A Instructional Cognitive-Apprenticeship-InspiredApproach ... · Scientific Writing and Reading By Yifat Ben-David Kolikant, David W. Gatchell, Penny L. Hirsch, and Robert A. Linsenmeier

A Cognitive-Apprenticeship-InspiredInstructional Approach for TeachingScientific Writing and Reading

By Yifat Ben-David Kolikant, David W. Gatchell, Penny L. Hirsch,and Robert A. Linsenmeier

We present an approach for integrating instruction of scien-tific writing and reading into undergraduate science courses,inspired by the pedagogical theory of cognitive apprenticeship.We demonstrate its implementation and describe a study ofstudents 'feedback that enabled us to elicit students' difficultiesand fine-tune the next application accordingly.

,�.) ,

Reading and writing scientificliterature is an indispensablepart of a scientist's work.Specifically, scientific lit-

erature is arguably the most importantcommunication channel within thescientific community, making availablefor practitioners the collective wisdomand knowledge of the community (Ko-prowski 1998; Rice 1998). We believethat one of the objectives of higher edu-cation in science and engineering is tointroduce students to the process of sci-entific research and, correspondingly, tointroduce students to the importance ofreading and writing scientific literaturefor their professional lives.

Much work has led to the develop-ment of diverse approaches for integrat-ing reading, writing, and presentingof scientific literature into science andengineering courses. Working with theliterature can serve as a means to achievea variety of goals, such as improvingconceptual understanding (Auerbach,Bourgeois, and Collins 2004; Janick-Buckner 1997) and enhancing commu-nication skills (Glaser 2000; Koprowski1998; Rice 1998). There are differentapproaches to teaching communicationskills. At some universities there areseparate courses taught by instructorswith either scientific backgrounds (Rice1998) or with contrasting backgroundsin writing, English, and communication(Chisman 1998), whereas at other uni-versities, instructors integrate the teach-

Yifat Ben-David Kolikant (yifatbd [email protected]) is an assistant professor in theSchool of Education at the Hebrew University of Jerusalem. David W. Gatchell ([email protected]) is a research associate in the Department of BiomedicalEngineering and VaNTH Engineering Research Center in Bioengineering EducationalTechnologies, Penny L. H-irsch ([email protected]) is associate director ofThe Writing Program, and RobertA. Linsenmeier ([email protected])isaprofessorin the Departments of Biomedical Engineering, Neurobiology, Physiology,and VaNTH Engineering Research Center in Bioengineering Educational Technologiesat Northwestern University in Evanston, illinois.

20 JOURNAL of COLLEGE SCIENCE TEACHING

ing of communication skills into theirscience courses. Students' assignmentsinvolve writing scientific biographies,papers (Smith 2001), proposals (Felzienand Cooper 2005), peer reviews (Ko-prowski 1998), and reviews of scientificpapers (Glaser 2000; Chisman 1998).

However, writing is a particularform of problem solving (Berkenkot-ter 2004; Flower and Hayes 1997).Novice writers, such as students,typically begin writing hoping thatthey will serendipitously articulate the"right" sentence(s), which will carrythem through the whole written draft.In contrast, experts use sophisticatedstrategies, such as setting and resettingwriting goals, generating ideas, explor-ing relationships among the ideas, andfinally connecting them in some kind ofanalytic framework aimed at a specificreader. Practice and assignments alonewill not turn novice writers into expertwriters. Students need scaffolded in-struction, as well as experience, to mas-ter higher-level skills in writing such assynthesis and argumentation (for moreinformation, see the extensive literatureon Writing Across the Curriculum, e.g.,Berkenkotter 2004; Bazerman and Rus-sell 1994).

Therefore, we present an inte-grated approach to teaching studentsto read, understand, and write scientificliterature. The approach is based on thecognitive apprenticeship instructionalmodel (CAIM'), which exposes studentsto the thinking processes that profession-als carry out in professional tasks, andallows them to experiment with expertstrategies while mentored in an authenticcontext (Collins, Brown, and Holum1991; Collins, Brown, and Newman1989). The CALM has been implement-ed successfully in other domains such asteaching mathematical problem solving(Schoenfeld 1987), and teaching readingand writing (Bereiter and Scardamalia1987). (For more examples, see Wilsonand Cole 1991.) We describe how thisapproach was used in a junior-levelanimal physiology class.

The cognitive apprenticeshipinstructional model (CALM)In the CAIM, teaching consists of the

following methods:* Modeling--an expert performs tasks

while reflecting on the thought pro-cesses involved, thereby making theseprocesses visible to apprentices.

* Coaching--an expert observesstudents carrying out a task andprovides scaffolded instruction.

* Articulation--students articulatetheir knowledge, reasoning, andproblem-solving processes.

* Reflection--students comparetheir work to that of the expert.

* Exploration--students are encour-aged to solve problems on their own.

Designing an instructional activitybased on the CAHIM requires analyzingan expert's knowledge beyond that ofcontent and focusing on the expert'sactivities, as well as heuristics andstrategies gained by experience. CAIMis well suited to writing instruction incollege science classes because highereducation brings together students,who are novice writers, with faculty,who possess expert knowledge, such asdiscipline-specific writing conventionsand where and how to search for scien-tific literature. In addition, faculty pos-sess important cognitive skills, such asanalysis and synthesis, to manipulateknowledge. Analysis, when associ-ated with scientific literature, involvesunderstanding how others contributedto a research question, identifyingwhether the evidence supports theauthors' claims, and assessing whetherthe authors' methods are valid. Synthe-sis, according to Bloom (1956), refersto creating a new idea by combininginformation from different sources,and in scientific writing involves re-organizing one's own knowledge, andthat of others, to answer the researchquestion. In theory, a CAlM-inspiredapproach should benefit novice writersbecause they can reflect on their ownapproach to writing and develop theircritical-thinking and writing skills,having seen how an expert engagesin analysis and synthesis. Yet, coach-ing and modeling require theoreticalknowledge and pedagogical practicesthat are not typically part of a scienceinstructor's repertoire.

Our instructional approachThe assignmentTo teach the process of scientificreading and writing in a contextthat resembled authentic scientificwork, students were assigned a scien-tific paper. Under the general topic of"animals' adaptations'" students wereasked to (1) choose an animal with aphysiological ability not possessed byhumans (e.g., hibernation, infraredsensation), (2) analyze two to fourarticles from the research literature onthis topic, and (3) compare the functionof this system with analogous systemsin other species, including humans.Because we wanted students to focuson tasks associated with the literature,the assignment did not include labora-tory experiments and thus the papersdid not include an empirical part.

Students began by writing aresearch proposal including (1) aresearch question, (2) a paragraph de-scribing their interest in the topic, and(3) two to four references to be used toanswer the question. The proposal wassubmitted to the instructor and to twopeers for review. Instructor feedbackhelped students examine the validityand feasibility of the proposed work.The peer review provided students anopportunity to reflect upon their ownwork and to play the role of reviewercommon in the scientific community.Next, students had three weeks towrite their papers.

The scaffoldingWe divided the anticipated work into tasksand outlined the scaffolding needed foreach task in tenms of practical informationand expert practices. Tasks were providedin various ways, as shown inTable 1. Scaf-folding included three documents (Dl,D2, and D3) that the instructor posted onthe course website and used in two oraldiscussion sessions (51 and S2). Also,the evaluation form (E) was posted priorto the assignment's deadline to encouragestudents to align their activities with oureducational goals (Table 3). (These formscanbe downloadedfiromtheVaNTHwxeb-site, http.'//www.vanth.org/curriculum/communications.asp.) Finally, theinstructor provided flexible office

NOVEMBER/DECEMBER 2006 21

hours to meet with students about theassignment.

Documents. In document Dlwe defined the assignment, providedstructure for the proposal and the paper,provided initial guidance (e.g., sampleresearch questions), and listed the futurescaffolds. Documents D2 and D3 sup-ported the activities required to writethe research proposal and the paper,respectively. Both documents includedstrategies to direct students to their nextactivity and to reason through the pro-cess. Excerpts like these exemplify thestrategies provided:

* "If you have an idea, your problemis going to be making the topic man-ageable by restricting your focus,"

* "If you have a rough idea for atopic, learning a little bit about itbefore launching into the seriousresearch literature is+ a good idea."

* "You need to keep your audience[other physiologists] in mind."

Discussion sessions. There wereno discussion sessions in this courseprior to the described intervention.Therefore, sessions S 1 and $2 weremade optional. Nonetheless, hay-

ing a separate session for each topicunderscored their importance to theassignment.

S 1 helped students search andfind literature relevant to their researchquestion. This session was useful, butnot innovative. S2 modeled the thoughtprocesses of an expert, namely the in-structor, and thus was essential to theCAIM. First, the instructor translatedthe meaning of analysis and synthesisinto the practical questions he askshimself when he surveys literature.

Next, he engaged students inan activity intended to illustrate the

The scaffolding design.

* Finding a topic

I . '-U.

U *U U. - -. * *UU . - - ��U8.

I I

1. Examples of research questions2. Information about journals useful for defining a question (orresearching a question)3. Strategies for finding a question from the literature4. Heuristics for exploring the topic before deciding on theactual resources

SiS1

S1$1

DI02

020:2

yes

yesyes

1 .A list of online field-specific search engines S1 02 -

* Finding relevant 2. A description of the limitation of general-purpose $1 D2 -

original literature (nonscientific) search engines*Gtigoiialieaue 3.A description of the procedure required for retrieving the Si D2 -SGeting rigial lteraure relevant literature

4. Heuristics used by scientific professionals to read the literature 52--

1 .A detailed structure of the proposal -D1

* Composing the proposal 2. Instructions for peer review S2--3. Explanation of experts' criteria for review 52--

* Analysis of the resources withrespect to their original goals,

aeswellasc qetiontepooe 1. Explanations and demonstrations of the meaning andrsacqusinexpected outcomes of each of the tasks within the context of 52 03 yes

* Synthesis of resources into the scientific reading/writing and the specific writing assignmentanswer to the research question

* Evaluation of the limitation ofthe answer

* Composing the paper

1. A description of the conventions of scientific writing:structure and mechanics2. Class activity to demonstrate the conventions inscientific literature3. An explanation of the gap between genuine scientific papersand students' papers, and the pedagogical reason underlyingthis difference4. Additional resources such as books and e-books that supportwriting

52

52

$2

03

03, E

03


structure and benefits of writing con-ventions used in physiology. To thisend, he brought scientific papers to theclass. Each student selected a differentpaper. They read the headings aloudas the instructor, in turn, wrote themon the blackboard. Obviously, mostof the papers had similar headings,e.g., abstract, introduction, methods,results, discussion, and conclusions.

Students were then asked to out-line the content of each section, guidedby questions such as, "What shouldyou expect to find in the introduc-tion?" The instructor organized theseresponses on the blackboard, outliningthe contents of each of the paper'sheadings.

He also directed students' atten-tion to writing norms. For example,results are typically organized aroundthe figures and tables, while the dis-cussion usually includes few, if any,figures. Moreover, within the discus-sion, he was able to demonstrate hisfamiliarity with the journals by guess-ing correctly, simply from students'descriptions of an article's placementof graphs and tables, which journalhad published which article. This wasa very dramatic and visible demon-stration of how thoroughly profes-sionals master the literature resourcesin their respective fields.

Finally, in order to ease the processof choosing resources from the litera-ture, the instructor articulated the think-ing process he follows when perusingthe literature, thereby demonstratingthe evaluative nature of this process.He first reads the introduction in orderto determine if the paper is relevantto his interest, then reads part of thediscussion section to see whether theresults support the hypothesis. He thenjumps to the results, focusing on thefigures and charts because they sum-marize the important data. Then, if thearticle suits his interest, he delves intothe details and reads the entire section.He scans the methods section, merelyverify,ing the reasonability of the ex-periments. If an innovative method isoffered, his reading will become morethorough and critical. The point of thisdemonstration was not to show that

Students' responses regarding the usefulness of the discussion rating eachcategory on a scale of 1 (no importance) to 4 (very important).

Analysis/synthesis 3.59 0.62 3.56 0.63

Writing 3.47 0.66 3.13 1.08

Reading 3.24 0.66 3.35 0.49

Evaluation of students' papers (rubric).

a. Clear explanation of question 23%

b. Explanation of why question is important 5%

c. Appropriate background to explain topic 18%

d. Appropriate use of references27

a. Clarity of analysis of original articles: question or hypothesis, 82%

methods, results, conclusions

b. Depth/quality of analysis of original articles 55%

c. Use of figures and/or tables to support explanation of 23%original articles

d. Discussion of points not clear in the original articles, if 0%applicable

a. Discussion of the limitations of original research articles 23%

b. Synthesis of an answer to the original question 18%

c. Synthesis of suggestions for future work and/or new 23%hypotheses, if applicable

d. Comparison of the animal under discussion to other animals 9%

a. Coherence of topics discussed 9%

b.Appropriateness of major articles 23%

a. No more than 6 pages, 1.5 line-spaced, 1 -inch margins, 11-point 0%or larger font (references, figures, and tables outside this limit.)

b. Legends for figures and tables 18%

c. References in consistent, complete format 27%

d. Spelling, punctuation, grammar 50%

e. Using words carefully and precisely 59%

f. Submission of original articles with your paper 0%


the instructor's strategy is superior, butrather to demonstrate that professionalshave a reading strategy, one which writ-ers need to consider. This is especiallytrue because professionals rarely readpapers from beginning to end, and skima great deal of literature.

Instructor evaluation ofstudent papersThe top of the evaluation form includeda shortened version of the assignment.Below that were listed the evaluationcriteria for the papers, presented in theleft column in Table 3. Grades wereawarded to each component of the pa-per based upon a list of subcriteria thatmatched the instructions. For example,the introduction was graded accordingto the following criteria:

* clarity of the question,* explanation of its importance,* appropriate background to explain

the topic, and* appropriate use of references.

Additionally, points were given for theoverall organization of the paper as wellas for mechanical issues, such as punc-tuation. The instructor wrote additionalcomments when points were taken off,as well as general comments on thebottom ofthe evaluation form and localcomments in the body of the paper.

Lastly, in both discussion ses-sions the instructor explicitly ex-plained to students the gap betweena professional review and his reviewof their work in order to help themalign their future learning trajectories.Specifically, the instructor eased thedemands in the discussion section, inparticular the subcriteria 3c and 3d, inwhich students were asked to evaluatethe limitation of their answers and tomake their own suggestions, whichrequire a broader knowledge of thefield. Therefore, students were penal-ized solely for ignoring these tasks orfor writing unreasonable arguments.

Study designIn order to fine-tune our instruction,we queried students at the beginningof the course about their experience


in writing scientific papers, as well astheir competency in associated taskssuch as searching for resources andciting them appropriately.

To better understand students'need for scaffolding, we asked them tofill out a survey anonymously after thesecond discussion session. We listedthe three parts of that session andasked students to rank the importanceof each from 1 (no importance) to 4(very important) for accomplishingthe assignment and for their profes-sional future, and to explain why.

The class had an enrollment of 42students. We received 17 responsesfrom the 24 students who participatedin that session. We calculated themean and standard deviation of theirresponses and specified the commonreasons as to why certain scaffoldingswere found to be useful (or not).

Next, to measure the gap betweenstudents' performance and the instruc-tor's expectations (reflected in therubric), we calculated the average andstandard deviation of the grades. Wealso collected a sample of 22 paperswith the instructor's evaluations andextracted the common difficulties.

FindingsStudents'prior experienceFor 63% of the students, this was theirfirst science course that had a smallenough enrollment to allow for a sub-stantial writing assignment. Only 7%had experienced more than two writingassignments. In addition, 40% said theydid not know where to look for recentwork done on a given topic, 86% did notknow where to find citations to a givenpaper, and about a third did not knowwhat to do in case a paper they neededwas not in the university library.

Studen ts"'opinions of theCAlM-inspired discussion sessionStudents' responses to the surveysregarding the usefulness of the seconddiscussion session are summarized inTable 2. The left column specifies thetopic discussed, while the second andthird present the mean and standarddeviation of students' responses ona 4-point r'cale and the importance

of each (1--no importance; 4--veryimportant).

Most students who participatedin the discussion session found ituseful. It addressed their short-termgoals by helping them to accomplishthe assignment.

The most common answer to whythe discussion session was relevant hadthe following pattern: "I will [or willnot] need it in my professional future asa [profession].' Interestingly, synthesisand analysis were perceived as impor-tant for both long-term and short-termgoals. However, writing was perceivedas less important for the future than forthe assignment. The dominant reason,as shown on the surveys, was that stu-dents believed that unless their futurewas in academia, they would not needwriting skills. Alternatively, readingwas perceived as being more importantfor long-term goals than for short-termgoals, with many students perceivingit as a general skill, important for anyscientifically literate profession.

Analysis of students' papersThe average of students' grades onthe papers was 91.88 and the standarddeviation was 4.48. We concludedthat, in general, students performedsatisfactorily on this assignment.

The grading rubric and subcriteriaallowed us to examine students' writ-ing achievement in greater detail. Theaverage of the sample of the 22 paperswe chose to evaluate more closely was92 and the standard deviation was 4.4.Thus, the sample was representative ofthe entire population. Table 3 presentsresuflts from the analysis of the evalua-tions of the sample. The numbered head-ings in the left column designate onecriterion, and the maximum number ofpoints that students could score is shownin parentheses. Below each criterion isan expanded list of specific subcriteria.The right column presents the distribu-tion of students who lost points in anyof the corresponding subcriteria.

Most students lost points in the areaof "content of literature review"--cri-terion (2) in the rubric (see Table 3). Inparticular, 82% of the students did notmeet the instructor's expectation regard-

ing the clarity of analysis of originalarticles (subcriterion 2a), and 55%performed unsatisfactorily regarding thedepth of the analysis (subcriterion 2b).In contrast, the small number of studentswho were penalized in the introductionsection implies that the scaffoldings ofthis task were satisfactory.

Additionally, a small number ofstudents' performances were penalizedin the discussion sections of their papers.The instructor added verbal commentsthat pinpointed invalid or incomplete in-ference chains, such as, "I am not clear towhat extent you told us about these"' and,"Would this be useful for humans?"

DiscussionThe fact that students performed at sucha high level on this assignment, giventheir limited experience and evidenceof the difficulties reported in previouswork (Troy, Hirsch, and Smith 2004),implies that our approach was useful.

A further analysis of students' per-formance revealed two major issues thatneed to be addressed. First, most studentsfound the scaffolding to be useful for theshort-term goals of accomplishing thepaper assignment; however, a significantnumber of students believed that the skillof scientific writing is necessary only fora career in research. Such an attitude mayhave diminished their engagement in thisassignment. In the next application, wewill address this issue by demonstratingthe life cycle of scientific work and theindispensable part played by the litera-ture, as well as the need to communicateclearly in all careers.

Second, many students were pe-nalized for their performance on thesection of literature review, yet wereable to synthesize an answer to theirresearch question from these papers.This implies that students understoodthe papers, but did not know how tocommunicate their knowledge. Thus,we need to fine-tune the scaffolded in-

struction in this area. In the next phase,we may provide students with questionsto assist them with summarizing andcritiquing the original literature.

Arguably, there are books that pro-vide the information needed to carry outa writing assignment. However, because

the assignment is time-consuming, read-ing an entire book might deter studentsrather than help. In our CAIM-inspiredapproach, the same knowledge was madeaccessible to students within a relevantcontext. Furthermore, the instructionincluded modeling of the instructor'sthinking process in similar situations,coaching students while providing feed-back; and clear instructions and criteria,thereby demonstrating the importanceof these skills to the instructor's ownprofessional life. U

AcknowledgmentThis work was supported by the Na-tional Science Foundation, Award#EEC-98 76363.

ResourceVaNTH ERG--www. vanth,.org.

ReferencesAuerbach, J.L., G.M. Bourgeois, and

T.R. Gollins. 2004. Do studentsbenefit? Writing-to-learn in a digitaldesign laboratory course. In Proceed-ings of the 34thA SEE/IEEE Frontiersin Education Conference. NewYork:IEEE.

Bazerman, G., and D.R. Russell.1994. Landmark essays on writingacross the curriculum. Davis, GA:Hermagoras Press.

Bereiter, G., and M. Scardamalia. 1987.The psychology of written com-position. Hillsdale, N J: LawrenceErlbaum.

Berkenkotter, G. 2004. Writing andproblem solving. In Languageconnections, writing and read-ing across the curriculum, eds.T. Fulwiler and A. Young, 33-44.Urbana, IL: National Gouncil ofTeachers of English. Available athttp.://wac. colostate. edu/books/language_connections.

Bloom, B.S., ed. 1956. Taxonomy ofeducational objectives: The classifi-cation of educational goals. Hand-book I, cognitive domain. New York,Toronto: Longmans, Green.

Chisman, J.K. 1998. Introducing collegestudents to the scientific literatureand the library. Journal of CollegeScience Teaching 28 (1): 39-42.

Collins, A., J.S. Brown, andA. Holum.1991. Gognitive apprenticeship:Making thinking visible. AmericanEducator 12 (6): 38-47.

Gollins, A., J.S. Brown, and S.E. New-man. 1989. Cognitive apprenticeship:Teaching the crafts of reading, writ-ing, and mathematics. In Knowing,learning, and instruction: Essaysin honor of Robert Glaser, ed. L.B.Resnick, 453-94. Hillsdale, NJ:Lawrence Erlbaum.

Felzien, L., and J. Cooper. 2005. Modelingthe research process. Journal of Col-lege Science Teaching 34 (6): 42-46.

Flower, L., and J.R. Hayes. 1997. Prob-lem solving strategies and the writingprocess. College English 39: 451.

Glaser, ES. 2000. Journal clubs--Asuccessful vehicle to science lit-eracy. Journal of College ScienceTeaching 29 (5): 320-24.

Janick-Buckner, D. 1997. Getting un-dergraduates to critically read anddiscuss primary literature. Journalof College Science Teaching 27(1): 29-32.

Koprowski, J.L. 1998. Sharpening thedraft of scientific writing. Journalof College Science Teaching 27(2): 133-35.

Rice, R.E. 1998. "Scientific writing"--A course to improve the writing ofscience students. Journal of CollegeScience Teaching 27 (4): 267-71.

Schoenfeld, A.H. 1987. What's all thefuss about metacognition'? In Cogni-tive science and mathematics educa-tion, ed. A.H. Schoenfeld, 189-215.Hillsdale, NJ: Lawrence Erlbaum.

Smith, G.R. 2001. Guided literatureexplorations. Journal of CollegeScience Teaching 30 (7): 465-69.

Troy, J., P. Hirsch, and H.D. Smith.2004. Team-based communicationexercises for biomedical engineer-ing juniors: Where to do it and whatworks. In Proceedings of the Ameri-can Society for Engineering Educa-tion (ASEE). Portland, OR: ASEE.

Wilson, B., and P. Cole. 1991. A re-view of cognitive teaching models.Educational Technology Researchand Development Journal 39 (4):47-64. Available online at www.cudenrver, edu/~bwilson


COPYRIGHT INFORMATION

TITLE: A Cognitive–Apprenticeship–Inspired InstructionalApproach for Teaching Scientific Writing and Reading

SOURCE: J Coll Sci Teach 36 no3 N/D 2006WN: 0630502498008

The magazine publisher is the copyright holder of this article and itis reproduced with permission. Further reproduction of this article inviolation of the copyright is prohibited. To contact the publisher:http://www.nsta.org/college

Copyright 1982-2006 The H.W. Wilson Company. All rights reserved.