Thus, a tremendous effort is needed to supportteachers in the adaptation to new programminglanguages and tools for teaching computer science,the adoption and implementation of new curricula,and in the constant enhancement of their pedagogicalapproaches.

Assimilating a new computer science curriculumrequires appropriate in-service training aimed atintroducing the curriculum and its didactic approach(Haberman & Ginat, 1999). In addition, teachersrequire on-going support in all aspects of classroomimplementation of the curriculum. According toHaberman, Lev, and Langley (2003) the assimilationof a new curriculum “should be accompanied by theestablishment of a collaborative ‘learning andcreating’ community of teachers. This communitywill begin its activity within an organized guidedcourse and continue with intra-school activity andinter-school cooperation” (p. 144).

42

2.6.3 The Role of ProfessionalAssociations in SupportingTeachers

In addition to occasional projects of massive widespreadtraining, national and regional teacher associations areplaying a considerable and growing role in both thedevelopment and support of new curricula. They arealso, especially in the case of computer science teachers,becoming the primary providers of professionaldevelopment opportunities (Driscoll, 1987; Inos &Quigley, 1995; Adagian, 1996). As Harris (1987) noted,participation in professional associations has a directimpact on professional effectiveness because it providesmany direct benefits to teachers. It:

• provides new expertise through professionaldevelopment workshops,

• establishes networking relationships thatprovide opportunities to share curricula,

• develops stronger personal commitment tostudents and the academic discipline, and

• increases the realization of the importance ofconnections with business and industry.

Professional associations also provide exposure tonew ideas, either through organized events such asconferences and workshops or through professionalpublications (Bell, 1983). As Romberg and Middleton(1995) argued, “without programs that fostercollaboration, teachers would not be able to discovernew teaching methods and ways of dealing withstudents and administration that their colleagues canand do provide” (p. 175).

Professional educational associations also provideessential support for specific academic disciplinesand play a key role in the development of leadershipwithin those disciplines. According to Jeffrey (1996)professional teacher associations support individualacademic disciplines by:

• encouraging the testing of new initiatives,• sharing expertise,• satisfying an almost insatiable need for

discipline-based professional development,• identifying key research needs,

• communicating the perspective of those in aparticular discipline to the wider educationalcommunity,

• recognizing trends in discipline content, and• putting strategies in place to ensure that

equitable standards are maintained.

The Computer Science Teachers Association (CSTA) forexample, is a membership organization that supportsand promotes the teaching of computer science andother computing disciplines at the K–12 level.Established in 2004, CSTA has already built a strongsystem of partnerships and carried out a number ofsuccessful projects (development of the ACM Model

Curriculum for K–12 Computer Science Education, the JavaEngagement for Teacher Training project, five ComputerScience and Information Technology symposia) thathave given it credibility and forged strong ties withthe community of computer science educators.

After examining the current research andexperiential classroom and administrative practices,the CSTA Board of Directors identified the followingkey components required to facilitate much neededimprovements to K–12 computer science education:

1. Communication/Partnership: Effective changerequires the engagement and support ofteachers, principals, school board staff,guidance counselors, administrators, parents,college faculty, College of Education faculty,legislators, and key industry partners.

2. Curriculum Standards: Consultation on,publication of, endorsement of, and adoptionof curriculum guidelines are needed toestablish the required body of knowledge andimprove consistency in classrooms nationally.

3. Curriculum Support: Models of success and thedevelopment and dissemination of teachingand learning materials help ensure theadoption of curriculum standards and addresscritical resource shortages. These must besupported by accessible and relevantprofessional development for teachers.

4. Research: Policy-makers (administrators andlegislators) require data to convince them that

43

it is time for a change and that proposedchange mechanisms will work.

5. Teacher Certification Standards: Ensuring thatthere are sufficient numbers of qualifiedteachers and that teacher certification standardsare consistent from state-to-state will improvepedagogy and course content nationally.

Like many other such associations internationally,CSTA has taken on the role of not only advocating forthe place of computer science within the larger highschool curriculum, but also providing the resourcesand support that practitioners require to continue toimprove their teaching (http://csta.acm.org/).

Machshava, (the Israeli National Center for HighSchool Computer Science Teachers)(http://cse.proj.ac.il), has assumed a similar role forIsraeli teachers (Lapidot, 2002). Its initiatives aredirected at similar goals, including:

• fostering professional leadership of computerscience teachers;

• helping create a professional community ofcomputer science teachers; identifyingcomputer science teachers’ needs;

• supporting local teacher centers; and • collecting and distributing computer science

education knowledge and experience.

2.7 CONCLUSION

For many years, there has been considerable debateover the place of computer science education in thehigh school curriculum (Stephenson, 2002). Whileresearchers and educators may hold differing opinionsconcerning when a program of computer scienceeducation should be implemented and how it shouldbe taught, the research consistently supports theconclusion that, like any other academic discipline, itmust be engaging and rigorous, or as Gal-Ezer et al.(1995) concluded, high school computer science “mustbe challenging, in the sense that it will not only teachthe foundations, but will also relate them to thepractical side of computing, and it should train the

students to deal with intellectually demanding tasks”(Gal-Ezer et al., 1995). Deciding what and how to teachis, however, just one element in ensuring that even thebest curriculum is actually implemented andsupported over time. The next chapter will, therefore,examine the multi-level issues that affect successfulcurriculum implementation.

2.8 REFERENCES

Adajian, L. B. (1996). Professional communities:Teachers supporting teachers. Mathematics Teacher,

89(4), 321–364.

Almstrum, V. L., Hazzan, O., & Ginat D. (Eds.).(2004). Special issue on import/exportrelationships to computer science educationresearch. Computer Science Education, 14(4).

Armoni, M., & Gal-Ezer, J. (2003). Non-determinismin CS high school curricula. Proceedings of the FIE

2003 Conference, Boulder, CO, F2C–18 – F2C–23.

Bell, D. (1983). Promoting business education throughteacher participation in professional associations.Business Education Forum/Yearbook, 37(8), 48–51.

Ben-Ari, M. (2001). Constructivism in computerscience education. Journal of Computers in

Mathematics and Science Teaching, 20(1), 45–73.

Ben-Ari, M. (2002). From theory to experiment topractice in CS education. Paper presented at Kolin

Kolistelut—Koli Calling: 2nd Annual Finnish/Baltic

Sea Conference on Computer Science Education, Koli,Finland, October 2002.

Ben-Ari, M. (2004). Situated learning in computerscience education, Computer Science Education,

14(2), 85–100.

Ben-David Kolikant, Y. (2004). Learning concurrency:Evolution of students’ understanding of

44

synchronization. The International Journal of Human

Computers Studies, 60(2), 243–268.

Ben-David Kolikant, Y., & Pollack, S. (2004a).Establishing computer science professional normsamong high school students. Computer Science

Education, 14(1), 21–35.

Ben-David Kolikant, Y., & Pollack, S. (2004b).Community-oriented pedagogy for inservice CSteacher training. Proceedings of the ITiCSE 2004

Conference, Leeds, United Kingdom, 191–195.

Ben-David Kolikant, Y. (2005). Students’ alternativestandards for correctness. Proceedings of the First

International Computing Education Research

Workshop, October 1–2, 2005, University ofWashington, Seattle, WA, 37–43.

Bergin, S., & Reilly, R. (2005). Programming: Factorsthat influence success. Proceedings of the SIGCSE

2005 Conference, St. Louis, Missouri, USA, 411–415.

Böszörmenyi, L. (2005). Teaching: People to people—about people: A plea for the historic and humanview. In R. T. Mittermeir (Ed.) From computer

literacy to informatics fundamentals, Proceedings of

International Conference on Informatics in Secondary

Schools—Evolution and Perspectives, ISSEP 2005,

Klagenfurt, Austria, March/April 2005, 93–103.

Campbell, P. F., & McCabe, G. P. (1984). Predicting thesuccess of freshmen in a computer science major.Communications of the ACM, 27(11), 1108–1113.

Cantwell Wilson, B., & Shrock, S. (2001). Contributingto success in an introductory computer sciencecourse: A study of twelve factors. SIGCSE Bulletin,

33(1), 184–188.

Cantwell Wilson, B. (2002). A study of factorspromoting success in computer science includinggender differences. Computer Science Education,

12(1–2), 141–164.

Cartelli, A. C. (2002). Computer science education inItaly: A survey. SIGCSE Bulletin, 34(4), 36–39.

Computer Science Teachers Association. (2005).Results of the national secondary computerscience survey. Retrieved October 16, 2005, fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html

Collofello, J. S. (2002). Creation, development andevaluation of an innovative secondary schoolsoftware development curriculum module.Proceedings of the 32nd ASEE/IEEE Frontiers in

Education Conference, Boston, MA.

Crombie, G., Abarbanel, T., & Anderson, C. (2000).All-female computer science: The positive effectsof single gender classes. The Science Teacher, March2000, 40–43.

Dagienë, V. (2005). Teaching information technologyin general education: Challenges andperspectives. In R. T. Mittermeir (Ed.) From

computer literacy to informatics fundamentals,

Proceedings of International Conference on Informatics

in Secondary Schools—Evolution and Perspectives,

ISSEP 2005, Klagenfurt, Austria, March/April2005, 53–64.

Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C.,Tucker, A., Turner, A. J., & Young, P. R. (1988). Draft report of the ACM Task Force on the

Core of Computer Science. New York, NY:Association for Computing Machinery.

Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C.,Tucker, A., Turner, A. J., & Young, P. R. (1989).Computing as a discipline. Communications of the

ACM, 32(1), 9–23.

Denning, P. J. (2004). Great principles in computing curricula. Proceedings of the SIGCSE

2004 Conference, Norfolk, Virginia, USA, 336–341.

45

Dorninger, C. (2005). Educational standards in schoolinformatics in Austria. In R. T. Mittermeir (Ed.)From computer literacy to informatics fundamentals,




du Boulay, B., O’Shea, T., & Monk, J. (1989). The blackbox inside the glass box: Presenting computingconcepts to novices. In E. Soloway & J. C. Spohrer(Eds.). Studying the novice programmer (pp.431–446). Hillsdale, NJ: Lawrence ErlbaumAssociates.

Driscoll, M. (1987). Ten case studies from a study of

exemplary mathematics programs. Reston, VA:National Council of Teachers of Mathematics.

Eurydice (Information Network on Education inEurope). (2005). Information and communication

technologies in education. Retrieved January 25,2005 from http://www.eurydice.org/Documents/Bibliographie/en/frameset_biblio_refs_en.html

Fadi, P. D., & McHugh, J. A. (2000). Problem-solvingmethodologies and the development of criticalthinking skills. Journal of Computer Science

Education, 14(1&2), 6–12.

Fadi, P. D., & McHugh, J. A. (2001). Prototypesoftware development tools for beginningprogramming. Journal of Computer Science

Education, 14(3&4), 14–20.

Felleisen, M., Findler, R. B., Flatt, M., &Krishnamurthi, S. (2002). The structure andinterpretation of the computer science curriculum.Journal of Functional Programming, 14(4), 365–378.

Fincher, S. (1999). What are we doing when we teachprogramming? Proceedings of the FIE1999

Conference, San Juan, Puerto Rico, 12a4–1–12a4–4.

Fincher, S., & Petre, M. (2004). Computer science

education research. London, UK: Routledge Falmer.

Fisher, A., & Margolis, J. (2002). Unlocking theclubhouse: The Carnegie Mellon experience.SIGCSE Bulletin, 34(2), 79–83.

Foley, J. (2004). Old challenges, new opportunities.Computing Research News, 16(4). RetrievedDecember 1, 2005 from http://www.cra.org/CRN/articles/sept04/foley.html

Franklin, R. (1987). What academic impact are highschool computing courses having on the entry-level computer science curriculum? Proceedings of

the SIGCSE 1987 Conference, St. Louis, Missouri,USA, 253–256.

Gal-Ezer, J., Beeri, C., Harel, D., & Yehudai, A. (1995).A high school program in computer science.Computer, 28(10), 73–80.

Gal-Ezer, J., & Harel, D. (1998). What (else) shouldCS educators know? Communications of the ACM,

41(9), 77–84.

Gal-Ezer, J., & Zeldes, A. (2000). Teaching softwaredesigning skills. Computer Science Education, 10(1),25–38.

Gal-Ezer, J., & Zur, E. (2002). The concept of “algorithmefficiency” in the high school CS curriculum.Proceedings of the 32nd ASEE/IEEE Frontiers in

Education Conference, Boston, MA, T2C1–T2C6.

Gal-Ezer, J., Vilner, T., & Zur, E. (2003). Characteristics

of students who failed (or succeeded) the introductory

CS course. Paper presented at the FIEE 2003

Conference. Boulder, CO. Retrieved December 4,2005 from http://fie.engrng.pitt.edu/fie2003

Ginat, D. (1996). Efficiency of algorithms forprogramming beginners. Proceedings of the SIGCSE

1996 Conference. Philadelphia, PA, 256–260.

46

Ginat, D. (2000). Colorful examples for elaboratingexploration of regularities in high school CS1.Proceedings of ITiCSE 2000, Helsinki, Finland,81–84.

Goldweber, M., Fincher, S., Clark, M., & Pears. A.(2004). The relationships between CS educationresearch and the SIGCSE community. SIGCSE

Bulletin, 36(1), 147–148.

Graham, S., & Latulipe, C. (2003). CS girls rock:Sparking interest in computer science anddebunking the stereotypes, Proceedings of the

SIGCSE 2003 Conference, Reno, Nevada, USA,322–326.

Grandell, L. (2005). High school students learninguniversity level computer science on the web—acase study of the DASK-model. Journal of

Information Technology Education, 4, 207–218.

Greening, T. (1998). Computer Science: Through theeyes of potential students, Proceedings of the ACSE

1998 Conference, Brisbane, Australia, 145–154.

Gries, D. (2002). Where is programming methodologythese days? SIGCSE Bulletin, 34(4), 5–7.

Gurbiel, E., Hardt-Olejniczak, G., & Kolczyk, E.(2005). Informatics and ICT in the Polisheducation system. In R. T. Mittermeir (Ed.) From





Gupta, U. G., & Houtz, L. E. (2000). High schoolstudents’ perceptions of information technologyskills and careers. Journal of Industrial Technology,

16(4), 2–8.

Guzdial, M. and Soloway, E. (2003). Computerscience is more important than calculus: The

challenge of living up to our potential. SIGCSE

Bulletin, 35(2), 5–8.

Haberman, B., & Ginat, D. (1999). Distance learningmodel with local workshop sessions applied to in-service teacher training, Proceedings of the ITiCSE

1999 Conference, Krakow, Poland, 64–67.

Haberman, B., & Ben-David Kolikant, Y. (2001).Activating ’black boxes’ instead of opening’zippers’—A method of teaching novices basic CSconcepts. Proceedings of the ITiCSE 2001 Conference,

Dublin, Ireland, 41–44.

Haberman, B., & Averbuch, H. (2002). The case of basecases: Why are they so difficult to recognize?Student difficulties with recursion. Proceedings of the

ITiCSE 2002 Conference, Aarhus, Denmark, 84–88.

Haberman, B., Shapiro, E., & Scherz, Z. (2002). Areblack boxes transparent?—High school students’strategies of using abstract data types. Journal of

Educational Computing Research, 27(4), 411–436.

Haberman, B., Lev, L., & Langley, D. (2003). Actionresearch as a tool for promoting teacherawareness of students’ conceptual understanding.Proceedings of the ITiCSE 2003 Conference,

Thessaloniki, Greece, 144–148.

Haberman, B. (2004a). High school students’ attitudesregarding procedural abstraction: Special issuedevoted to recent research projects of secondaryinformatics education. Education and Information

Technologies, 9(2), 131–145.

Haberman, B. (2004b). How learning logicprogramming affects recursion comprehension.Computer Science Education, 14(1), 37–53.

Haberman, B., Averbuch, H., &. Ginat, D. (2005). Is itreally an algorithm?—The need for explicitdiscourse. Proceedings of the ITiCSE 2005

Conference, Lisbon, Portugal, 74–78.

47

Hagan, D., & Markham, S. (2000). Does it help tohave some programming experience beforebeginning a computing degree program? SIGCSE

Bulletin, 32(3), 25–28.

Harris, J. W. (1987). Teacher and leader—why we mustbe both. Vocational Education Journal, 62(5), 26–28.

Hazzan O., Impagliazzo, J., Lister, R., & Schocken, S.(2005). Using history of computing to addressproblems and opportunities, Proceedings of the

SIGCSE 2005 Conference, St. Louis, MO, 126–127.

Hazzan, O., & Lapidot, T. (2004a). Construction of aprofessional perception in the “Methods ofTeaching Computer Science” course. SIGCSE

Bulletin, 36(2), 57–61.

Hazzan, O., & Lapidot, T. (2004b). The practicum incomputer science education: Bridging the gapbetween theoretical knowledge and actualperformance. SIGCSE Bulletin, 36(4), 47–51.

Holmboe, C., McIver, L., & George, C. (2001). Researchagenda for computer science education. Proceedings

of the 13th Workshop of the Psychology of Programming

Interest Group, Bournemouth, UK, 207–233.

Inos, R. H., & Quigley, M. A. (1995). Synthesis of the

research on educational change. Part 4: The teacher’s

role. Honolulu, HI: Pacific Region EducationalLaboratory.

International Society for Technology in Education(ISTE). (2002) National educational technology standards

for students. Retrieved December 1, 2005, fromhttp://www.iste.org/netsImpagliazzo, J., & Lee, J. A.N. (2004). Using computing history to enhanceteaching. Proceedings of the History of Computing in

Education 2004 Conference, Toulouse, France, 165–176.

Jeffrey, N. (1996). The responsibility of professionalassociations. Paper presented at A Meeting of the

Minds: ITEC Virtual Conference 1996, Australia.

Kavander, T., & Salakoski, T. (2004). Where have allthe flowers gone? Computer science education ingeneral upper secondary school. Proceedings of the

Kolin Kolistelut—Koli Calling Conference, Koli,Finland, 112–115.

Kuznetsov, A. A., & Beshenkov, S. A. (2005). Russianeducational standards of informatics andinformatics technologies (ICT): Aims, content,perspectives. In R. T. Mittermeir (Ed.) From computer





Lapidot, T., & Hazzan, O. (2003). Methods of teachingcomputer science courses for prospective teachers.SIGCSE Bulletin, 35(4), 29–34.

Latulipe, C., & Graham, S. (2005). Degaming thecurriculum: A failed experiment. RetrievedDecember 10, 2005, from: http://hci.uwaterloo.ca/students/clatulip/research.html

Lee, P. (2004). The computer science brain drain: A call

to revitalize computer science education. RetrievedApril 23, 2004, from https://www.wiki.cs.cmu.edu/public/uploads/Main/it-talent.pdf

Levy, D., & Lapidot, T. (2002). Shared terminology,private syntax: The case of recursive descriptions.Proceedings of the ITiCSE 2002 Conference, Aarhus,Denmark, 89–93.

Loidl, S., Muhlbacher, J., & Scauer, H. (2005).Preparatory knowledge: Propaedeutic ininformatics. In R. T. Mittermeir (Ed.) From





Merritt, S. (1995). Reflections of a computer scientist

48

for teachers and teacher educators. In J. D. Tinsleyand T. J. van Weert (Eds.). Proceedings of the World

Conference on Computers in Education VI (4790486).London: Chapman and Hall for the InternationalFederation of Information Processing.

Micheuz, P. (2005). 20 years of computers andinformatics in Austria’s secondary academicschools. In R. T. Mittermeir (Ed.) From computer



Schools—Evolution and Perspectives, ISSEP 2005,Klagenfurt, Austria, March/April 2005, 20–31.

Mitchell, W. (2002). Information technologyeducation: One state’s experience. Journal of

Computing in Small Colleges, 17(4), 123–132.

Mittermeir, R. T. (Ed.) (2005). From computer literacy to

informatics fundamentals. International Conference on

Informatics in Secondary Schools-—Evolution and

Perspectives, ISSEP 2005, Klagenfurt, Austria,March/April 2005.

Moorman, P., & Johnson, E. (2003). Still a strangerhere: Attitudes among secondary school studentstowards computer science. Proceedings of the

ITiCSE 2003 Conference, Thessaloniki, Greece,193–197.

Morris, J. H. and Lee, P. (2004). The incrediblyshrinking pipeline is not just for women anymore.Computing Research News, 16(1), 20.

Muller, O., Haberman, B., & Averbuch, H. (2004). (Analmost) pedagogical pattern for pattern-basedproblem-solving instruction. Proceedings of the

ITiCSE 2004 Conference, Leeds, United Kingdom,102–106.

Muller, O. (2005). Pattern oriented instruction and theenhancement of analogical reasoning. Proceedings of

The First International Computing Education Research

Workshop, October 1–2, 2005. Seattle, WA, 57–67.



O’Lander, R. (1996). Factors effecting high schoolstudent’s choice of computer science as a major.Proceedings of the Symposium on Computers and the

Quality of Life, Philadelphia, Pennsylvania, USA,25–31.

Paz, T., & Lapidot, T. (2004). Emergence of automatedassignment conceptions in a functionalprogramming course. Proceedings of the ITiCSE

2004 Conference, Leeds, United Kingdom, 181–185.

Payton, F. C. (2003). Rethinking the digital divide.Communications of the ACM, 46(6), 89–91.

Peckham, J., DiPippo, L., Reynolds, J., Paris, J.,Monte, P., & Constantinidis, P.A. (2000), Firstcourse in computer science: The discipline is more than programming. Journal of Computing

in Small Colleges, 15(5), 223–230.

Perkins, D., Schwartz, S., & Simmons, R. (1988).Instructional strategies for the problems of noviceprogrammers. In R. E. Mayer (Ed.), Teaching and

learning computer programming (pp. 153–178).Hillsdale, NJ: Lawrence Erlbaum Associates.

Pham, B. (1997). The changing curriculum ofcomputing and information technology inAustralia. Proceedings of the ACSE 1997 Conference,

Melbourne, Australia, 149–154.

Poirot, J. L. (1979). Computer education in thesecondary school: Problems and solutions.Proceedings of the SIGCSE 1979 Conference, Dayton,Ohio, 101–104.

Poirot, J. L., Taylor, H. G., & Norris, C. A. (1988).Retraining teachers to teach high school computerscience. Communications of the ACM, 37(7), 912–917.

49

Pollack, S., & Scherz, Z. (2005). Supporting projectdevelopment in CS – the effect on intrinsic andextrinsic motivation. Proceedings of the 10th PEG

Conference, Tampere, Finland, 143–148.

Ramberg, P. (1986). A new look at an old problem:Keys to success for computer science students.ACM SIGCSE Bulletin, 18(3), 36–39.

Ragonis, N., & Ben-Ari, M. (2005). A long-terminvestigation of the comprehension of OOPconcepts by novices. Computer Science Education,

15(3), 203–221.

Rich, L., Perry, H., & Guzdial, M. (2004). A CS1course designed to address interests of women.Proceedings of the SIGCSE 2004 Conference, Norfolk,Virginia, USA, 190–194.

Reiter, A. (2005). Incorporation of informatics inAustrian education: The project “Computer-Education-Society” in the school year 1984/85. InR. T. Mittermeir (Ed.) From computer literacy to

informatics fundamentals, Proceedings of International

Conference on Informatics in Secondary Schools—

Evolution and Perspectives, ISSEP 2005, Klagenfurt,Austria, March/April 2005, 4–19.

Roberts, E. (2004). The dream of a common language:The search for simplicity and stability incomputer science education. Proceedings of the

SIGCSE 2004 Conference, Norfolk, Virginia, USA,115–119.

Roberts, E,. & Halopoff, G. (2005). Computer ScienceTeachers Association analysis of high schoolsurvey data. Retrieved December 2, 2005 fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html

Romberg, T. A., & Middleton, J. A. (1995).Conceptions of mathematics and mathematicseducation held by teachers. In N. Webb & T. A.Romberg (Eds.). Collaboration as a process for

reform. New York: Teachers College Press.

Rountree, N., Vilner, T., Wilson, B., & Boyle, R. (2004).Predictors for success in studying CS, panelsession. Proceedings of the SIGCSE 2004 Conference,

Norfolk, Virginia, USA, 145–146.

Rosenthal, T., Suppes, P., & Ben-Zvi, N. (2002).Automated evaluation methods with attention toindividual differences—A study of a computer-based course in C, Proceedings of the ASEE/IEEE

Frontiers in Education Conference, Boston, MA.

Sabin, M., Higgs, B., Riabov, V., & Moreira, A. (2005).Designing a pre-college computing course. Journal

of Computing in Small Colleges, 20(5), 123–132.

Scherz, Z., & Haberman, B. (2005). Mini-projectsdevelopment in computer science—Students’ useof organization tools. Informatics in Education, 4,

307–319.

Schollmeyer, M. (1996). Computer programming inhigh school vs. college. Proceedings of the SIGCSE

1996 Conference, Philadelphia, PA, USA, 378–382.

Sims-Knight, J. E., & Upchurch, R. L. (1993). Teachingsoftware design: A new approach to high schoolcomputer science. Paper presented at The Annual

Meeting of the American Educational Research

Association, Atlanta, GA. Retrieved November 1,2005 from http://www2.umassd.edu/cisw3/people/faculty/rupchurch/

Sleeman, D., Putnam, R. T., Baxter, J. A, & Kuspa, L.(1989). A summary of misconceptions of highschool basic programmers. In E. Soloway & J. C.Spohrer (Eds.). Studying the novice programmer (pp.301–314). Hillsdale, NJ: Lawrence ErlbaumAssociates.

Stephenson, C. (2002). High school computer scienceeducation. Journal of Computer Science Education,

Annual, 30–40.

50

51

Stevenson, D. E. (1993). Science, computationalscience and computer science: At a crossroads.Proceedings of the 1993 ACM Conference on

Computer Science, Indianapolis, IN, 7–14.

Stiller, E., & LeBlanc, C. (2003). Creating new computerscience curricula for the new millennium. Journal of


Taylor, H. G., & Mounfield, L. C. (1989). The effect ofhigh school computer science, gender, and workon success in college computer science. ACM

SIGCSE Bulletin, 21(1), 195–198.

Taylor, H. G., & Mounfield, L. C. (1991). An analysis ofsuccess factors in college computer science: Highschool methodology is a key element. Journal of

Research on Computing in Education, 24(2), 240–245.

The College Board. (2005). Advanced Placement report

to the nation. Retrieved December 22, 2005, fromhttp://www.collegeboard.com/about/news_info/ap/2005/index.html





Tucker, A., McCowan, D., Deek, F., Stephenson, C.,Jones, J., & Verno, A. (2004). Implementationchallenges for K–12 computer science curriculum.Proceedings of the SIGCSE 200404 Conference,

Norfolk, VA, 334–335.

Vandenberg, S., & Wollowski, M. (2000). Introducingcomputer science using a breadth-first approachand functional programming. Proceedings of the

SIGCSE 2000 Conference, Austin, TX, USA,180–184.

Ventura, P., & Ramamurthy, B. (2004). Wanted: CS1students. No experience required. Proceedings of

the SIGCSE 2004 Conference, Norfolk, Virginia,USA, 240–244.

CHAPTER THREE

USING FRANK’S FRAMEWORK TO EXPLORE THEDEVELOPMENT AND IMPLEMENTATION OF HIGH SCHOOL

COMPUTER SCIENCE CURRICULA IN FIVE COUNTRIES

52

3.0 INTRODUCTION

In recent decades, computers have come to occupy apivotal place in our work, personal, and home lives.Accordingly, the field of computer science hasexpanded to encompass the technical, innovative,conceptual, and psychological ramifications of thepresence of the computer at work and at home. Muchresearch (National Research Council, 1999) confirmsthe urgent need to improve the level of publicunderstanding of computer science as an academicand professional field. In fact, the lack of a currentcommitment to computer science education at alllevels is recognized as a major factor in the shortage ofcomputer science professionals that continues to affectindustries world-wide (Bureau of Labor Statistics,2005). Many students receive their first exposure tomany scientific disciplines in high school (secondaryschool). Lack of support and implementation ofappropriate learning opportunities in computerscience for students at this level therefore significantlyaffects the long-term national economic viability, notjust in the high tech industries, but in all industries.

Although a significant amount of research has beenconducted on teaching specific computer scienceconcepts or using specific software tools for instruction,minimal attention has been paid to developing a solidset of instructional and content standards for pre-college computer science. As Tucker, Deek, Jones,McCowan, Stephenson, and Verno. (2003) noted:

Computer science is an established discipline atthe collegiate and post-graduate levels. Oddly, theintegration of computer science concepts into theK-12 curriculum has not kept pace in the UnitedStates. As a result, the general public is not as

well educated about computer science as it shouldbe, and a serious shortage of informationtechnologists at all levels exists and may continueinto the foreseeable future. (p. 3)

For this reason, the United States may find itself atconsiderable disadvantage when compared to othercountries that have already implemented or arecurrently developing a comprehensive computerscience curriculum for K–12 education.

This paper uses Frank’s (1972) framework forcritiquing educational policy and reform to analyzethe transcript of a National Science Foundation-sponsored panel entitled Finding a Computing

Curriculum that Fits: the United States and Beyond

(Verno, Chiles, Gal-Ezer, Martin & Stephenson, 2005).The panel, which took place at the NationalEducational Computing Conference (NECC) inPhiladelphia in June 2005, brought togethercurriculum experts from five countries: Canada,Israel, Scotland, South Africa, and the United States.Its goal was to discover elements that could beidentified as contributing to the successfuldevelopment and implementation of a national highschool computer science curriculum.

3.1 METHODOLOGY

3.1.1 Method of Analysis

This paper collects the observations of five educatorsfrom five different countries who participated in apanel at NECC in 2005. Panel organizers selected thecountries based upon their varying levels of

commitment to a national high school computerscience curriculum. The individual panelists wereselected on the strength of their individual and nationalcurriculum development efforts, their personalactivism in supporting excellence in computer scienceeducation nationally, and their ability to illuminate keyimplementation issues and strategies. They alsorepresented various stakeholder constituencies,including researchers, curriculum developers, andeducators. The following table contains a list of thepanelists and the countries they represented.

Data for this study is drawn from a single source: arecorded and then professionally prepared transcript ofthe one-hour panel session. Employing the constantcomparative method (Strauss, 1987), data analysisbegan with the first presentation and continuedthroughout the document. A coding framework forthree main categories was drawn directly from thestructure of the panel presentation (curriculum models,curriculum dissemination and implementation, andassessment). This framework was further revised andexpanded as additional themes emerged. To ensureaccuracy of reporting, the panelists and the panel chairreviewed the transcript drafts. The transcript was thencoded, and 58 illustrative segments ranging in lengthfrom 30 to 187 words were copied from the word-processed transcript into a database file.

Analysis began with the grouping of similarlycoded items (e.g., implementation obstacles). Thetranscript was then systematically analyzed bysearching for instances of all coded items in eachspeaker’s segment of the presentation. The data werefirst analyzed by presenter and then across all fivepresenters. All of the coded data were then furthersorted and compared using Frank’s (1972) frameworkfor critiquing educational policy and reform.

3.1.2 Employing a Framework

Evaluating any educational phenomenon is a complextask and can be approached from many differenttheoretical perspectives or frameworks, each of whichmay provide a variety of analytical tools andperspectives. The benefit of such a framework, asarticulated by Kubow and Fossum (2003), is that it canbe used to analyze the complex factors, motivations,and ambiguities shaping educational policy andreform. It is important to note, however, that no singleframework can truly encompass education’scomplexity and, in fact, that even the best frameworkswill “limit reflective inquiry to a select number of itemsfor analysis, thereby privileging some information andfactors over others and allowing some insights to begained while excluding from view other factors thatmight influence interpretation of the issue” (p. 235).

This paper uses Frank’s (1972) framework, whichprovides a set of three criteria or elements: policy

effectiveness, theoretical adequacy, and empirical validity.The policy effectiveness criterion requires an analysis ofpolicies and programs so that sufficient supports havebeen put into place to ensure that intended outcomesare met. In the case of a computer science curriculum,for example, this would involve an examination ofthe fiscal and structural mechanisms that have beenput in place to support curriculum development andimplementation and to ensure students have access to appropriate learning/resource materials. Thedetermination of theoretical adequacy can be seen as restingupon the extent to which the strategies used are sufficientand appropriate to achieve the expected outcomes. Theempirical validity criterion can be applied to questionsrelating to the extent to which the evaluation of the policyinitiatives are grounded in research, that is, whetherscientifically proven methods have been employed todetermine the success or failure of the reform.

3.1.3 Understanding Context

Before analyzing the observations of the panelparticipants, it is essential to place them within their

53

PANELIST COUNTRY/PROVINCE

Anita Verno (Chair) United States

Chris Stephenson Canada (Ontario)

Judith Gal-Ezer Israel

Jackie Martin Scotland

Michael Chiles South Africa

specific educational contexts. The following sectiontherefore provides a brief description of the highschool computer science curriculum in each of the fivecountries represented. It should be noted that whilemany of the countries do not identify specific years ofinstruction by a grade level (for example Grade 10),the panelists attempted to give what would be theU.S. grade equivalent (where possible) to make iteasier to compare curricula across countries.

3.1.3.1 Canada

In Canada, K–12 education policy and curriculum aredeveloped and administered at the provincial level.While many provinces have implemented some formof Information and Communications Technologycurriculum, Ontario has mandated a ComputerStudies curriculum consisting of Computer Science andComputer Engineering courses (Ministry of Education,1999 & 2000). The Ontario curriculum consists of a setof eight courses beginning in Grade 9. The firstcourse (Integrated Technologies) is intended as ageneral introduction to computing technologies,including computer applications and a small amountof programming. In Grade 10, the curriculum dividesinto two distinct streams: the Computer andInformation Science stream and the ComputerEngineering stream.

The Computer and Information Science streamconsists of three courses to be taught in Grades 10–12.Each course must include a minimum of 110 contacthours. The course for Grade 10 introduces students tosoftware design concepts and programmingfundamentals (e.g. sequence, selection, andrepetition) as well as the relationship amongnetworks, operating systems, and applicationssoftware. The course for Grade 11 includes the studyof data control and data structures, problem solvingand programming, the software development cycle,and operating systems. The Grade 12 course requiresstudents to design and implement algorithms andprograms, use software development and diagnostictools, and use file management techniques in a

project setting. In addition, some of these coursescontain learning expectations relating to computingethics and careers.

The Computer Engineering stream consists of fourcourses to be taught in Grades 10–12. The singleGrade 10 course provides an introduction tocomputer hardware and the control of externalcomponents and includes the study of logic gates,internal numbering and character representationsystems, operating systems and networks, andfundamental programming concepts. Students inGrade 11 have a choice of two Computer Engineering

courses. The Grade 11 college/university preparationcourse is intended for students who want to studycomputer science or engineering at the college oruniversity level. This course focuses on the use ofcomputer hardware and software to solve problemsfrom an engineering perspective, requiring studentsto construct systems that use computer programs tointeract with hardware and to install and configurecomputer hardware and software components. TheGrade 11 workplace preparation course, however, isintended for students who wish to improve theirpractical understanding of hardware and softwareoperations, networks, and operating systems. Thiscourse focuses on installation, maintenance, andrepair procedures for computer systems andnetworks. The Grade 12 Computer Engineering courserequires students to develop and construct systemsand design and implement computer programs todrive real-world devices and to develop a thoroughunderstanding of networking hardware, protocols,and configurations.

3.1.3.2 Israel

Unlike Canada’s provincial curricula, Israel has anational curriculum for computer science. In Israel,the Ministry of Education sets all educational policyand then implements it with the aid of professionalcommittees and supervisors. Israel’s curriculum forhigh school computer science (Gal-Ezer, J. & Harel,D., 1999) was developed on the basis of the following

54

underlying principles:• Computer Science is a full-fledged scientific

subject.• The program should concentrate on the key

concepts and foundations of the field. • The program should focus on lasting concepts

of computer science, not on changingtechnology.

• Two different programs are needed withdiffering levels of requirements.

• Each of the programs should have requiredunits and electives.

• Conceptual and experimental issues should beinterwoven throughout the program.

• Two different programming paradigms shouldbe taught to provide different ways ofalgorithmic thinking or different ways ofsolving problems.

• A well-equipped and well-maintained computerlaboratory is mandatory.

• New course material must be written for allparts of the program by different teams indifferent academic institutions. The teams musthave computer scientists on board, as well ashigh school teachers and researchers incomputer science education.

• To be certified to teach computer science,teachers must have an adequate formalcomputer science education (at least anundergraduate degree in computer science).

The 450-hour curriculum is divided into five units of90 hours each. The first three units are intended forstudents seeking an introduction to computer science.Students wishing to continue with their study ofcomputer science at the university level would taketwo additional advanced units. The first unit(Fundamentals 1) covers the foundations of computerscience (problem solving, algorithmic thinking, andalgorithmic solutions to algorithmic problems). Thesecond unit (Fundamentals 2) introduces moreadvanced concepts such as recursion and two-dimensional arrays. The third unit (Applications or a

Second Paradigm) provides students with a choice of

elective courses. Students intending to major incomputer science can choose among coursesincluding: Logic Programming, Computer Organization

and Assembly Language, Information Systems, orGraphics. Students planning to complete only threeunits usually opt for an Applications course. The unitfour course (Software Design) focuses on abstract datatypes and data structures. The final unit course(Theory), again provides students with a choice of twoelectives: Computational Models or Numerical Analysis.

Although these operate as distinct courses within thecurriculum, they all focus on the key concepts andfoundations of computer science: algorithms,abstraction, and system design, and each of thecourses that comprise the curriculum emphasize thesefoundations (Gal-Ezer, Beeri, Harel, & Yehudai, 1995).

3.1.3.3 Scotland

In Scotland, the government provides recommendedcurriculum guidelines for schools in all areas. Whilethese guidelines are not mandatory, adherence islargely guaranteed through the linking of finalexamination content for Grades 10–12 to therecommended curriculum. Scotland’s computingcurriculum has been in place since 1984, with severalminor revisions and a major revision in 1999. TheScottish computing curriculum for high schools (TheScottish Parliament Information Centre, 1999)consists of a flexible network of courses covering abroad range of topics. It’s designed to both educatepupils about computers and also to provideapplication skills. At every stage of the curriculumframework, teaching, learning, and assessment followtwo principles: knowledge and understanding andthe application of that knowledge in a problem-solving scenario. Each course in the curriculumconsists of a combination of mandatory and requiredunits (three in total for each course). Studentevaluations are drawn from a combination ofcoursework, course projects, and a final examination.

In each grade, students choose from a selection ofcourses depending upon their interests and their

55

performance in preceding courses. Like Israel,Scotland offers a multi-level curriculum consisting ofa Standard Grade strand and a Higher strand. TheStandard Grade strand consists of an Intermediate 1Computing Studies course, an Intermediate 2 Computing

Studies course, and an Intermediate 2 Information

Systems course. The Higher strand consists of a Higher

Computing course, a Higher Information Systems

course, an Advanced Higher Computing course, andan Advanced Higher Information Systems course.Performance in the first Standard Grade Computing

Studies course is assessed at three levels: foundational,general, and credit. Pupils who achieve a foundationalcredit would progress from the first Standard GradeComputing Studies course to the Intermediate 1

Computing Studies course. Pupils who achieve ageneral credit would progress from the first Standard

Grade course to the Intermediate 2 course. Success inthis course would allow them to proceed to theHigher strand the following year. Students whoachieve a credit grade in the first Standard Grade

would move directly to the Higher Computing orInformation systems courses. Success in these courseswould then allow them to move onto an AdvancedHigher course. The Standard Grade course is a two-year course requiring 160 instructional hours while theIntermediate, Higher, and Advanced Higher courses areone-year courses with 120 hours of instructional time.

Scotland, however, is unique among therepresentatives in that it provides a strand of threecourses referred to as the Access Cluster, specificallydesigned for pupils who have disadvantages inlearning. The Access 1 and 2 courses concentrate onteaching life skills using computer technology.Students who succeed in the Access 1 and 2 coursescan go on to do Access 3, a very basic computingskills course. From this course they can go on to doIntermediate 1 and perhaps Intermediate 2 in Grade 12.

3.1.3.4 South Africa

As is the case with Israel and Scotland, educationalpolicy in South Africa is determined at the national

level, but like Ontario, it is implemented at theprovincial level. Over the past 10 years, the SouthAfrican National Department of Education has beeninvolved in a complete revision of its curricula (fromGrades 1–12), including its various computingcurricula (Department of Education, 2005a, 2005b).The new computing curricula have been designed tobe scenario-based, and so, where possible, examplesare drawn from everyday life and integrated withother subject areas. As in Ontario, South Africa’scomputer studies curriculum consists of two distinctstrands: Computer Applications Technology and

Information Technology. The Computer Applications

Technology strand focuses on end-user applicationsfor Grades 10–12. The Information Technology strand,on the other hand focuses on the use of informationand communications technologies in social andeconomic applications, with an emphasis on systemanalysis, problem solving, logical thinking, andinformation management and communicationthrough the use of various software developmenttools (e.g. object-oriented programming). Studentswho opt to take the Information Technology strand willtake one course per year in each of Grades 10, 11,and 12. All subjects in the new curricula to beintroduced in 2006 will have 4 to 5 hours ofinstruction time per week with roughly 200 days (or40 weeks) in the year available for school work. Ofthis tuition time, some will be used for contact time,some for practical (lab) work, and some for formalassessment (tests and examinations).

Course contents include algorithmic design,hardware and system software, networking, humancomputer interfacing, programming, data structures,databases and spreadsheets, testing and userinterfaces, and management information systems.The Information Technology course is organized aroundfour learning outcomes with the expectation that apercentage of the time in each course will be spent ineach.

• Hardware and system software (20%)• E-communications (12%)• Social and ethical issues (8%)• Programming and software development (60%).

56

As the student progresses through the courses, thecontent and level of complexity for each of theselearning outcome areas increases appropriately untilthe point where, at the end of Grade 12, the studentsare expected to be able to design and implement arelatively complex, real-world application.

3.1.3.5 United States

Although there is, at present, no national curriculumfor high school computer science education, theComputer Science Teacher’s Association (CSTA) isactively promoting the Model Curriculum for K–12

Computer Science Education published by theAssociation for Computing Machinery (Tucker, Deek,Jones, McCowan, Stephenson, & Verno, 2003). Thismodel curriculum is based upon the principle thatstudents require an understanding of the place ofcomputer science in the modern world and that thediscipline of computer science interleaves bothprinciples and skills.

The ACM Model Curriculum consists of four levelsof courses with the expectation that all studentsshould complete at least the first two levels. Level 1(Foundations of Computer Science) is recommended fordelivery at the K–8 level and consists of modulesintended for use in conjunction with existing studiesacross the curriculum. The learning outcomes for thislevel are drawn primarily from the National

Educational Technology Standards (ISTE, 2002) butinclude additional outcomes relating to problemsolving and algorithmic thinking. Level II (Computer

Science in the Modern World) is intended as a firstcomputer science course for all high school students.It provides a broad overview of the discipline withthe goal of preparing students for the technologicalworld. Conceptual content includes a fundamentalunderstanding of operating systems, networks, theInternet, problem solving, programming, careers, andissues in computing ethics. The Level III course(Computer Science as Analysis and Design) is a pre-Advanced Placement course in that it focuses onscientific and engineering principles. This course is

intended for students interested in pursuing moreadvanced studies in Computer Science, Engineering,or Engineering Technology at the college oruniversity level. Finally, the Level IV course isintended as a special projects course that wouldallow students to focus on a specific area ofcomputing in which they are especially interested. Itis intended to encompass a broad range ofspecialized courses such as AP Computer Science orcourses leading to industry certification.

3.2 EXPLORING CURRICULUMDEVELOPMENT ANDIMPLEMENTATION USINGFRANK’S FRAMEWORK

3.2.1 Policy Effectiveness

Frank (1972) posits that evaluating the feasibility of agiven educational policy or change requiresconsideration of three key aspects of policyeffectiveness: support and dissent, monetary andother costs, and the consequences of options.Analysis of the panel transcript illustrates that thepanelists from the five participating countriesprovide ample evidence of Frank’s argumentconcerning the importance of these aspects as theyrelate to the implementation of a high schoolcomputer science curriculum.

The issue of support and dissent, for example,appears frequently in the panelists’ comments,primarily in relation to discussions of the originatingsite of, or impetus for, the curriculum change. Thepanelists from Canada, Israel, and South Africa, forexample, note that the computer science curriculumrevision in their countries was very much a top-downprocess. Stephenson, for example, notes that inOntario:

In 1999 a new government was elected and theydecided, “Let’s start all over again in everysubject area in the secondary or high school level.We’re going to rewrite every single piece ofcurriculum in the province.” And so courses were

57

developed and implemented one grade level peryear. (p. 6)

Panelists Gal-Ezer (Israel) and Chiles (South Africa) alsoindicate that the computer science curriculum reform intheir countries were similarly conceived of and driven bythe government body responsible for education. Martin,however, indicates that while the Scottish governmentplays a central role in the development of the nationalcurriculum, its adoption and implementation is a matterof choice rather than mandate.

Unlike the rest of the UK, the Scottish curriculum isnot actually statutory. It is the responsibility of headteachers, but most schools do tend to follow thecurriculum guidelines that were set down by ourgovernment as though they were statutory. (p. 10)

Despite the top-down decision-making evident in themajority of the represented countries, severalpanelists also noted their governments attempted toimprove the viability of proposed educationalreforms by engaging the support of key stakeholdergroups at several stages of the process. Stephensonreports that in Ontario, for example, the governmentworked closely with the Association for ComputerStudies Educators (ACSE) in the design of the newcomputer science curriculum. It allowed ACSE toselect the team members with the goal of ensuringthat the curriculum would be written by practitionerswith a thorough understanding of the knowledgebase of the subject area.

Stephenson further indicates that once thedocument was written, key stakeholder groups,including representatives from industry and parentgroups, were also brought together to review anddiscuss the new curriculum prior to itsimplementation. A similar curriculum vettingprocess, specifically designed to address issues ofboth rigor and equity, was conducted in South Africa,about which Chiles notes:

The implementation and review process involvedmany groups of people, including the teachersthemselves, teacher unions (teacher unions arevery strong in our country), the Human Rights

Commission (because of the inequities of the pastwe had to include what was happening withinhuman rights), and the tertiary institutions. (p. 23)

The involvement of tertiary or post-secondaryinstitutions was also a key element in the policyimplementation process in Israel, where teams ofdiscipline specialists subjected the curriculum to arigorous review even after its implementation, and inScotland where universities played a key role in on-going professional development for teachers.

According to Verno, however, the situation in theUnited States differs markedly from that of the othercountries represented on the panel. The United Statesis conspicuous in its lack of engagement withcomputing curriculum policy at both the federal andstate levels. In this absence of governmental action,efforts to define a national curriculum for computerscience in the United States are being driven by theComputer Science Teachers Association (CSTA), aprofessional membership organization representingK–12 computing teachers. In its efforts to promote arigorous and consistent computer science curriculum,CSTA plays a prime role in the continueddevelopment and dissemination of the ACM Model

Curriculum for K–12 Computer Science Education. It isalso developing and disseminating supportingresources and teaching and learning materials, andworking to improve teacher access to professionaldevelopment through workshops and symposia.

As further evidence of the accuracy of Frank’sframework, the discussion also demonstratesconsiderable panelist engagement with issues relatingto funding and resource access. With the exception ofthe United States (where there has been nogovernmental support for the development ofimplementation of a national computer sciencecurriculum) all of the panelists note the extent towhich the goals of the curriculum reform could onlybe achieved with significant financial investment bythe government. Gal-Ezer, for example, indicates thatin Israel, “The Ministry of Education very generouslybudgeted the program. The budgets covered thedevelopment of the curriculum and the development

58

of the materials, including the salary of thedevelopers” (p. 19). Martin notes a similar level ofon-going support from the Scottish government:

Schools were given a massive government cashinjection mainly to fund multimedia technology,which was going to prove to be the most costlypart of the curriculum to implement. But it wasstill not enough, and half way through they gaveus another 80 million pounds. (pp. 21–22)

Many of the panelists, however, note that access toresources was problematic. Chiles indicates thatimplementation of the new computer sciencecurriculum in South Africa has been seriouslyimpeded by a lack of infrastructure funding, whichserves to exacerbate inequities already present in theeducation system.

Funding is a major problem in that thegovernment provides no funding whatsoever toschools in terms of implementation of many ofthe subjects. They will provide somethingtowards the training, but in terms of theimplementation, the physical infrastructure that isrequired, there is no funding whatsoever. And soyou can imagine what is going to happen in asituation such as we have currently, that theadvantaged schools—the previously whiteschools that come from richer communities—willhave all of the equipment; the disadvantagedschools—previously the black and Indian, coloredcommunities—will have very little funding andwill not be able to provide the equipmentthemselves. (p. 25)

Problems with hardware were also a factor in theinitial implementation of the Scottish curriculum as,according to Martin, the Exam Board computersystem required to process the results of studentexaminations (which are key to student placement atuniversities) was not installed until the day it was tobe used, and then was found to have bugs, whichresulted in incorrect marks or no marks at all beinggenerated for a considerable number of students.

In both Ontario and Scotland, access to

curriculum policy documents and support materialswas problematic. According to Stephenson, Ontarioteachers responsible for implementing the newcurriculum had difficulty accessing the curriculumdocuments.

The curriculum documents were all posted onlineand every school received just one set ofdocuments for all courses. So if you had seventeachers in your department, they were sharingone set of curriculum [documents] … Thesedocuments were received in the schools anywherebetween September 15th and September 30th, andschool started September 6th. (p. 15)

While all of the panelists articulate the importance ofadequate funding for teacher training as a key factorin successful curriculum implementation, theexperiences of the panelists differ significantly on thisissue. With the exception of the United States, all ofthe governments represented on the panel providedsome funding for teacher training. Martin noted, forexample, that teacher training on the Scottishcurriculum was provided at national, regional, andlocal levels.

We had documentation sent to centers fordiscussion and that was followed-up by nationalmeetings with headmasters and headmistresses,and regional meetings with principle teachers ofthe various subjects. Some teachers were secondedto become national trainers. They went around toregional meetings training principle teachers. Wedid this for a year and a half, and it worked verywell. The final stage of dissemination took place inthe form of in-service training days. Normallythese would take place within a school, but for thepurpose of standardization, what happened wasthat, within a local authority, one school wouldhost an in-set for all the subject teachers withinthat local authority, and they attended the samein-set. (p. 20)

According to Chiles, although to a much morelimited extent, the government of South Africa alsoprovided training for teachers implementing the new

59

Grade 10 computer science course: The Grade 10 training will take place across thecountry … So the teachers will be receiving oneweek’s training for a course which is going to lastfor three years. Many of the teachers, in fact, areunder-qualified and consequently have to gothrough a qualification phase, and the intent is totry and do that within one week. I don’t thinkthat that is a task that is going to be easily dealtwith. And so we’re hoping that many of thetertiary institutions ought to be offering what arecalled Advanced Certificates of Education. (pp.23–24)

The government was also establishing partnershipswith universities to address the issue of on-goingteacher training.

Verno also highlights the role of partnerships withcolleges and universities in the successful curriculumimplementation of a national computer sciencecurriculum, especially in the absence of governmentfunding or support. In particular, she notes thatCSTA funds and oversees two teacher professionaldevelopment programs: Java Engagement for TeacherTraining (JETT) and Teacher Engagement forComputer Science (TECS), which involve collegesand universities across the country providingworkshops and on-going mentoring for local highschool computer science educators.

As Kubow and Fossum (2003) indicated, theassessment of policy effectiveness in accordance withFrank’s criteria also requires consideration of thepolicy implementation timeframe. While the panelistsfrom Israel, Scotland, and South Africa indicate thattheir new computer science curricula were developedand implemented on a reasonable timeframe,according to Stephenson, the government’s rush todevelop and implement a new curriculum forOntario was considerably more problematic.

I think everyone described most accurately thewhole process as ready, fire, aim. “We’re going todo this new curriculum, we’re going to do it rightnow, oh boy we better plan for it.” And, so therewere issues. (p. 15)

3.2.2 Theoretical Adequacy

According to Frank’s critical framework, determiningthe theoretical adequacy of an education policymandate requires an examination of the linkagebetween the outcomes desired and the strategiesemployed in three areas: the reasoning andjustification associated with the issue, thecontextualization of the rationales with respect tocurrent and pressing needs, and the effect oftangential factors. An examination of the transcript ofthe international computer science panel sessionreveals, however, that in all cases, significantly moreattention is being paid to setting forward theexpectations than to determining if the strategies putinto place are actually achieving those outcomes.

In both Ontario and Israel, the panelists indicatethat computer science is seen as a full-fledgedscientific discipline and efforts to revise thecurriculum are situated in the desire to increase itsrigor. In Ontario’s case, Stephenson notes:

Many of the provinces went very much toward ageneral IT curriculum, cutting out the rigor of theirComputer Science program. But a few provinces,including Ontario have gone in a differentdirection. They chose to make their ComputerScience curriculum both more extensive and morerigorous. We stuck to the Computer Scienceframework, and we looked at expanding it. (p. 6)

According to Martin, changes to the Scottishcomputing curriculum were similarly situated in thedesire to make the program more rigorous. In thiscase, however, this was driven largely by Universityconcerns regarding student preparedness for post-secondary programs. In South Africa, however, newcurriculum policies required serious consideration ofissues of equity and ethnicity that are at thefoundation of all educational policy development andimplementation in that country.

There is a perception that Computer Studies isexclusively for the privileged white community,and so we do have a problem in terms ofinclusion of the other race groups. (p. 25)

60

This concern with balancing rigor and equity isfurther evidenced in both South Africa and Scotlandby additional policies that provide the opportunityfor students for whom the increased rigor isproblematic to take a slower, less rigorous approachto the same material. As Martin explains, thiscommitment to differentiated instruction is perceivedas central to Scotland’s commitment to providingequitable learning opportunities for all of its citizens.

In Scotland we’re very, very proud of our policyof inclusion and we firmly believe that educationshould be accessible to all. There is no upperleaving age to school in Scotland. We havestudents in our classes who are adults who havecome back to school, and we also have quite awide range of alternative assessment techniquesto enable pupils with special needs to participatein national exams. (pp. 22-23)

While commitment to a more equitable system can beseen as a primary rationale for educational policies inScotland and South Africa, in many countries thesepolicies are also rationalized in terms of the long-termeconomic needs of the country. This is most clearlyexemplified by Chiles of South Africa, who states:

In terms of the expected impact of the new courseson the economy, on employment opportunities,and on future studies, we have tried as far aspossible in terms of the development of the courseto take into account what is happening in businessand industry. (pp. 24–25)

and Verno of the United States who argues:

I think it’s time for us to make a broadcommitment to CS education at the K–12 level, soour students can compete in this globalenvironment that we are now part of. (p. 29)

The panelists’ considerations of the tangential effectsof new policies on education seem to be manifestedmost consistently in areas relating specifically toteacher preparations and qualifications. The panelistsindicate many countries place insufficient impetus on

the need to ensure that teachers’ skills are consistentwith the policies. Gal-Ezer, notes that in Israel, forexample, teachers who were teaching computerscience without a bachelor’s degree in computerscience prior to the curriculum revision were requiredto take 10 additional courses in order to qualify tocontinue teaching the subject in high schools.Stephenson also indicates that in Ontario, the neweducational policies created a series of challenges forwhich the government had not adequately planned.

When you change the curriculum you also get awhole lot of ancillary issues. They [thegovernment] never passed legislation to addressthe teacher certification issues, or addressed theissues such as mandatory courses and theshortage of teachers. As a result, the professionalassociation continues to shoulder theresponsibility of providing the majority ofopportunities for professional development toimplement the curriculum for the teachers. (p. 16)

Chiles of South Africa, also poignantly illustrates thateducation never takes place in isolation and that theworld outside can bring its own devastatingtangential effects despite the best-planned reforms.

One of the major problems that we have ineducation is HIV/AIDS. As you know, it’spandemic in Africa. In South Africa we have adouble-edged sword in that not only are many ofour teachers dying of HIV/AIDS, but many of theofficials or the workforce in business andcommerce are also dying, and where do they gettheir trained people from? Education. So we’relosing teachers to HIV/AIDS on the one sidedirectly, and we’re losing teachers to industry andcommerce because of HIV/AIDS on the otherside. And that is a major, major problem in ourcountry. (p. 26)

3.2.3 Empirical Validity

Frank (1972) posits that incorporating the differencesbetween past and present plans are central to

61

ensuring the empirical validity of a given educationalreform. Specifically, he identifies three areas thatmust be taken into account for this determination:

• whether the interpretations of circumstances areconsistent among stakeholders,

• what signs or changes related to an issue mightbe relevant, and

• whether they will be broadly recognized, andthe strengths and limitations of these means ofjudging.

The transcript of the international computer sciencepanel indicates, however, that, among all of Frank’scategories for critiquing educational policy andreform, empirical validity is the least represented inthe panelists’ reflections.

It is clear from Stephenson’s comments that thedecision to revise the computer science curriculum inOntario was made because the provincialgovernment had decided to revise the entire highschool curriculum, and that this decision wasconsistent with a cyclical pattern of political decision-making rather than of broader stakeholderperceptions that change was needed.

In 2006 all of the Computer Science courses willundergo review again. It starts all over. That hasto do with the fact that the government changesevery four years. So by about their second yearin, it’s like, “Let’s really change something. Let’sstart with education.” So the process continues.(p. 17)

While Martin (Scotland), Gal-Ezer (Israel), andChiles (South Africa) also note that the decision toreform the high school computer science curriculumin their countries was similarly government-driven,Chiles, however, indicates that the new curriculumpolicies in South Africa were also driven by abroader desire among multiple stakeholder groupsfor educational reform that would right the wrongsof apartheid, which ended in 1994: “South Africa hasin the last 10 years been through major curriculumreform. When we say major, everything thathappened before 1994 has literally been dropped and

is being revised” (p. 12). For all of the panelists, the determination of

whether or not the educational reform policies wouldachieve their intended aims and how this is to bemeasured seemed open to question. While bothStephenson (Canada) and Gal-Ezer (Israel) indicatethat the reforms were intended to improve thescientific rigor of the curriculum, Stephensoncontends that the Ontario government is focusedentirely on changing how the teachers assess thestudents rather than on indicators of student success.Gal-Ezer (Israel) and Martin (Scotland) however,indicate that student performance on standardizedtests is the marker by which their countries measurethe success of the reforms.

Although educational reform policy developmentis clearly a top-down process in the countriesrepresented by the panelists, the task of determiningthe long-term effect of these policies appears to fallmore often to external bodies than to the governmentitself. In some cases the relationship between thesebodies is a formal one, as Martin states:

We also have a national public body calledLearning and Teaching Scotland, which isseconded computing teachers and teachers fromother subjects and they support the EducationDepartment by reviewing, assessing, anddeveloping the curriculum. (p. 22)

while in others, as Stephenson indicates, theresponsibility is undertaken by an arms-length, andoften educator-driven, organization:

There was no [government] emphasis onassessing whether these courses were actuallyworking … again, the professional association, thesubject association, did that assessment todetermine where the problems were. (p. 17)

From these comments it is possible to conclude that,at least in countries represented by the panel, whereeducational policy reform is clearly top-down, andoften politically driven, surprisingly little attention ispaid to determining whether or not the broad goalsof the reform have actually been achieved.

62

3.3 CONCLUSION

Applying Frank’s framework for critiquingeducational policy and reform to an examination ofefforts to reform the computer science curriculum infive countries provides an interesting perspective oncomputer science curriculum reform in specific andon curriculum reform as a whole in that it highlightsareas of strength and weakness that may have long-term effects on the extent to which the reformsaccomplish what they had ostensibly been designedto achieve. As this examination demonstrates, evenwhere the reforms are top-down and politicallydriven, chances of successful implementation can beimproved when sufficient attention is paid to policyeffectiveness (especially relating to support bystakeholder groups and sufficient resource allocation),theoretical adequacy (the balance of justifications andtangential effects) and empirical validity (whether andhow the long-term effects of the policy reform aremeasured). And while it is clear that none of thecountries represented on the panel can claim an error-free policy development and reform process, each,through the honest and open examination of itssuccesses and failures, provides beneficial insights forother countries willing to learn the lessons provided.

3.4 REFERENCES

Bureau of Labor Statistics. (2005) U.S. Department of

Labor occupational outlook handbook. RetrievedJanuary 12, 2005, from http://www.bls.gov/oco/ocos110.htm

Department of Education. (2005a). Guidelines for the

development of learning programmes: Computer

Applications Technology. Pretoria Department ofEducation, South Africa.

Department of Education. (2005b). Guidelines for the

development of learning programmes: Information

Technology. Pretoria Department of Education,South Africa.

Frank, A. G. (1972). Sociology of development andunderdevelopment of society. In J. D. Cockcroft,A. G. Frank, & D. L. Johnson (Eds.). Dependence

and underdevelopment: Latin America’s political

economy (pp. 321–397). Garden City, NY: Anchor.

Gal-Ezer, J., & Harel, D. (1999). Curriculum andcourse syllabi for high-school computer scienceprogram, Computer Science Education, 9(2), 114–147.

Gal-Ezer, J., Beeri C., Harel D., & Yehudai, A. (1995).A high-school program in computer science.Computer 28(10), 73–80.

International Society for Technology in Education(ISTE). (2002) National educational technology

standards for teachers. Retrieved December 1, 2005,from http://www.iste.org/nets

Kubow, P. K., & Fossum, P. R. (2003). Comparative

education: Exploring issues in international context.

Upper Saddle River, NJ: Pearson Education.

Ministry of Education. (1999). Technological education: The

Ontario curriculum grades 9 and 10. Toronto, ON: Minis-try of Education, Government of Ontario, Canada.

Ministry of Education. (2000). Technological education:

The Ontario curriculum grades 11 and 12. Toronto,ON: Ministry of Education, Government ofOntario, Canada.



Strauss, A. (1987). Qualitative analysis for social

scientists. New York: Cambridge University Press.

The Scottish Parliament Information Centre. (1999).Education in Scotland. Retrieved June 5, 2005 fromhttp://www.scottish.parliament.uk/business/research/pdf_subj_maps/smda-10.pdf

63





Verno, A., Chiles, M., Gal-Ezer, J., Martin, J., &Stephenson, C. (2005). Finding a curriculum that

fits: The United States and Beyond. Panel presentedat the National Educational ComputingConference. Philadelphia, PA.

64

4.0 INTRODUCTION

As the previous three chapters have shown, designingand successfully implementing a high schoolcomputer science curriculum is an exceedinglycomplex task involving multiple stakeholders. In theUnited States, this process is made even more complexby an educational system in which decision-makingand funding responsibilities are dispersed andfragmented, even to the level of individual schools.This does not mean, however, that it is not possible forthe United States to launch a much-needed effort toimprove high school computer science education; onlythat doing so requires a level of understanding andcommitment that encompasses leaders at every level,from the federal leaders who put forth new legislationto the state and district education administrators whointerpret and ensure compliance with policy to theindividual teachers who make it a classroom reality.

This chapter reviews what the previous chaptershave shown, with the goal of formulating strategiesand recommendations that can bring about real,fundamental, and lasting improvements to high schoolcomputer science education. These improvements willhelp provide students with the scientific knowledgebase required to prepare them for success in the 21stcentury and will ensure that this country can producea future workforce capable of the innovations that willcontinue to drive the global economy.

4.1 HOW DO WE DESIGN A BETTERCURRICULUM?

As demonstrated in Chapter Two, the educationalresearch clearly shows that high school computer

science education suffers because the discipline itselfseems to defeat all efforts to pin it down to one,easily understood definition. This has resulted in aninconsistent self-image, an incorrect public image,and serious misconceptions among both educationaldecision-makers and the broader public about thescope and focus of the discipline. While in the best ofall possible worlds, attempting to establish a single,universally accepted definition might seem like themost practical solution, attempting to develop such adefinition would inevitably lead to more heatedarguments among academics, scientists, and industryleaders. In addition, even if an agreement could bereached, the continual evolution of the theory andtechnology upon which it is based would render itimmediately out of date.

4.1.1 Ten Core Principles

Fortunately, research shows us that it is possible tocome to understand computer science and to define acurriculum for its study at the high school level thatsupports the understanding that computer science isa science. Here is a list of ten principles for asuccessful computer science curriculum.

• A high school computer science curriculum must;• focus on the scientific principles that underlie

that science; • develop students’ familiarity with the key

principles of the discipline, includingabstraction, complexity, modularity, andreusability;

• focus on teaching problem-solvingmethodologies and critical thinking skills;

• help students develop a wide range of cognitive

65

CHAPTER FOUR

STRATEGIES FOR THE SUCCESSFUL DESIGN ANDIMPLEMENTATION OF A NATIONAL HIGH SCHOOL

COMPUTER SCIENCE CURRICULUM

capabilities and practical skills, independent ofspecific technologies;

• be comprehensive enough to provide studentswith a broad overview of the field and sense ofthe history of the discipline so they can betterappreciate how computers have been used toaddress real problems (It must address thebreadth of the discipline and its manifestationin multiple areas of scientific engagementincluding hardware design, networks, graphics,databases and information retrieval, computersecurity, software design, programminglanguages, logic, programming paradigms,artificial intelligence, applications ininformation technology and informationsystems, and social issues.);

• deal explicitly with the design and analysisprocess so that students develop anunderstanding of the structure of computersystems and the processes involved in theirdesign, development, and maintenance;

• enable students to learn and to scaffold newideas, concepts, and skills across a series ofcourses that provide age-appropriate learningoutcomes;

• be taught in such a way that the contents areengaging to all students regardless of genderand ethnicity;

• interweave both conceptual and experimentalissues so that students come to understand boththe theoretical underpinnings of the disciplineand how that theory influences practice; and

• not confuse computer science with computerliteracy (Teaching students how to use varioussoftware applications is not teaching computerscience.).

4.1.2 Key Concepts

Like all scientific disciplines, computer scienceconsists of a set of key concepts that are fundamentalto its understanding and practice, and shouldtherefore form the basis of any high school

curriculum. The current body of research on computerscience education, (as discussed in Chapter Two) hasidentified several concepts key to the teaching ofcomputer science, including the following:

• Algorithm design: The use of a precise, step-by-step process for solving a programming problem.

• Recursion: A programming function or methodthat repeatedly calls itself until a specifiedcondition is met. (Note while recursion wassingled out in the research, it is important tonote that it is just one of many concepts thatcould be considered under the broader categoryof problem solving.)

• Abstraction and data types: Abstraction is aprocess of picking out (abstracting) commonfeatures of objects and procedures. A datatype isa name or label for a set of values and someoperations that one can perform on that set ofvalues. Programming languages implicitly orexplicitly support one or more datatypes. Thesetypes may act as a statically or dynamicallychecked constraint, ensuring valid programs fora given language.

• Program correctness and efficiency: Determiningprogram correctness involves ensuring that theprogram runs in strict accordance with itsspecification. Program efficiency relates to theefficient use of resources and is generallycontained in two properties: speed (the time ittakes for an operation to be completed), andspace (the memory or non-volatile storage usedup by the construct).

• Program testing and debugging: Testing is aprocess used to help identify the correctness,completeness, and quality of developedcomputer software. Debugging is a methodicalprocess of finding and reducing the number ofbugs, or defects, in a computer program or apiece of electronic hardware thus making itbehave as expected.

The problem with these research-identified concepts,however, is that they provide an extremely limitedview of the discipline of computer science. A more

66

comprehensive delineation of core computer scienceconcepts, however, can be found in the NationalResearch Council report entitled Being Fluent with

Information Technology (National Research Council,1999). This report puts forth a vision of InformationFluency that extends beyond the use of today’stechnology in one’s own field (Information Literacy), toinclude the use of algorithmic thinking to solveproblems and the ability to independently learn tomaster new technologies as they evolve. Towards thisend, Being Fluent with Information Technology identifies10 key concepts:

• Computers: Key aspects of computing, includingthe program as a sequence of steps, the processof program interpretation, computing memory,and peripheral devices.

• Information systems: The general structuralfeatures of an information system includinghardware and software components, processes,and interfaces.

• Networks: Key attributes and aspects ofinformation networks including their physicalstructure, wide-area networks and logicalstructure.

• Digital representation of information: The use ofbinary code.

• Information organization: General concepts ofinformation organization including searchingand retrieving.

• Modeling and abstraction: General techniques forrepresenting real-world phenomena ascomputer models, for example arrays and listsof procedures.

• Algorithmic thinking and programming: Conceptsof algorithmic thinking including repetition(iteration or recursion), data organization(records, arrays, lists), generalization, andsoftware design processes.

• Universality: The understanding that anycomputational task can be performed by anycomputer.

• Limitations of information technology: Assessingwhat information technology can be appliedand when it should be applied.

• Societal impact of information and information

technology: Social concerns relating to the use ofinformation technology such as privacy,intellectual property, security, online etiquette,hacking, and free speech.

4.1.3 Example of a ComprehensiveCurriculum

It is often helpful to put these kinds of recommenda-tions into perspective by providing an example of an actual curriculum that incorporates them. Thissection examines the Model Curriculum for K–12

Computer Science Education (Tucker, Deek, Jones,McCowan, Stephenson, & Verno, 2004) previouslydescribed in Chapter Two, with the goal ofdemonstrating how the ten principles and conceptsoutlined above can be used as the foundation for asolid curriculum framework.

• Focus on the scientific principles that underlie

that science. The authors of the Model

Curriculum for K–12 Computer Science Education

clearly state the study of computer science isindeed a scientific endeavor and that the goal oftheir curriculum framework is to introduce theprinciples and methodology of computerscience to all students.

• Develop students’ familiarity with the key

principles of the discipline, including students

abstraction, complexity, modularity, and

reusability skills. The Model Curriculum for K–12

Computer Science Education consists of four levelsof learning, each of which attempts to addressthese key principles at an appropriate learninglevel for students. Both the Level II course,Computer Science in the Modern World and theLevel III course, Computer Science as Analysis and

Design, require students to develop anunderstanding of hierarchy and abstraction andto master the basic principles of software design,including modularity and reusability. Theflexible framework also provides for a Level IVTopics in Computer Science course that could

67

include, among other options, an AP course thatwould cover modularity and reusability at anincreased level of complexity or a project coursethat would focus on some aspect of objectoriented programming and design.

• Focus on teaching problem-solving methodologies

and critical thinking skills. All of the coursesoutlined in the Model Curriculum for K–12

Computer Science Education contain requirementsfor problem solving and critical thinking.Learning Level I (covering grades 1–8), forexample, requires students to understand therelationship between logic and problem solvingin the real world, while Level II requiresstudents to gain a conceptual understanding ofthe basic steps in algorithmic problem solving.Level III also requires students to understandthe limits of computing by recognizingproblems that are computationally unsolvable.

• Help students develop a wide range of cognitive

capabilities and practical skills, independent of

specific technologies. The design of the Model

Curriculum for K–12 Computer Science Education iscentered in the authors’ belief that students needto understand the interleaving of computerscience principles and skills. Students are thenexpected to apply these skills (especiallyalgorithmic thinking) in their problem-solvingactivities in all educational subjects. In addition,the learning outcomes specified in thecurriculum are intended to ensure that studentsare prepared to be knowledgeable users andcritics of computers as well as designers andbuilders of computer applications.

• Address the breadth of the discipline and its

manifestation in multiple areas of scientific

engagement (hardware design, networks, graphics,

databases and information retrieval, computer

security, software design, programming languages,

logic, programming paradigms, artificial intelligence,

applications in information technology and

information systems, and social issues). Learningoutcomes for all of these areas are included inall of the courses specified by the Model

Curriculum for K–12 Computer Science Education.

• Deal explicitly with the design and analysis.

Every course in the Model Curriculum for K–12

Computer Science Education contains specificlearning expectations relating to problem solving,critical thinking, software design, and analysis.

• Enable students to learn and to scaffold new ideas,

concepts, and skills across a series of courses that

provide age-appropriate learning outcomes. TheModel Curriculum for K–12 Computer Science

Education builds a solid framework of conceptualand empirical learning experiences for students.By incorporating all of the National EducationalTechnology Standards (NETS) developed by theInternational Society for Technology in Education(ISTE, 2002) into its Level I section, it supports andexpands the national standards for educationaltechnology learning in grades K–8. It thenprovides a set of three age and learning-levelappropriate courses that allow students toprogress from foundational knowledge to mastery.

• Be taught in such a way that the contents are

engaging to all students regardless of gender and

ethnicity. While the Model Curriculum for K–12

Computer Science Education does not prescribespecific pedagogical approaches, it doeshighlight the importance of increasing the levelof computer science knowledge for all students,and most especially for those who are membersof underrepresented groups.

• Interweave both conceptual and experimental

issues so that students come to understand both

the theoretical underpinnings of the discipline

and how that theory influences practice. Eachhigh school course described in the Model


includes a set of conceptual learning outcomesas well as practical laboratories designed toprovide students with hands-on learning.

• Not confuse computer science with computer literacy.

Teaching students how to use various softwareapplications is not teaching computer science. Theintroductory section of the Model Curriculum for

K–12 Computer Science Education explicitly

68

69

KEY FLUENCY CONCEPT COURSES LEARNING OUTCOME

Computers Level II Principles of computer organization and the major components(input, output, memory, storage, processing, software, operatingsystem, etc.)

Fundamentals of hardware design

Level III Hardware and systems: logic, gates and circuits, binary arithmetic,machine and assembly language, operating systems, user interfaces,compilers

Information systems Level II Principles of computer organization and the major components(input, output, memory, storage, processing, software, operatingsystem, etc.)

Level III Design for usability

Networks Level II The basic components of computer networks (servers, fileprotection, routing protocols for connection/communication, spoolers and queues, shared resources, and fault-tolerance)

Digital representation of Level II The notion of hierarchy and abstraction in computing, including information high-level languages, translation (compilers, interpreters, linking),

machine languages, instruction sets, and logic circuits

The connection between elements of mathematics and computerscience, including binary numbers, logic, sets, and functions

Level III Topics in discrete mathematics: logic, functions, sets, and theirrelation to computer science

Information organization Level II The notion of hierarchy and abstraction in computing, includinghigh-level languages, translation (compilers, interpreters, linking),machine languages, instruction sets, and logic circuits

The connection between elements of mathematics and computerscience, including binary numbers, logic, sets, and functions

Level III Levels of language, software, and translation: characteristics ofcompilers, operating systems, and networks

describes the distinctions between computerscience and information technology and betweeninformation literacy and information fluency.

An examination of the learning outcomes included inthe Model Curriculum for K–12 Computer Science

Education also indicates the extent to which thiscurriculum framework complies with the essentialconcepts of information fluency as set forth in theBeing Fluent with Information Technology report. The

following table illustrates coverage of these conceptsin the Level II Computer Science in the Modern World

course, and the Level III Computer Science asAnalysis and Design course. (The Level I Foundations

of Computing course is not included in this tablebecause this is a pre-high school course. The Level IVTopics in Computer Science course is excluded from this table because this course can take many forms,including an AP Computer Science course, or acourse geared toward professional certification.)

70

Modeling and abstraction Level II Managing complexity through top-down and object-oriented design

Procedures and parameters

Level III Methods (functions) and parameters

Recursion

Objects and classes (arrays, vectors, stacks, queues, and their usesin problem-solving)

Event-driven and interactive programming

Algorithmic thinking and Level II Examples (like programming a telephone answering system) that programming identify the broad interdisciplinary utility of computers and

algorithmic problem solving in the modern world

The basic steps in algorithmic problem-solving (problem statementand exploration, examination of sample instances, design, programcoding, testing and verification)

Sequences, conditionals, and loops (iteration)

Level III Fundamental ideas about the process of program design andproblem solving, including style, abstraction, and initial discussionsof correctness and efficiency as part of the software design process.

Principles of software engineering: software projects, teams, thesoftware life cycle

Universality Level II The notion of computers as models of intelligent behavior (as foundin robot motion, speech and language understanding, and computervision), and what distinguishes humans from machines

Limitations of information Level III The limits of computing: what is a computationally “hard” problem?technology (e.g., ocean modeling, air traffic control, gene mapping) and

what kinds of problems are computationally unsolvable (e.g., thehalting problem)

Societal impact of Level II Ethical issues that relate to computers and networks (including information and security, privacy, intellectual property, the benefits and drawbacks information technology of public domain software, and the reliability of information on the

Internet), and the positive and negative impact of technology onhuman culture

Identification of different careers in computing (e.g., informationtechnology specialist, Web page designer, systems analyst,programmer, CIO)

Level III Social issues: software as intellectual property, professional practice

Careers in computing: computer scientist, computer engineer,software engineer, information technologist.

While it may be one among many possible options, theModel Curriculum for K–12 Computer Science Education

serves as an excellent example of a comprehensive andyet flexible high school computer science curriculumframework. Designed to reflect both the breadth of thediscipline and the key knowledge needs of today’sstudents, it allows schools to focus on core knowledgewhile at the same time incorporating local needs andexisting program strengths.

4.2 HOW DO WE ENSURESUCCESSFULIMPLEMENTATION?

The extent to which a curriculum meets students’learning needs is a key element in decisions aboutwhat to teach. The curriculum itself, however, willhave no real effects on student learning outcomesunless it is broadly and successfully implemented atthe classroom level. As demonstrated in ChapterThree, however, the successful implementation of ahigh school computer science curriculum is acomplex process requiring effort and buy-in at everylevel of the educational system. This section exploresthe key factors for effective curriculumimplementation and provides practical strategies forsuccess in three key areas: policy effectiveness,theoretical effectiveness, and empirical validity.

4.2.1 Policy Effectiveness

The sad reality is that very few educational policyinitiatives effectively generate systemic changesimply because there is a fundamental disconnectbetween those responsible for creating the policiesand those responsible for making them happen.Often this gap results from a lack of understanding ofthe educational environment, students, and teachers.Sometimes it is the result of arrogance. It is always aprecursor to failure.

Here, however, are five essential elements thatwill help ensure that a new or revised curriculum is

translated into classroom reality in a way that trulybenefits students and the broader society.

1 Support: The new curriculum initiative musthave both top-down and bottom-up support.Whether it is driven by the federal or stategovernment (top-down), by school district, orby teachers and parents (bottom-up) everyonemust agree that change is necessary andeveryone must be committed to making ithappen. This also means that major changeagents must be in place at all of these levels toensure a continued level of enthusiasm andsupport.

2 Stakeholder Buy-in: External groups must have arole in the curriculum review process if they areto support the implementation process. Thismeans that groups such as teachers unions,professional associations, parent councils, post-secondary institutions, and business andindustry should be invited to take part in thereview process and to help create consensusthat the new curriculum is in concert withlarger educational, social, and economic goals.

3 Resources: Schools, teachers, and students mustbe provided with the resources required tosuccessfully implement the new curriculum.These resources include access to theappropriate hardware and software (it does notalways have to be the latest and the greatest butit has to work) and learning resources (e.g.textbooks, reference resources, andmanipulatives).

4 Professional Development: Teachers must receivethe training necessary to allow them to masternot only the content of the new curriculum buteffective teaching strategies for delivering thatcontent to students.

5 Timeframe: Often governments or individualstry to rush the implementation process, eitherbecause they do not understand how long ittakes to achieve real systemic change, or becausethe appearance of doing something is moreimportant than actually doing it right. Everystep toward the successful implementation of a

71

new curriculum takes time. Giving less timethan truly needed to accomplish any step alongthe path from vision to reality can condemn theentire process to failure.

4.2.2 Theoretical Effectiveness

Attempts to implement new curricula or simplyimprove existing ones also fail because the goals arenot clear or the strategies used to achieve them areinsufficient or inappropriate. As the comments of theinternational computer science curriculum experts inChapter Three demonstrate, the problem often lies inthe fact that the curriculum developers wereattempting to achieve multiple and conflicting goals.For example, attempts to increase the academic rigorof a course or subject area may actually discouragestudents (especially those who might shy away froma program such as computer science) from pursuingstudies in this area, thus undercutting the goal ofmaking the curriculum more equitable. Curriculumdesigners must also deal with the trade-off betweenflexibility and the realities of school scheduling. Forexample, while a curriculum designed to include abroad array of possible courses may appear to offerstudents increased opportunities to explore theirinterests and abilities, in reality, it could ensure thatno courses are offered simply because there are notenough students in any one course in a given schoolto allow that school to staff it.

Here, therefore, are three suggestions to helpensure your new or revised computer sciencecurriculum will be theoretically effective.

1 Be explicit: Begin the curriculum developmentprocess with a clear statement of the intendedoutcomes of the curriculum.

2 Examine all assumptions: Do not assume that agiven strategy will achieve one or morelearning outcomes. Also, do not assume thatbecause two goals are worth achieving that theywill not somehow be in conflict with each otherat the implementation level.

3 Do a school-level reality check: Ensure that those

who will ultimately be responsible forimplementing a new or revised curriculumhave an opportunity to provide feedback on themodel before the content (i.e., specific learningoutcomes and strategies) is written. Once thecontent has been written, ask for their feedbackagain and make this feedback a driving force ofthe pre-implementation revision.

4.2.3 Empirical Validity

One of the most surprising things the research reviewreveals, is how little effort is put into measuring theultimate, long-term success of a curriculum initiative.With the exception of the government of Israel, theresearch provides little indication that othergovernments have funded research to determine if anew computing curriculum for high schools hasactually achieved its goals. This should not besurprising since the original curriculum models formost of the curricula that have been developed todate were not grounded in existing educationalresearch. This lack of attention to research mayexplain why so few educational change initiativesachieve their stated goals and why so many teachersrespond to announcements on impending changeswith extreme skepticism.

Here, therefore, are three suggestions to helpensure your new or revised computer sciencecurriculum will be empirically valid.

• Do the research first: Ensure that the curriculummodel and content are solidly ground in thecurrent body of international educationalresearch. The research can be very helpful inpointing out mistakes there is no need to repeat.

• Plan for the long-term: Ensure that everyoneinvolved in the project at every level of theeducational system understands that systemicchanges will take time and will require a long-term commitment of attention and resources.

• Measure your success and correct when

necessary: Establish short-term and long-termmetrics for success and perform appropriate

72

quantitative and qualitative studies todetermine if they have been met. If early resultsshow less success than hoped for, do nothesitate to make appropriate adjustments.

4.2.4 Example of a ComprehensiveImplementation Plan

Recommendations regarding successful curriculumimplementation can sometimes be clearer whenexamined in relation to an actual project that attemptsto put them into practice. This section examines oneparticular project involving administrators, teachers,and students in the Los Angeles Unified SchoolDistrict (LAUSD).

The Computer Science Equity Collaborative(CSEC) was initiated by a National ScienceFoundation-funded research team and was formed toaddress disparities in representation in high schoolcomputer science along lines of race, social class, andgender. Jointly sponsored by UCLA’s GraduateSchool of Education, UCLA’s Henry Samueli Schoolof Engineering and Applied Sciences, and theLAUSD, CSEC began with a three-year, multi-site,longitudinal study of the computer science pipelinein LAUSD high schools. This study revealed seriousproblems in the district’s computer science educationprogram, including the lack of a teacher computerscience pipeline, a dearth of computing resources forteachers and students, and a general sense ofmisunderstanding of the computer science disciplineby students, teachers, counselors, and administrators.This research led to the creation of a partnershipfocused on addressing these issues throughprofessional development for teachers and increasedsupport services for students studying AdvancedPlacement Computer Science (AP CS).

• Support: The partnership between the UCLAGraduate School of Education researchers, thepost-secondary engineering faculty, and theLAUSD representatives ensured that teachersand students would be provided with manydifferent kinds of support from several sources.

This support included both pedagogical andtechnical support.

• Stakeholder Buy-in: The CSEC collaborativeallowed for multiple roles for stakeholders thatencompassed knowledge of computer science,insight on the instructional design leadershipand vision, and an essential awareness of thecomplexities of race and gender in computing.The Dean of the School of Engineering providedoverall support for diversifying the high schoolcomputer science pipeline and articulated theconnections between this high school programand his own goals for a more culturally diverseenrollment in computer science at UCLA.LAUSD’s Director of Science, Todd Ullah,provided passionate leadership for the project atthe district level with his vision and hiswillingness to create change and improvecomputer science education in the district. JohnKwan, of LAUSD’s Instructional TechnologyDivision provided much-needed planning,implementation, and coordination between thedistrict, teachers, and schools. Schooladministrators also provided assistance byidentifying potential AP CS teachers.

• Resources: The Graduate School of Educationprovided the foundational research andcontributed to all areas of curriculum and teachersupport. Under the direction of Jane Margolis andJoanna Goode, researchers from the GraduateSchool of Education provided the facilitationbetween LAUSD and the Engineering School.They provided the research-based evidence todrive the reform effort and worked with bothinstitutions to co-develop a model of change thatwould increase representation in computerscience classes for females and students of colorwhile also accounting for the complex ways inwhich schools operate. Margolis, who was thePrincipal Investigator of the three-year researchproject, carried out much of the coordinating andpublicity work, including serving as lead grantwriter, attending conferences, and makingindustry connections. Joanna Goode coordinated

73

and taught at the monthly AP Readiness programand served as lead instructor and planner in theprofessional development program. She alsodeveloped the curricular materials, coordinatedwith teachers via email, provided instructionalmaterials, and administered evaluations. After theinitial summer institute, follow-up researchrevealed that teachers found the existing districtcomputer science curriculum to be a poor matchfor the needs of students in learning computerscience and preparing for the AP exam. Armedwith qualitative data demonstrating theinadequacy of learning resources, Goodesurveyed available introductory computer sciencecurriculum and recommended an AP-specificcomputer science curriculum for use by LAUSDteachers. With the approval the District, LAUSDand UCLA shared the cost of this curriculum.

• The School of Engineering providedassistance with logistical issues, such asproviding lab space and classrooms for highschool teachers and students. A lecturer from theSchool of Engineering also served as a primaryinstructor for both the summer professionaldevelopment program and monthly APReadiness sessions for teachers and students.

• LAUSD provided district-level coordinationincluding: identifying and recruiting teachers forthe program, enrolling teachers, providingsubstitute teachers, and providing transportation.The school district also supported the addition ofAP computer science courses to schools’ masterschedules. District staff also took the lead for thelogistical and curricular planning of the summerprofessional development program.

• Professional Development: CSEC beganpreparing teachers to teach the AP CS course byproviding a summer program, but soondetermined that one or two summer weeks wasnot sufficient. As a result, they then began amonthly AP Readiness program to support theteachers’ ongoing knowledge and pedagogydevelopment throughout the year. The CSECteam also worked with the teachers, rather than

just trying to “train” them. They maintainedflexibility in program scheduling toaccommodate emerging issues, concerns, andmisunderstandings. They also made a point ofacknowledging many of the teachers in theroom as experts—respecting them while alsoproviding a safe space for new AP CS teachersto wrestle with computer science concepts.Finally, they helped build leadership within theteaching community by allowing teachers totake on new roles in helping other teacherswhile improving their own teaching practices.

• Timeframe: The project’s three-year researchprogram exposed many of the underlying issuesof access, equity, and the rigor of computerscience in LA schools and provided a plan forhow to address them. Pending new funding, theresearchers predict that the project will becompleted in 2009.

Several additional elements of the program helped toensure that it met the criteria for empirical validity.

• Do the research first: All of the interventionsproposed in this program followed from theextensive three-year study preceding projectimplementation. The project also involvedspecialists from several disciplines to ensurethat several different kinds of expertise wouldbe brought to help interpret the research and tosolve the problems that the data uncovered.

• Plan for the long-term: All of the partners in theproject were committed to maintaining theimplementation plan. Project organizers admit,however, that they have concerns about thesustainability of this effort and the pressure ofoutside forces that have the potential tomarginalize the importance of computerscience. They note that the national educationaldialogue (and policy) now is not about access,rigor, deep thinking, scientific reasoning, orinquiry. Rather, it is about failing schools, anarrow focus on literacy and numeracy (asdetermined by testable topics) accountability,and so on. In this climate. it is difficult to

74

maintain computer science as a vital part of theK–12 educational system.

• Measure your success and correct when

necessary: The CSEC team is using several datapoints to evaluate the success of the project.They are collecting and analyzing data onannual computer science enrollment numbers(e.g., how has enrollment changed) by overallcourses, overall enrollment, and also by genderand race. They also collected data at eachmeeting with teachers and students to examinethe change in their understandings and practicesover time. They continue to conduct research todetermine teacher professional development andinvolvement, and if the programs aresustainable after the research grants expire.

The CSEC project was designed to ensure sustainabilityby involving multiple levels of stakeholder buy-in andsupport. Unlike many efforts to improve K–12computing education, this project provides a relevantexample of a change initiative that is solidly groundedin research (a three-year study) and clear in its focusand objectives (to improve equity of access bytraditionally underrepresented students of color andfemales to computer science education by increasingthe opportunities for students to take the AP CScourse). The CSEC project leaders were also careful toacknowledge and include teachers, to recognize andsupport their craft knowledge, to encourage anddevelop their leadership, and to support their ongoingdevelopment as exemplary educators.

4.3 HOW DO WE IMPROVE HOWTEACHERS TEACH?

One of the challenges decision-makers face whenattempting to improve existing curricula or introducenew ones is determining the extent to which they canand should also attempt to shape how teachers teach.Pedagogy, however, is a complex phenomenon, bothscience and art, and as such it has proven almostimpossible to quantify or replicate.

4.3.1 Five Qualities of ExemplaryTeachers

Although we do not currently have a guaranteedprocess for ensuring that all teachers are greatteachers, research into computer science educationhas shown that truly effective teachers demonstrateseveral key qualities that help them engage and teachstudents more effectively. Here are five key qualitiesof exemplary high school computer science teachers.

• Problem-solving approach: Exemplary computerscience teachers use a problem-solvingapproach that allows students to examineproblems from different angles and perspectivesand formulate solutions.

• Real-world focus: Exemplary computer scienceteachers motivate students by having themcreate real-world artifacts with an intendedaudience and encouraging them to understandthe essential link between the problem, the user,and the solution

• Explicit emphasis on design: Exemplarycomputer science teachers explicitly teach anduse the software design process, ensuring thatstudents master the steps involved in designing,creating, testing, and debugging software.

• A welcoming environment: Exemplary computerscience teachers make their classroom a welcomingenvironment for all students (especially youngwomen and minority students) and find creativeways to engage all students with examples andexercises that are relevant to their lives.

• Modeling life-long learning: Exemplary computerscience teachers serve as role models for theirstudents by continuing to enhance their ownteaching and technology skills and by exploringnew ideas and new technologies.

4.3.2 Seven Systemic ChangesRequired to Improve Teaching

For teachers, the challenge of becoming andremaining exemplary educators is often exacerbated

75

by a system of pre-service education and teachercertification that makes it so difficult to become acomputer science teacher that it actually discourageshighly qualified individuals from applying in the firstplace. In many states, there is no pre-service programthat specifically prepares high school computerscience teachers. In many school districts, there are nocore knowledge guidelines for computer scienceteachers. In many schools, teachers are expected totake on teaching responsibilities in subject areas forwhich they have little or no preparation (ComputerScience Teachers Association, 2005).

The responsibility for ensuring that all highschool computer science teachers are knowledgeable,well prepared, and continuously engaged must beshared at all levels of the education system. Itrequires a profound commitment, a coordinated setof policies, and the allocation of appropriate andadequate resources. Here is a list of seven policychanges that would substantially improve computerscience education in the United States.

• Mastery of core knowledge: High school computerscience teachers should be required to havecompleted an undergraduate degree in computerscience or a comparable degree program.

• Standardized pre-service programs: All teacherpreparation programs should be required toadhere to the National Council for Accreditationof Teacher Education (NCATE) standards forhigh school computer science educators.

• Certification standards: State teacher certificationrequirements for high school computer scienceteachers should adhere to a consistent (andenforced) national standard that would allow forgreater clarity and mobility from state to state.

• Professional development: School districts shouldprovide regular professional development forcomputer science teachers to allow them tokeep their knowledge and skills current.

• Focus on teaching: School districts should employa sufficient number of technical specialists withresponsibility to ensure that computer hardware,networks, and software is maintained, freeingteachers to concentrate on their teaching.

• Competitive compensation: Salaries for computerscience teachers should be commensurate withthose offered in industry to ensure that the bestpossible candidates prepare and apply forteaching positions.

• Professional affiliation: All high school,computer science teachers should be membersof professional associations that support theirdiscipline-based knowledge and provide ateaching community that mentors andcelebrates them.

4.4 CALL TO ACTION

Computer science has increasingly become a coreknowledge requirement for all educated citizens. Likethe more traditional sciences, it provides an essentialunderstanding of the world around us. Even if we arenot aware of it, computers are part of almost everyaspect of our lives (e.g., banking, health care,shopping, and travel) and it is vital that weunderstand their capabilities and their limitations. Allstudents need a basic level of knowledge about howto use computers safely and securely. Computerscience also teaches students how to be innovativeand solve problems and, in this way, helps them bebetter prepared for college and careers.

There is also a broader national interest at stake.U.S. government labor forecasts clearly indicate thatwe are not producing enough computer sciencegraduates to meet the needs of industry or to competein an increasingly global economy (Bureau of LaborStatistics, 2005). In addition, fewer college studentsare enrolling in computer science courses, and fewergraduates with computer science degrees are going onto earn their PhDs (Taulbee, 2003). If we do not beginto address these pipeline issues immediately and atthe earliest possible educational level, our ability tocompete in the global economy and to participate inthe great scientific and industrial developments of thisnew century will continue to diminish.

Addressing these issues is not just a school issue,it is an issue that can only be addressed with vision,

76

action, and commitment at all levels of the politicaland educational systems. The following sectionsprovide suggestions for ways in which we can worktogether to achieve long-term, systemicimprovements to computer science education. Theseinclude federal government policy makers, stategovernment policy makers, school district policymakers, school principals, teachers, university andcollege faculty, and business and industry leaders.

4.4.1 Federal Government PolicyMakers

Here are 10 ways federal-level policy makers andleaders can improve high school computer scienceeducation:

• Talk about the importance of computer sciencewhen you are speaking at events and with otherfederal leaders.

• Include computer science in the 21st centuryskills conversation. When people talk about theimportance of Science, Technology, Engineeringand Math (commonly referred to as “STEM”skills), emphasize and clarify the “T.” Makingthe most of technology education is not justwiring schools or teaching students to usecomputers. That is only half the battle.

• Insist that funded STEM initiatives includerequirements to address issues in K–12computer science education.

• Support new initiatives that provide funding forresearch that will help improve computerscience education.

• Support new initiatives that provide funding forcomputer science teacher training.

• Put forward or support an initiative thatencourages states to rationalize their computerscience teacher certification requirements. Thiswill encourage the development of a larger,stronger, and better-qualified educationalworkforce.

• Put forward or support an initiative thatencourages states to ensure that computer

science teachers have computer sciencequalifications.

• Make sure conversations about thecompetitiveness of the United States encouragestudents to go into computer science careersand explain why it is so important for ournation to be a leader in this field.

• Talk to people in your community aboutoutsourcing technology jobs. Explain how thistrend can be combated by focusing on theeducation of future workers in our own country.

• Help improve the public’s understanding of themany exciting, varied, and cutting edge jobsavailable in the field of computer science.

4.4.2 State Government Policy Makers

Here are 10 ways state-level policy makers and leaderscan improve high school computer science education:

• Ask what local school authorities are doing toensure students are learning what they need toget the jobs of the future and require all highschool students to take at least one computerscience course.

• Identify how much is being spent on computerscience education in your state and determine ifthere have been cutbacks on computer scienceeducation.

• Promote funding for computer science courseequipment and materials.

• Ensure that schools of education and on-the-jobprofessional development opportunities areadequately training computer science teachersin your area.

• Ensure that certifications requirements forcomputer science teachers guarantee thatteachers have the appropriate knowledge baseto teach in this discipline.

• Provide programs for individuals transitioningfrom the information technology industry toeducation that offer an opportunity to acquirethe educational knowledge and training neededto become an exemplary teacher.

77

• Talk to your local leaders and communitymembers about what they see as futurecomputer science job needs.

• Encourage businesses to provide support andmentoring opportunities for schools. Explainthe importance of business’ role in theeducation of future workers.

• Talk to people in your community aboutoutsourcing technology jobs. Explain how thistrend can be combated by focusing on theeducation of future workers in your state.

• Point community members to the resources ofnonprofit groups, such as CSTA, that providecurriculum models and other resources tosupport computer science teachers.

4.4.3 School District Policy Makers

Here are 10 ways school district policy makers,curriculum policy leaders, and instructionaltechnology specialists can improve high schoolcomputer science education:

• Identify how much is being spent on computerscience education in your school district anddetermine if there have been cutbacks oncomputer science education.

• Promote funding for computer science courses,equipment, and materials.

• Put policies in place to ensure that onlyqualified teachers are teaching computer sciencecourses and enforce those policies.

• Review your current computer sciencecurriculum to determine whether it providessufficient opportunities for students to gain theknowledge and skills they need to succeed in anincreasingly computerized world.

• Review your current computer science curriculumto determine whether it is both academicallyrigorous and welcoming to all students.

• Create and support professional developmentopportunities for computer science teachers toensure that their technical and pedagogicalskills keep pace with change.

• Help teachers keep their focus on teaching byproviding technical support staff who areresponsible for maintaining and upgrading thehardware and software.

• Provide opportunities for computer scienceteachers across the district to meet with andmentor each other.

• Work with school counselors to ensure that theyare providing students with accurate andappropriate information about careeropportunities in computing and the educationalpathways students must take to achieve theirlong-term goals.

• Point community members to the resources ofnonprofit groups, such as CSTA, that providecurriculum models and other resources tosupport computer science teachers.

4.4.4 School Principals

Here are 10 ways school principals can improve highschool computer science education:

• Identify how much is being spent on computerscience education in your school and determineif there have been cutbacks on computer scienceeducation.

• Promote funding for computer science courses,equipment, and materials.

• Ensure that only qualified teachers are teachingcomputer science courses.


• Work with your computer science teachers toensure that their courses are both academicallyrigorous and welcoming to all students.

• Create and support professional developmentopportunities that help computer scienceteachers ensure that their technical andpedagogical skills keep pace with studentlearning needs.

78

• Help teachers keep their focus on teaching byproviding technical support staff who areresponsible for maintaining and upgrading boththe hardware and software.

• Provide opportunities for computer scienceteachers to meet with colleagues from across thedistrict, the state, and the nation.

• Work with school counselors to ensure they areproviding students with accurate andappropriate information about careeropportunities in computing and the educationalpathways students must take to achieve theirlong-term goals.

• Point teachers to the resources of nonprofitgroups, such as CSTA, that provide curriculummodels and other resources to supportcomputer science teachers.

4.4.5 Teachers

Here are 10 ways teachers can improve high schoolcomputer science education:


• Ensure that your courses are both academicallyrigorous and welcoming to all students.

• Model life-long learning to your students byseeking out and taking advantage of relevantprofessional development opportunities.

• Lobby for funding for computer sciencecourses, equipment, and materials.

• Lobby for your program by speaking to parentgroups, other teachers, and individual studentsabout computer science.

• Help dispel the myth that there are no jobopportunities in computer science. There aremany exciting and cutting-edge jobs available.

• Work with school counselors to ensure that theyare providing your students with accurate andappropriate information about career

opportunities in computing and the educationalpathways students must take to achieve theirlong-term goals.

• Create opportunities to meet with computerscience teachers from across the district so thatyou can share resources and strategies.

• Become a member of a professional association,such as CSTA, that provides curriculum models,support materials, mentoring, and communityand that helps you continue to develop yourteaching and leadership skills.

4.4.6 University and College Faculty

Here are 10 ways university and college faculty incomputing and engineering programs and in schoolsof education can improve high school computerscience education:

• Support high school computer scienceeducation by requiring students entering youruniversity to have taken at least one high schoolcomputer science course.

• Create different entry points into your programsfor students with differing levels of high schoolcomputing experience.

• Provide local high school teachers and studentswith opportunities to interact with your faculty,graduate students, and undergraduate studentsby inviting them to visit your institution oroffering to send faculty and students to speak atlocal schools.

• Provide teachers and students with detailedinformation about your computer science andengineering programs and the skills incomingstudents require to ensure their success at yourinstitution.

• Create opportunities for your female andminority students to mentor high schoolstudents whose life experiences are similar totheir own.

• Create and support professional developmentopportunities that help computer scienceteachers ensure that their technical and

79

pedagogical skills keep pace with change.• Provide teachers, students, and school

counselors with information about careers incomputer science and the higher educationpathways required for students to meet theirlong-term goals.

• Ensure that schools of education are adequatelypreparing pre-service teachers to teachcomputer science.

• Help teachers purchase new technologyresources by working with them to developgrant applications.

• Ensure that your faculty are aware of the Model


and that they play an active role in professionalorganizations such as CSTA that supportcomputer science teachers.

4.4.7 Business and Industry Leaders

Here are 10 ways business and industry leaders canimprove high school computer science education:

• Make sure conversations about thecompetitiveness of the United States encouragestudents to go into computer science careersand explain why it is so important for ournation to be a leader in this field.

• Help improve the public’s understanding of themany exciting, varied, and cutting edge jobsavailable in the field of computer science.

• Include computer science in the 21st centuryskills conversation. When people talk about the importance of Science, Technology,Engineering and Math (commonly referred to as “STEM” skills), emphasize and clarify the“T.” Making the most of technology educationis not just wiring schools or teaching students to use computer applications. That is only halfthe battle.

• Talk to people in your community aboutoutsourcing technology jobs. Explain how thistrend can be combated by focusing on theeducation of future workers in our own country.

• Build partnerships with schools that promoteand demonstrate excellence in computer scienceeducation.

• Support the national implementation ofcurriculum standards such as the ACM Model

Curriculum for K–12 Computer Science and the

National Educational Technology Standards.

• Require that all computer education projectsreceiving funding from your organizationdemonstrate consistency with these nationalstandards.

• Fund new research initiatives that will helpimprove computer science education.

• Fund new teacher professional developmentinitiatives for computer science teachers.

• Support an initiative that encourages states toensure that computer science teachers havecomputer science qualifications.

4.5 CONCLUSION

Our goal for this chapter was to provide practicalsolutions for a complex problem, and while wecannot claim that any of our lists are exhaustive, theydo provide a powerful set of suggestions that, if putinto practice, will significantly improve computerscience education in high schools. The reality is thatwe are at somewhat of a crossroads. We know thatthe world continues to change and that many of thechanges we face are related to the burgeoningpossibilities and consequences of our growingdependence upon and interrelation with computingtechnology. Maintaining our ability to meet thechallenges of the present and future require us tothink very carefully about the kind of knowledge ourstudents need to grow and to succeed. Supportingand improving high school computer scienceeducation and ensuring that the opportunities itprovides are open to all students requires a multi-level commitment. It is a commitment we must makeif our schools are to continue to provide relevanteducation and our society is to continue to solveproblems on the cutting edge of innovation.

80

81

4.6 REFERENCES


Labor Occupational Outlook Handbook. RetrievedJanuary 12, 2005, from http://www.bls.gov/oco/ocos110.htm

Computer Science Teachers Association. (2005).Results of the national secondary computerscience survey. Retrieved October 16, 2005, fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html


standards for students. Retrieved December 1, 2005,from http://www.iste.org/nets



Taulbee, O. E. (2003). 2002-2003 Taulbee study:



Tucker, A., Deek, F., Jones, J., McCowan, D., Stephenson,C. and Verno, A. (2003). A Model curriculum for K–12

computer science: Report of the ACM K–12 Education Task

Force Computer Science Curriculum Committee. NewYork, NY: Association for Computing Machinery

ACM K–12 Task Force. (2005). ACM K–12 Task Force

Curriculum Survey Results. Retrieved December 22,2005, from http://csta.acm.org/Research/sub/TuckerSurveyResults.html

Adajian, L. B. (1996). Professional communities:Teachers supporting teachers. Mathematics Teacher,

89(4), 321–364.

Almstrum, V. L., Hazzan, O., & Ginat D. (Eds.).(2004). Special issue on import/exportrelationships to computer science educationresearch. Computer Science Education, 14(4).

Armoni, M., & Gal-Ezer, J. (2003). Non-determinismin CS high school curricula. Proceedings of the FIE

2003 Conference, Boulder, CO, F2C–18 – F2C–23.

Ballantyne, R., McLean, S. V., & Macpherson, I.(2003). Knowledge and skills required for creating a

culture of innovation: Supporting innovative teaching

and learning practices. Brisbane: Faculty ofEducation, Queensland University of Technology.

Bauch, P. A., & Goldring, E. B. (2000). Teacher workcontext and parent involvement in urban highschools of choice. Educational Research and

Evaluation, 6(1), 1–23.

Bell, D. (1983). Promoting business education throughteacher participation in professional associations.Business Education Forum/Yearbook, 37(8), 48–51.

Ben-Ari, M. (2001). Constructivism in computerscience education. Journal of Computers in

Mathematics and Science Teaching, 20(1), 45–73.

Ben-Ari, M. (2002). From theory to experiment topractice in CS education. Paper presented at Kolin

Kolistelut—Koli Calling: 2nd Annual Finnish/Baltic

Sea Conference on Computer Science Education, Koli,Finland, October 2002.

Ben-Ari, M. (2004). Situated learning in computerscience education, Computer Science Education,

14(2), 85–100.

Ben-David Kolikant, Y. (2004). Learning concurrency:Evolution of students’ understanding ofsynchronization. The International Journal of Human

Computers Studies, 60(2), 243–268.

Ben-David Kolikant, Y., & Pollack, S. (2004a).Establishing computer science professional normsamong high school students. Computer Science

Education, 14(1), 21–35.

Ben-David Kolikant, Y., & Pollack, S. (2004b).Community-oriented pedagogy for inservice CSteacher training. Proceedings of the ITiCSE 2004

Conference, Leeds, United Kingdom, 191–195.

Ben-David Kolikant, Y. (2005). Students’ alternativestandards for correctness. Proceedings of the First

International Computing Education Research

Workshop, October 1–2, 2005, University ofWashington, Seattle, WA, 37–43.

Bergin, S., & Reilly, R. (2005). Programming: Factorsthat influence success. Proceedings of the SIGCSE

2005 Conference, St. Louis, Missouri, USA, 411–415.

Böszörmenyi, L. (2005). Teaching: People topeople—about people: A plea for the historic andhuman view. In R. T. Mittermeir (Ed.) From


Proceedings of International Conference on

Informatics in Secondary Schools—Evolution and

Perspectives, ISSEP 2005, Klagenfurt, Austria,March/April 2005, 93–103.

82

BIBLIOGRAPHY


Labor occupational outlook handbook. RetrievedJanuary 12, 2005, from http://www.bls.gov/oco/ocos110.htm

Campbell, P. F., & McCabe, G. P. (1984). Predicting the success of freshmen in a computer sciencemajor. Communications of the ACM, 27(11),1108–1113.

Cantwell Wilson, B., & Shrock, S. (2001). Contributingto success in an introductory computer sciencecourse: A study of twelve factors. SIGCSE Bulletin,

33(1), 184–188.

Cantwell Wilson, B. (2002). A study of factorspromoting success in computer science includinggender differences. Computer Science Education,

12(1–2), 141–164.

Cartelli, A. C. (2002). Computer science education inItaly: A survey. SIGCSE Bulletin, 34(4), 36–39.

Collofello, J. S. (2002). Creation, development andevaluation of an innovative secondary schoolsoftware development curriculum module.Proceedings of the 32nd ASEE/IEEE Frontiers in

Education Conference, Boston, MA.

Computer Science Teachers Association. (2005).Results of the national secondary computer science

survey. Retrieved October 16, 2005, fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html

Computer Science Teachers Association (2005a).Strategic Plan. Retrieved December 22, 2005, fromhttp://csta.acm.org/About/sub/CSTAStrategicPlan.html/

Crombie, G., Abarbanel, T., & Anderson, C. (2000).All-female computer science: The positive effects of single gender classes. The Science

Teacher, March 2000, 40–43.

Dagienë, V. (2005). Teaching information technologyin general education: Challenges and perspectives.In R. T. Mittermeir (Ed.) From computer literacy to




Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C.,Tucker, A., Turner, A. J., & Young, P. R. (1988).Draft report of the ACM Task Force on the Core of

Computer Science. New York, NY: Association forComputing Machinery.

Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C.,Tucker, A., Turner, A. J., & Young, P. R. (1989).Computing as a discipline. Communications of the

ACM, 32(1), 9–23.

Denning, P. J. (2004). Great principles in computingcurricula. Proceedings of the SIGCSE 2004

Conference, Norfolk, Virginia, USA, 336–341.

Department of Education. (2005a). Guidelines for the

development of learning programmes: Computer

Applications Technology. Pretoria Department ofEducation, South Africa.

Department of Education. (2005b). Guidelines for the

development of learning programmes: Information

Technology. Pretoria Department of Education,South Africa.

Dorninger, C. (2005). Educational standards in schoolinformatics in Austria. In R. T. Mittermeir (Ed.) From

computer literacy to informatics fundamentals, Proceedings

of International Conference on Informatics in Secondary

Schools—Evolution and Perspectives, ISSEP 2005,Klagenfurt, Austria, March/April 2005, 66–69.

du Boulay, B., O’Shea, T., & Monk, J. (1989). The blackbox inside the glass box: Presenting computingconcepts to novices. In E. Soloway & J. C. Spohrer(Eds.). Studying the novice programmer (pp. 431–446).

83

Hillsdale, NJ: Lawrence Erlbaum Associates.

Driscoll, M. (1987). Ten case studies from a study of

exemplary mathematics programs. Reston, VA:National Council of Teachers of Mathematics.

Eurydice (Information Network on Education inEurope). (2005). Information and communication

technologies in education. Retrieved January 25, 2005from http://www.eurydice.org/Documents/Bibliographie/en/frameset_biblio_refs_en.html

Fadi, P. D., & McHugh, J. A. (2000). Problem-solvingmethodologies and the development of criticalthinking skills. Journal of Computer Science

Education, 14(1&2), 6–12.

Fadi, P. D., & McHugh, J. A. (2001). Prototype softwaredevelopment tools for beginning programming.Journal of Computer Science Education, 14(3&4), 14–20.

Felleisen, M., Findler, R. B., Flatt, M., &Krishnamurthi, S. (2002). The structure andinterpretation of the computer science curriculum.Journal of Functional Programming, 14(4), 365–378.

Fincher, S. (1999). What are we doing when we teachprogramming? Proceedings of the FIE1999

Conference, San Juan, Puerto Rico, 12a4–1–12a4–4.

Fincher, S., & Petre, M. (2004). Computer science

education research. London, UK: Routledge Falmer.

Fisher, A., & Margolis, J. (2002). Unlocking theclubhouse: The Carnegie Mellon experience.SIGCSE Bulletin, 34(2), 79–83.

Foley, J. (2004). Old challenges, new opportunities.Computing Research News, 16(4). RetrievedDecember 1, 2005 from http://www.cra.org/CRN/articles/sept04/foley.html

Frank, A. G. (1972). Sociology of development andunderdevelopment of society. In J. D. Cockcroft,

A. G. Frank, & D. L. Johnson (Eds.). Dependence

and underdevelopment: Latin America’s political

economy (pp. 321–397). Garden City, NY: Anchor.

Franklin, R. (1987). What academic impact are highschool computing courses having on the entry-level computer science curriculum? Proceedings of

the SIGCSE 1987 Conference, St. Louis, Missouri,USA, 253–256.

Fullan, M. (2002). The change leader. Educational


Gal-Ezer, J., Beeri, C., Harel, D., & Yehudai, A. (1995).A high school program in computer science.Computer, 28(10), 73–80.

Gal-Ezer, J., & Harel, D. (1999). Curriculum andcourse syllabi for high-school computer scienceprogram, Computer Science Education, 9(2), 114–147.

Gal-Ezer, J., & Harel, D. (1998). What (else) should CSeducators know? Communications of the ACM,

41(9), 77–84.

Gal-Ezer, J., & Zeldes, A. (2000). Teaching softwaredesigning skills. Computer Science Education, 10(1),25–38.

Gal-Ezer, J., & Zur, E. (2002). The concept of“algorithm efficiency” in the high school CScurriculum. Proceedings of the 32nd ASEE/IEEE

Frontiers in Education Conference, Boston, MA,T2C1–T2C6.

Gal-Ezer, J., Vilner, T., & Zur, E. (2003). Characteristics

of students who failed (or succeeded) the introductory

CS course. Paper presented at the FIEE 2003

Conference. Boulder, CO. Retrieved December 4,2005 from http://fie.engrng.pitt.edu/fie2003

Ginat, D. (1996). Efficiency of algorithms forprogramming beginners. Proceedings of the SIGCSE

1996 Conference. Philadelphia, PA, 256–260.

84

Ginat, D. (2000). Colorful examples for elaboratingexploration of regularities in high school CS1.Proceedings of ITiCSE 2000, Helsinki, Finland, 81–84.

Goldweber, M., Fincher, S., Clark, M., & Pears. A.(2004). The relationships between CS educationresearch and the SIGCSE community. SIGCSE

Bulletin, 36(1), 147–148.

Graham, S., & Latulipe, C. (2003). CS girls rock:Sparking interest in computer science anddebunking the stereotypes, Proceedings of the

SIGCSE 2003 Conference, Reno, Nevada, USA,322–326.

Grandell, L. (2005). High school students learninguniversity level computer science on the web—acase study of the DASK-model. Journal of

Information Technology Education, 4, 207–218.

Greening, T. (1998). Computer Science: Through theeyes of potential students, Proceedings of the ACSE

1998 Conference, Brisbane, Australia, 145–154.

Gries, D. (2002). Where is programming methodologythese days? SIGCSE Bulletin, 34(4), 5–7.

Gurbiel, E., Hardt-Olejniczak, G., & Kolczyk, E. (2005).Informatics and ICT in the Polish educationsystem. In R. T. Mittermeir (Ed.) From computer





Gupta, U. G., & Houtz, L. E. (2000). High schoolstudents’ perceptions of information technologyskills and careers. Journal of Industrial Technology,

16(4), 2–8.

Guzdial, M. and Soloway, E. (2003). Computerscience is more important than calculus: Thechallenge of living up to our potential. SIGCSE

Bulletin, 35(2), 5–8.

Haberman, B., & Ginat, D. (1999). Distance learningmodel with local workshop sessions applied to in-service teacher training, Proceedings of the ITiCSE

1999 Conference, Krakow, Poland, 64–67.

Haberman, B., & Ben-David Kolikant, Y. (2001).Activating ’black boxes’ instead of opening’zippers’—A method of teaching novices basic CSconcepts. Proceedings of the ITiCSE 2001 Conference,

Dublin, Ireland, 41–44.

Haberman, B., & Averbuch, H. (2002). The case of basecases: Why are they so difficult to recognize?Student difficulties with recursion. Proceedings of the

ITiCSE 2002 Conference, Aarhus, Denmark, 84–88.

Haberman, B., Shapiro, E., & Scherz, Z. (2002). Areblack boxes transparent?—High school students’strategies of using abstract data types. Journal of

Educational Computing Research, 27(4), 411–436.

Haberman, B., Lev, L., & Langley, D. (2003). Actionresearch as a tool for promoting teacherawareness of students’ conceptual understanding.Proceedings of the ITiCSE 2003 Conference,

Thessaloniki, Greece, 144–148.

Haberman, B. (2004a). High school students’ attitudesregarding procedural abstraction: Special issuedevoted to recent research projects of secondaryinformatics education. Education and Information

Technologies, 9(2), 131–145.

Haberman, B. (2004b). How learning logicprogramming affects recursion comprehension.Computer Science Education, 14(1), 37–53.

Haberman, B., Averbuch, H., &. Ginat, D. (2005). Is itreally an algorithm?—The need for explicitdiscourse. Proceedings of the ITiCSE 2005 Conference,

Lisbon, Portugal, 74–78.

Hagan, D., & Markham, S. (2000). Does it help tohave some programming experience before

85

beginning a computing degree program? SIGCSE

Bulletin, 32(3), 25–28.

Harris, J. W. (1987). Teacher and leader—why we mustbe both. Vocational Education Journal, 62(5), 26–28.

Hazzan O., Impagliazzo, J., Lister, R., & Schocken, S.(2005). Using history of computing to addressproblems and opportunities, Proceedings of the

SIGCSE 2005 Conference, St. Louis, MO, 126–127.

Hazzan, O., & Lapidot, T. (2004a). Construction of aprofessional perception in the “Methods ofTeaching Computer Science” course. SIGCSE

Bulletin, 36(2), 57–61.

Hazzan, O., & Lapidot, T. (2004b). The practicum incomputer science education: Bridging the gapbetween theoretical knowledge and actualperformance. SIGCSE Bulletin, 36(4), 47–51.

Holmboe, C., McIver, L., & George, C. (2001). Researchagenda for computer science education. Proceedings

of the 13th Workshop of the Psychology of Programming

Interest Group, Bournemouth, UK, 207–233.

Impagliazzo, J., & Lee, J. A. N. (2004). Using computinghistory to enhance teaching. Proceedings of the

History of Computing in Education 2004 Conference,

Toulouse, France, 165–176.

Information Technology Association of America.(2002). Bouncing Back: Job skills and the continuing

demand for IT workers. Arlington, VA: InformationTechnology Association of America.

Inos, R. H., & Quigley, M. A. (1995). Synthesis of the

research on educational change. Part 4: The teacher’s role.

Honolulu, HI: Pacific Region Educational Laboratory.


standards for students. Retrieved December 1, 2005,from http://www.iste.org/nets


standards for teachers. Retrieved December 1, 2005,from http://www.iste.org/nets

Jeffrey, N. (1996). The responsibility of professionalassociations. Paper presented at A Meeting of the

Minds: ITEC Virtual Conference 1996, Australia.

Kavander, T., & Salakoski, T. (2004). Where have allthe flowers gone? Computer science education ingeneral upper secondary school. Proceedings of the

Kolin Kolistelut—Koli Calling Conference, Koli,Finland, 112–115.

Kubow, P. K., & Fossum, P. R. (2003). Comparative

education: Exploring issues in international context.

Upper Saddle River, NJ: Pearson Education.

Kuznetsov, A. A., & Beshenkov, S. A. (2005). Russianeducational standards of informatics andinformatics technologies (ICT): Aims, content,perspectives. In R. T. Mittermeir (Ed.) From computer





Lapidot, T., & Hazzan, O. (2003). Methods of teachingcomputer science courses for prospective teachers.SIGCSE Bulletin, 35(4), 29–34.

Latulipe, C., & Graham, S. (2005). Degaming thecurriculum: A failed experiment. RetrievedDecember 10, 2005, from: http://hci.uwaterloo.ca/students/clatulip/research.html

Lee, P. (2004). The computer science brain drain: A call to

revitalize computer science education. RetrievedApril 23, 2004, from https://www.wiki.cs.cmu.edu/public/uploads/Main/it-talent.pdf

Levy, D., & Lapidot, T. (2002). Shared terminology,private syntax: The case of recursive descriptions.

86

Proceedings of the ITiCSE 2002 Conference, Aarhus,Denmark, 89–93.

Loidl, S., Muhlbacher, J., & Scauer, H. (2005).Preparatory knowledge: Propaedeutic ininformatics. In R. T. Mittermeir (Ed.) From





McLaughlin, M. W. (1992). How district communitiesdo and do not foster teacher pride. Educational


Merritt, S. (1995). Reflections of a computer scientistfor teachers and teacher educators. In J. D. Tinsleyand T. J. van Weert (Eds.). Proceedings of the World

Conference on Computers in Education VI (4790486).London: Chapman and Hall for the InternationalFederation of Information Processing.

Micheuz, P. (2005). 20 years of computers andinformatics in Austria’s secondary academicschools. In R. T. Mittermeir (Ed.) From computer









Mitchell, W. (2002). Information technologyeducation: One state’s experience. Journal of


Mittermeir, R. T. (Ed.) (2005). From computer literacy to

informatics fundamentals. International Conference on

Informatics in Secondary Schools-—Evolution and

Perspectives, ISSEP 2005, Klagenfurt, Austria,March/April 2005.

Moorman, P., & Johnson, E. (2003). Still a stranger here:Attitudes among secondary school studentstowards computer science. Proceedings of the ITiCSE

2003 Conference, Thessaloniki, Greece, 193–197.

Morris, J. H. and Lee, P. (2004). The incrediblyshrinking pipeline is not just for women anymore.Computing Research News, 16(1), 20.

Muller, O., Haberman, B., & Averbuch, H. (2004). (An almost) pedagogical pattern for pattern-basedproblem-solving instruction. Proceedings of the

ITiCSE 2004 Conference, Leeds, United Kingdom,102–106.

Muller, O. (2005). Pattern oriented instruction and theenhancement of analogical reasoning. Proceedings of

The First International Computing Education Research

Workshop, October 1–2, 2005. Seattle, WA, 57–67.



O’Lander, R. (1996). Factors effecting high schoolstudent’s choice of computer science as a major.Proceedings of the Symposium on Computers and the

Quality of Life, Philadelphia, Pennsylvania, USA,25–31.

Paz, T., & Lapidot, T. (2004). Emergence of automatedassignment conceptions in a functionalprogramming course. Proceedings of the ITiCSE

2004 Conference, Leeds, United Kingdom, 181–185.

Payton, F. C. (2003). Rethinking the digital divide.Communications of the ACM, 46(6), 89–91.

87

Peckham, J., DiPippo, L., Reynolds, J., Paris, J.,Monte, P., & Constantinidis, P.A. (2000), Firstcourse in computer science: The discipline is morethan programming. Journal of Computing in Small

Colleges, 15(5), 223–230.

Perkins, D., Schwartz, S., & Simmons, R. (1988).Instructional strategies for the problems of noviceprogrammers. In R. E. Mayer (Ed.), Teaching and

learning computer programming (pp. 153–178).Hillsdale, NJ: Lawrence Erlbaum Associates.

Pham, B. (1997). The changing curriculum ofcomputing and information technology inAustralia. Proceedings of the ACSE 1997 Conference,

Melbourne, Australia, 149–154.

Poirot, J. L. (1979). Computer education in the secondaryschool: Problems and solutions. Proceedings of the

SIGCSE 1979 Conference, Dayton, Ohio, 101–104.

Poirot, J. L., Taylor, H. G., & Norris, C. A. (1988).Retraining teachers to teach high school computerscience. Communications of the ACM, 37(7),912–917.

Pollack, S., & Scherz, Z. (2005). Supporting projectdevelopment in CS – the effect on intrinsic andextrinsic motivation. Proceedings of the 10th PEG

Conference, Tampere, Finland, 143–148.

Ramberg, P. (1986). A new look at an old problem:Keys to success for computer science students.ACM SIGCSE Bulletin, 18(3), 36–39.

Ragonis, N., & Ben-Ari, M. (2005). A long-terminvestigation of the comprehension of OOPconcepts by novices. Computer Science Education,

15(3), 203–221.

Rich, L., Perry, H., & Guzdial, M. (2004). A CS1course designed to address interests of women.Proceedings of the SIGCSE 2004 Conference, Norfolk,Virginia, USA, 190–194.

Reiter, A. (2005). Incorporation of informatics inAustrian education: The project “Computer-Education-Society” in the school year 1984/85. InR. T. Mittermeir (Ed.) From computer literacy to




Roberts, E. (2004). The dream of a common language:The search for simplicity and stability in computerscience education. Proceedings of the SIGCSE 2004

Conference, Norfolk, Virginia, USA, 115–119.

Roberts, E,. & Halopoff, G. (2005). Computer ScienceTeachers Association analysis of high schoolsurvey data. Retrieved December 2, 2005 fromhttp://csta.acm.org/Research/sub/CSTANationalSurvey2004.html

Romberg, T. A., & Middleton, J. A. (1995).Conceptions of mathematics and mathematicseducation held by teachers. In N. Webb & T. A.Romberg (Eds.). Collaboration as a process for

reform. New York: Teachers College Press.

Rountree, N., Vilner, T., Wilson, B., & Boyle, R. (2004).Predictors for success in studying CS, panelsession. Proceedings of the SIGCSE 2004 Conference,

Norfolk, Virginia, USA, 145–146.

Rosenthal, T., Suppes, P., & Ben-Zvi, N. (2002).Automated evaluation methods with attention toindividual differences—A study of a computer-based course in C, Proceedings of the ASEE/IEEE

Frontiers in Education Conference, Boston, MA.

Sabin, M., Higgs, B., Riabov, V., & Moreira, A. (2005).Designing a pre-college computing course. Journal

of Computing in Small Colleges, 20(5), 123–132.

Sargent, J. (2004). An overview of past and projectedemployment changes in the professional IToccupations. Computing Research News, 16(3), 1–22.

88

Scherz, Z., & Haberman, B. (2005). Mini-projectsdevelopment in computer science—Students’ useof organization tools. Informatics in Education, 4,

307–319.

Schollmeyer, M. (1996). Computer programming in high school vs. college. Proceedings of the

SIGCSE 1996 Conference, Philadelphia, PA, USA,378–382.

Shackelford, R. (2005). Why can’t smart people figure out

what to do about computing education. Presentationat the CSTA/ISTE Computer Science andInformation Technology Symposium,Philadelphia, PA.

Sims-Knight, J. E., & Upchurch, R. L. (1993). Teachingsoftware design: A new approach to high schoolcomputer science. Paper presented at The Annual

Meeting of the American Educational Research

Association, Atlanta, GA. Retrieved November 1,2005 from http://www2.umassd.edu/cisw3/people/faculty/rupchurch/

Sleeman, D., Putnam, R. T., Baxter, J. A, & Kuspa, L.(1989). A summary of misconceptions of highschool basic programmers. In E. Soloway & J. C.Spohrer (Eds.). Studying the novice programmer (pp.301–314). Hillsdale, NJ: Lawrence ErlbaumAssociates.

Stephenson, C. (2002). High school computer scienceeducation. Journal of Computer Science Education,

Annual, 30–40.

Stevenson, D. E. (1993). Science, computationalscience and computer science: At a crossroads.Proceedings of the 1993 ACM Conference on

Computer Science, Indianapolis, IN, 7–14.

Stiller, E., & LeBlanc, C. (2003). Creating newcomputer science curricula for the newmillennium. Journal of Computing in Small Colleges,

18(5), 198–209.

Strauss, A. (1987). Qualitative analysis for social

scientists. New York: Cambridge University Press.

Taulbee, O. E. (2003). 2002–2003 Taulbee study:



Taylor, H. G., & Mounfield, L. C. (1989). The effect ofhigh school computer science, gender, and workon success in college computer science. ACM

SIGCSE Bulletin, 21(1), 195–198.

Taylor, H. G., & Mounfield, L. C. (1991). An analysisof success factors in college computer science:High school methodology is a key element.Journal of Research on Computing in Education,

24(2), 240–245.

The College Board. (2005). AP report to the nation.

Retrieved December 22, 2005, from http://www.collegeboard.com/about/news_info/ap/2005/index.html

The Scottish Parliament Information Centre. (1999).Education in Scotland. Retrieved June 5, 2005 fromhttp://www.scottish.parliament.uk/business/research/pdf_subj_maps/smda-10.pdf





Vandenberg, S., & Wollowski, M. (2000). Introducingcomputer science using a breadth-first approachand functional programming. Proceedings of the

SIGCSE 2000 Conference, Austin, TX, USA,180–184.

Ventura, P., & Ramamurthy, B. (2004). Wanted: CS1students. No experience required. Proceedings of

89

90

the SIGCSE 2004 Conference, Norfolk, Virginia,USA, 240–244.

Verno, A., Chiles, M., Gal-Ezer, J., Martin, J., &Stephenson, C. (2005). Finding a curriculum that

fits: The United States and Beyond. Panel presentedat the National Educational ComputingConference. Philadelphia, PA.

WHAT IS THE COMPUTER SCIENCE TEACHERS ASSOCIATION (CSTA)?

The Computer Science Teachers Association, a limited liability company under the auspices of ACM, has beenorganized to serve as a focal point for addressing several serious (crisis level) issues in K–12 computer scienceeducation, including:

• Lack of administrative, curricular, funding, professional development and leadership support for teachers• Lack of standardized curriculum• Lack of understanding of the discipline and its place in the curriculum• Lack of opportunities for teachers to develop their skills and interests• The above issues result in:• A profound sense of isolation, and• Dropping enrollment in college level computer science programs

There are other organizations that address use of technology across the curriculum, but only CSTA speaksdirectly and passionately for high school computer science.

OUR MISSIONCSTA is a membership organization that supports and promotes the teaching of computer science and othercomputing disciplines at the K–12 level by providing opportunities for teachers and students to betterunderstand the computing disciplines and to more successfully prepare themselves to teach and to learn.

OUR GOALSCSTA's organizational and educational goals include:

• Helping to build a strong community of CS educators who share their knowledge• Providing teachers with opportunities for high quality professional development• Advocating at all levels for a comprehensive computer science curricula• Supporting projects that communicate the excitement of CS to students and improve their understanding

of the opportunities it provides• Collecting and disseminating research about computer science education• Providing policy recommendations to support CS in the high school curriculum, • Raising awareness that computer science educators are highly qualified professionals with skills that

enrich the educational experience of their students

OUR SCOPEThe scope of the organization includes:

• High school (all aspects of computer science education)• Elementary and middle school (introducing problem solving and algorithmic thinking)• College/university (to establish better transition for high school programs and provide a greater

level of support to high school teachers)• Business and industry (supporting computer science education and teachers)

WHO SHOULD JOIN?All High School & Middle School Computer Science TeachersAll K-12 Computer Applications TeachersAll individuals interested (and passionate) about K–12 CS EducationThe first year of your CSTA membership is FREE!

JOIN US VIA…Web: http://csta.acm.org/Phone: 800.342.6626Email: [email protected]

The Following Companies are Gold Level Sponsors of CSTA

IBM

IS A

REG

ISTE

RED

TR

ADEM

ARK

OF

INTE

RN

ATIO

NAL

BU

SIN

ESS

MAC

HIN

ES C

OR

POR

ATIO

N; M

ICR

OSO

FT IS

A R

EGIS

TER

ED T

RAD

EMAR

K O

F M

ICR

OSO

FT C

OR

POR

ATIO

N; S

UN

IS

A R

EGIS

TER

ED T

RAD

EMAR

K O

F SU

N M

ICR

OSY

STEM

S, IN

C.

91

NOTES

84

NOTES

85