Building on a Base: tools, practices, and implications from

physics education research (PER)

S.J. PollockN.D. FinkelsteinPhysics Department

Thanks for support from: Pew/Carnegie CASTL,NSF CCLINSF STEM-TPAPS: PhysTEC

Overview

• Physics Education Research (PER)

Rapid growth, subfield of physics• A Physicist’s History: Research on student concepts (Arons, McDermott, ...)

Concept Inventories (Halloun, Hestenes , Hake, ...)

Curriculum (Washington, Maryland, Mazur, many...) Theoretical Frames (Redish, diSessa, many...)

Theoretical frames

Student concepts and engagement

Curricular reforms

Data

Classroom practice

Building on a base

structurePieces Coherence

By Authority Independent(experiment)

learning

COGNITION AND INSTRUCTION (physics), David Hammer

Novice Expert

Formulas & “plug ‘n chug”

Concepts & Problem Solving

content

think about science like a scientist

What’s our goal?

APS

In recent years, physics education research has emerged as a topic of

research within physics departments. ... The APS applauds and supports

the acceptance in physics departments of research in physics

education.

-The American Physical Society

Statement 99.2 Research in Physics Education (May 1999)

Professional recognition

• Journals (AJP, and Physical Review)

• NSF funding

• >50 institutions with PER groups

Data on student conceptions

Interviews/open questions

(e.g. Arons, McDermott, ...)

• Prior knowledge

• Basis for surveys and curriculum reform

CLASS

CURRIC

STUDENT

DATA

THEORY

A possible “tilting” development

• Force Concept Inventory (Hestenes, Wells, Swackhamer, Physics Teacher 20, (92) 141, Halloun and Hestenes)

• Multiple choice survey, (pre/post)

• Experts (especially skeptics!) =>

necessary (not sufficient) indicator of conceptual understanding.

CLASS

CURRIC

STUDENT

DATA

THEORY

Sample question

Force Concept Inventory (FCI)

R. Hake, ”…A six-thousand-student survey…” AJP 66, 64-74 (‘98).

<g> = post-pre 100-pre

traditional lecture

FCI I CLASS

CURRIC

STUDENT

DATA

THEORY

Trad’l Model of EducationInstruction via

transmissionIndividual Content (E/M)transmissionist

CLASS

CURRIC

STUDENT

DATA

THEORY

Where does this come from?

• Our classes

Force Concept Inventory (FCI)

R. Hake, ”…A six-thousand-student survey…” AJP 66, 64-74 (‘98).

<g> = post-pre 100-pre

red = trad, blue = interactive engagement

FCI II

CLASS

CURRIC

STUDENT

DATA

THEORY

PER Theoretic Background

Instruction

via transmissionIndividual Content (E/M)transmissionist

Individual

Prior knowledge

Content (E/M)Construction

constructivistbasic constructivist

J. Piaget - Swiss psychologist (1896-1980)Students: are active in the educational process

construct understanding based on prior knowledgelearn through individual development

CLASS

CURRIC

STUDENT

DATA

THEORY

Value of FCI

• Based on research

• Refocus on concepts

• Quantitative basis for comparing curricula

• Wake up call

CLASS

CURRIC

STUDENT

DATA

THEORY

Force Concept Inventory (FCI)

R. Hake, ”…A six-thousand-student survey…” AJP 66, 64-74 (‘98).

<g> = post-pre 100-pre

Fa03/Sp04Fa98

red = trad, blue = interactive engagement

FCI at CU

CLASS

CURRIC

STUDENT

DATA

THEORY

Next steps

Conceptual survey development www.flaguide.org

Attitudes/student epistemology

Research on student understanding -> guide to curricular reforms -> incorporate cognitive theories

CLASS

CURRIC

STUDENT

DATA

THEORY

Attitudes and Beliefs

VASS, MPEX, CLASS, ... (e.g. Saul, Redish, PER@C,...)

Assessing the “hidden curriculum”

Examples:Examples: ““I study physics to learn knowledge that will be I study physics to learn knowledge that will be useful in life.”useful in life.”““TTo learn physics, I only need to memorize solutions to sample problems”

CLASS

CURRIC

STUDENT

DATA

THEORY

CLASS pre/post

0

20

40

60

80

100

0 20 40 60 80 100

Unfavorable

Favorable

Overall Pre

Indep. Pre

Coher. Pre

Conc. Pre

R. App. Pre

R. Care. Pre

Math Pre

Effort Pre

Skept. Pre

Overall Post

Indep. Post

Coher. Post

Conc. Post

R. App. Post

R. Care Post

Math Post

Effort Post

Skept. Post

W. Adams 2003, replicating Redish, Steinberg, Saul AJP 66 p. 212 (‘98)

(Typical) attitude shifts

CLASS pre/post

0

20

40

60

80

100

0 20 40 60 80 100

Unfavorable

Favorable

Overall Pre

Indep. Pre

Coher. Pre

Conc. Pre

R. App. Pre

R. Care. Pre

Math Pre

Effort Pre

Skept. Pre

Overall Post

Indep. Post

Coher. Post

Conc. Post

R. App. Post

R. Care Post

Math Post

Effort Post

Skept. Post

Concepts

Reality

W. Adams 2003, replicating Redish, Steinberg, Saul AJP 66 p. 212 (‘98)

(Typical) attitude shifts

Shift (%) (“reformed” class)

-6

-8

-12

-11

-10

-7

-17

+5(All ±2%)

CLASS categories

• Real world connect...

• Personal interest........

• Sensemaking/effort...

• Conceptual................

• Math understanding...

• Problem Solving........

• Confidence................

• Nature of science.......

Engineers: -12

Phys Male: +1Phys Female: -16

CLASS

CURRIC

STUDENT

DATA

THEORY

But it’s possible to do better

Conceptual Understanding

35

45

55

65

75

g<=.25 0.25<g<=0.5 0.5<g<=0.75 0.75<g<=0.9 0.9<g<=1

Learning GainsLow learning gain <---------> high learning gain

Blue= preRed= post

Data from instructor attending (somewhat) to “hidden curriculum”)

CLASS

CURRIC

STUDENT

DATA

THEORY

Expectations/Beliefs matter

0

10

20

30

40

50

60

0-40 (N=24) 40-60 (N=74) 60-80(N=189)

80-100(N=44)

Pre-Overall Favorable Score

g<=0.3 0.3<g<=0.8 g>0.8

low <--------------------------------------> high

pre CLASS (overall)

CLASS

CURRIC

STUDENT

DATA

THEORY

Curriculum reformConcepTests (Mazur) (easy to implement) Tutorials (McDermott) (modest infrastructure)

Workshop physics (Laws) (resource intensive)

And many more - can’t do justice! Interactive Lect Demos (Thornton, Sokoloff) Problem solving (Van Heuvelen, Heller,...)

Based on empirical researchNext generation: cognitive theory as well.

CLASS

CURRIC

STUDENT

DATA

THEORY

Topic U. Wash.

no tutorial

U. Wash.

with tutorial

CU

with tutorial

Newton’s law & tension 25% 50% 55%

Newton & constraints 45% 70% 45%/75%

Force diagrams 30% 90% 95%

Newton’s III law 15% 70% 70%

Combine Newton’s laws 35% 80% 80%

ReproducibilityPrimary/secondary implementation of “Tutorials”

Rounding all results to nearest 5%

UW data from McDermott, Shaffer, Somers, Am. J. Phys. 62(1), 46-55 (94)

CLASS

CURRIC

STUDENT

DATA

THEORY

Summary

• State of PER: beyond “reflective teaching”

• Data driven

• Published/publishable results

• Reproducible across institutions

• Changing culture of departments (?!)

CLASS

CURRIC

STUDENT

DATA

THEORY

Discussion!

• Starting ideas...– What sorts of practices occur in engineering /

based on what sort of research/theoretical framing?– What assessment tools are there?– How well codified is the discipline / goals of

instruction?

The end

See: www.flaguide.orgper.colorado.eduwww2.physics.umd.edu/~redish/Book/

Impact of peer instruction

FCI scoresPhys 1110 Fa '03

0

10

20

30

40

50

60

70

0 7 13 20 27 33 40 47 53 60 67 73 80 87 93 100

Score (%)

# of students

FCI Pre

FCI Post

CU reformed course Fa 03

Traditional vs. Interactive Engagement(From Hake, see earlier ref, AJP 66, 64-74 (‘98)

%gain vs %pretest

Correlating rest of course score to tut hw (Sp04: N=513, r=.65)

0

10

20

30

40

50

60

70

80

0 20 40 60 80 100Tutorial HW score

Remaining grade

(85 max)

g known (N=383, r=.58)g unknown (N=130, r=.65)

Impact of tutorials