1
DEVELOPMENT OF THE 3DLOP The ThreeDimensional Learning Observation Protocol (3DLOP) will characterize the three dimensions in teaching (in development) and what teaching activities are occurring in a class meeting (complete) (Matz, 2015). Teaching activities: clicker questions, tasks, interactions, lecture, administration, and miscellaneous. Preliminary analyses show that courses incorporating 3D questions on exams tend to be active, but active courses do not necessarily incorporate 3D questions on exams (Fig. 3). DISCUSSION Coding exams with the 3DLAP reveals a shift in how some faculty are approaching assessments in gateway science courses. Different departmental cultures have contributed to, and help explain why, some courses exhibit change and others do not: Changes in introductory cell and molecular biology were supported by a substantial, permanent increase in TA/LA funding. Changes in general chemistry were supported by the departmentwide adoption of the researchbased CLUE curriculum (Cooper, 2013). A dynamic system of interactions, going beyond resources provided for this project, have contributed to these outcomes, including faculty hires, related externallyfunded projects, and demonstrated institutional commitment to STEM teaching and learning. CONCLUSIONS We have shown that assessments in some gateway courses increasingly reflect scientific practices and core ideas as the result of a system of transformation efforts at Michigan State University. Additional ongoing work includes studying persistence rates and the relationships between threedimensional assessment, instruction, and student learning. ACKNOWLEDGEMENTS: We thank the Association of American Universities, the Deans of the College of Natural Science, and all MSU faculty members who participated in disciplinary discussions and provided materials for this study. REFERENCES: (1) Cooper, M., & Klymkowsky, M. (2013). Journal of Chemical Education, 90(9), 1116– 1122. (2) Cooper, M. M., et al. (2015). Science, 350(6258), 281282. (3) Gwet, K. L. (2008). British Journal of Mathematical and Statistical Psychology, 61(1), 2948. (4) Laverty, J. T., et al. (2016). PloS one, 11(9), e0162333. (5) Matz, R. L., et al. Paper presented at NARST, April 1114, Chicago, IL. (6) National Research Council. (2012). Washington, DC: The National Academies. AFFILIATIONS: 1 Hub for Innovation in Learning and Technology, 2 Physics and Astronomy, 3 Chemistry, 4 STEM Transformation Institute, 5 Plant Biology, 6 Microbiology and Molecular Genetics, 7 Biochemistry and Molecular Biology; a Michigan State University, b Florida International University, c Grand Valley State University, d Kansas State University DATA COLLECTION & ANALYSIS – ASSESSMENTS We collected four years of midterm and final exams from all major introductory biology, chemistry, and physics lecture courses (Table 1). In total, we analyzed 4,023 questions from 134 unique exams, fully representing all 185 course sections of the eight relevant courses. Disciplinary coding groups achieved acceptable interrater reliability as measured by Gwet’s AC 1 (Gwet, 2008) and percent agreement. RESULTS – ASSESSMENTS RQ: How has the fraction of assessment items that reflect scientific practices, crosscutting concepts, and core ideas changed over time? Figure 2. Fraction of exam points that reflect the three dimensions over time in A) Bio I, B) Bio II, C) Chem I & II, and D) PhyAlgebra & Calculus I and II. Points are scaled by the number of students who took each exam. RQ: How have student grades and Dgrade, Fgrade, and withdrawal (DFW) rates changed in these courses of interest over time? THREEDIMENSIONAL LEARNING A 2012 National Research Council report introduced the idea of three dimensional learning as a vision for science and engineering education. Scientific practices: What scientists do, e.g. analyze and interpret data Crosscutting concepts: Concepts with common applications across fields, e.g. structure and function Core ideas: Explanatory and generative disciplinary ideas, e.g. evolution drives the diversity of life GATEWAY COURSE TRANSFORMATION The goal of our transformation project is to develop and enact a shared vision for undergraduate gateway biology, chemistry, and physics courses by engaging faculty in disciplinary discussions about the three dimensions, among other activities (Cooper, 2015). We are characterizing changes in instruction (video observations) and assessments (exams). Figure 1. Schematic of integrating the three dimensions. DEVELOPMENT OF THE 3DLAP The ThreeDimensional Learning Assessment Protocol (3DLAP) characterizes the potential for assessment items to elicit evidence of threedimensional learning (Laverty, 2016). Developed through iterations of literature review, discussion with disciplinary experts, and review of existing assessment items. Criteria were developed for each scientific practice, crosscutting concept, and core idea, e.g., for analyzing and interpreting data: 1. Question gives a scientific question, claim, or hypothesis to be investigated. 2. Question gives a representation of the data (e.g., table, graph, list of observations) provided to answer the question or test the claim or hypothesis. 3. Question gives an analysis of the data or asks student to analyze the data. 4. Question asks student to interpret the results or assess the validity of the conclusions in the context of the question, claim, or hypothesis. Characterizing Scientific Practices, Crosscutting Concepts, and Core Ideas in Science Assessments Rebecca L. Matz 1,a , Marcos D. Caballero 2,a , Justin H. Carmel 3,4,b , Diane EbertMay 5,a , Cori L. FataHartley 6,a , Debbie G. Herrington 3,c , James T. Laverty 2,d , Lynmarie A. Posey 3,a , Ryan L. Stowe 3,a , Jon R. Stoltzfus 7,a , Sonia M. Underwood 3,4,b , Melanie M. Cooper 3,a Table 1. Numbers of exams and questions coded with the 3DLAP by discipline. Discipline Exams Questions Individual questions Clusters Total Biology 68 2237 266 2503 Chemistry 32 718 93 811 Physics 34 479 230 709 Total 134 3434 589 4023 Table 2. Student outcomes: Average grades and DFW rates in introductory courses. Approx.annual enrollment Avg. Grade DFW Year 0 Year 3 Year 0 Year 3 Biology I (cell/mol) 2,400 2.3 2.6 31% 21% BiologyII (org/pop) 1,000 3.0 3.0 11% 13% Chemistry 5,000 2.3 2.9 35% 17% Physics 4,500 3.0 3.0 8% 9%

2017-11-03 AAC&U poster (1) 45.pdfDEVELOPMENT*OF*THE*3D.LOP! The$Three&Dimensional$Learning$Observation$Protocol$(3D&LOP)$will$ characterize$the$three$dimensions$in$teaching$(in$development)$and$

Embed Size (px)

Citation preview

Page 1: 2017-11-03 AAC&U poster (1) 45.pdfDEVELOPMENT*OF*THE*3D.LOP! The$Three&Dimensional$Learning$Observation$Protocol$(3D&LOP)$will$ characterize$the$three$dimensions$in$teaching$(in$development)$and$

DEVELOPMENT  OF  THE  3D-­‐LOP§ The  Three-­‐Dimensional  Learning  Observation  Protocol  (3D-­‐LOP)  will  

characterize  the  three  dimensions  in  teaching  (in  development)  and  what  teaching  activities  are  occurring  in  a  class  meeting  (complete)  (Matz,  2015).

§ Teaching  activities:  clicker                                                                                                                                              questions,  tasks,  interactions,                                                                                                                          lecture,  administration,  and                                                                                                                    miscellaneous.  

§ Preliminary  analyses  show                                                                                                                                                    that  courses  incorporating  3D                                                                                                                                            questions  on  exams  tend  to  be                                                                                                                                  active,  but  active  courses  do                                                                                                                                                not  necessarily  incorporate  3D                                                                                                                                            questions  on  exams  (Fig.  3).  

DISCUSSION§ Coding  exams  with  the  3D-­‐LAP  reveals  a  shift  in  how  some  faculty  are  

approaching  assessments   in  gateway  science  courses.§ Different  departmental  cultures  have  contributed  to,  and  help  explain  

why,  some  courses  exhibit  change  and  others  do  not:  § Changes  in  introductory  cell  and  molecular  biology  were  

supported  by  a  substantial,  permanent  increase  in  TA/LA  funding.§ Changes  in  general  chemistry  were  supported  by  the  

department-­‐wide  adoption  of  the  research-­‐based  CLUE  curriculum  (Cooper,  2013).

§ A  dynamic  system  of  interactions,  going  beyond  resources  provided  for  this  project,  have  contributed  to  these  outcomes,  including  faculty  hires,  related  externally-­‐funded  projects,  and  demonstrated  institutional  commitment  to  STEM  teaching  and  learning.

CONCLUSIONSWe  have  shown  that  assessments   in  some  gateway  courses  increasingly  reflect  scientific  practices  and  core  ideas  as  the  result  of  a  system  of  transformation  efforts  at  Michigan  State  University.  Additional  ongoing  work  includes  studying  persistence  rates  and  the  relationships  between  three-­‐dimensional  assessment,  instruction,  and  student  learning.

ACKNOWLEDGEMENTS: We  thank  the  Association  of  American  Universities,   the  Deans  of  the  College  of  Natural  Science,  and  all  MSU  faculty  members  who  participated  in  disciplinary  discussions  and  provided  materials  for  this  study.REFERENCES: (1)  Cooper,  M.,  &  Klymkowsky,  M.  (2013).  Journal  of  Chemical  Education,  90(9),  1116–1122.  (2)  Cooper,  M.  M.,  et  al.  (2015).  Science,  350(6258),  281-­‐282.  (3)  Gwet,  K.  L.  (2008).  British  Journal  of  Mathematical  and  Statistical  Psychology,  61(1),  29-­‐48.  (4)  Laverty,  J.  T.,  et  al.  (2016).  PloSone, 11(9),  e0162333.  (5)  Matz,  R.  L.,  et  al.  Paper  presented  at  NARST,  April  11-­‐14,  Chicago,  IL.  (6)  National  Research  Council.  (2012).  Washington,  DC:  The  National  Academies.AFFILIATIONS: 1Hub  for  Innovation  in  Learning  and  Technology,  2Physics  and  Astronomy,  3Chemistry,  4STEM  Transformation  Institute,  5Plant  Biology,  6Microbiology  and  Molecular  Genetics,  7Biochemistry  and  Molecular  Biology;  aMichigan State  University,  bFlorida International  University,  cGrand Valley  State  University,  dKansas State  University

DATA  COLLECTION  &  ANALYSIS  – ASSESSMENTS§ We  collected  four  years  of  midterm  and  final  exams  from  all  major  

introductory  biology,  chemistry,  and  physics  lecture  courses  (Table  1).§ In  total,  we  analyzed  4,023  questions  from  134  unique  exams,  fully  

representing  all  185  course  sections  of  the  eight  relevant  courses.  

§ Disciplinary  coding  groups  achieved  acceptable  inter-­‐rater  reliability  as  measured  by  Gwet’s AC1 (Gwet,  2008)  and  percent  agreement.  

RESULTS  – ASSESSMENTSRQ:  How  has  the  fraction  of  assessment   items  that  reflect  scientific  practices,  crosscutting  concepts,  and  core  ideas  changed  over  time?

Figure  2.  Fraction  of  exam  points  that  reflect  the  three  dimensions  over  time  in  A)  Bio  I,  B)  Bio  II,  C)  Chem I  &  II,  and  D)  Phy-­‐Algebra  &  -­‐Calculus  I  and  II.  Points  are  scaled  by  the  number  of  students  who  took  each  exam.  

RQ:  How  have  student  grades  and  D-­‐grade,  F-­‐grade,  and  withdrawal  (DFW)  rates  changed  in  these  courses  of  interest  over  time?  

THREE-­‐DIMENSIONAL  LEARNINGA  2012  National  Research  Council  report  introduced  the  idea  of  three-­‐dimensional  learning  as  a  vision  for  science  and  engineering  education.§ Scientific  practices:  What  scientists  do,  e.g.  analyze  and  interpret  data§ Crosscutting  concepts:  Concepts  with  common  applications  across  

fields,  e.g.  structure  and  function§ Core  ideas:  Explanatory  and  generative  disciplinary  ideas,  e.g.  

evolution  drives  the  diversity  of  life

GATEWAY  COURSE  TRANSFORMATIONThe  goal  of  our  transformation  project  is  to  develop  and  enact  a  shared  vision  for  undergraduate  gateway  biology,  chemistry,  and  physics  courses  by  engaging  faculty  in  disciplinary  discussions  about  the  three  dimensions,  among  other  activities  (Cooper,  2015).  We  are  characterizing  changes  in  instruction  (video  observations)  and  assessments  (exams).

Figure  1. Schematic  of  integrating  the  three  dimensions.

DEVELOPMENT  OF  THE  3D-­‐LAP§ The  Three-­‐Dimensional  Learning  Assessment  Protocol  (3D-­‐LAP)  

characterizes  the  potential  for  assessment   items  to  elicit  evidence  of  three-­‐dimensional  learning  (Laverty,  2016).

§ Developed  through  iterations  of  literature  review,  discussion  with  disciplinary  experts,  and  review  of  existing  assessment   items.

§ Criteria  were  developed  for  each  scientific  practice,  crosscutting  concept,  and  core  idea,  e.g.,  for  analyzing  and  interpreting  data:1. Question  gives  a  scientific  question,  claim,  or  hypothesis  to  be  

investigated.2. Question  gives  a  representation  of  the  data  (e.g.,  table,  graph,  list  of  

observations)  provided  to  answer  the  question  or  test  the  claim  or  hypothesis.

3. Question  gives  an  analysis  of  the  data  or  asks  student  to  analyze  the  data.

4. Question  asks  student  to  interpret  the  results  or  assess  the  validity  of  the  conclusions  in  the  context  of  the  question,  claim,  or  hypothesis.

Characterizing  Scientific  Practices,  Crosscutting  Concepts,  and  Core  Ideas  in  Science  Assessments

Rebecca  L.  Matz1,a,  Marcos  D.  Caballero2,a,  Justin  H.  Carmel3,4,b,  Diane  Ebert-­‐May5,a,  Cori  L.  Fata-­‐Hartley6,a,  Debbie  G.  Herrington3,c,  James  T.  Laverty2,d,  Lynmarie  A.  Posey3,a,  Ryan  L.  Stowe3,a,  Jon  R.  Stoltzfus7,a,  Sonia  M.  Underwood3,4,b,  Melanie  M.  Cooper3,a

Table  1.  Numbers  of  exams  and  questions  coded  with  the  3D-­‐LAP  by  discipline.

Discipline ExamsQuestions

Individual  questions Clusters TotalBiology 68 2237 266 2503

Chemistry 32 718 93 811Physics 34 479 230 709Total 134 3434 589 4023

Table  2.  Student  outcomes:  Average  grades  and  DFW  rates  in  introductory  courses.  

Approx.  annual  enrollment

Avg.  Grade DFWYear  0 Year  3 Year  0 Year  3

Biology  I  (cell/mol) 2,400 2.3 2.6 31% 21%Biology  II  (org/pop) 1,000 3.0 3.0 11% 13%Chemistry 5,000 2.3 2.9 35% 17%Physics 4,500 3.0 3.0 8% 9%