2011 AAPT Summer Meeting Omaha Nebraska Making Homework

2011 AAPT Summer Meeting Omaha, Nebraska Making Homework Easier to Do Than to Copy Gerd Kortemeyer Michigan State University

Copying Homework Too Good? Do all work independently Continuum Good? Gray area Productive collaboration, give and take Bad. Collaboration, mostly one-way Ugly! Blind copying, no understanding

Copying Homework • Why do students copy homework? § § § They don’t understand the material? They don’t have enough time? They don’t see the value of doing it? They are not aware of the consequences? It’s considered an acceptable form of “cheating”? § It’s so easy to do!

Why do students copy homework? • They don’t understand the material? § Homework may not address student difficulties • It should clearly be formative assessment • It should be helping the students learn, not test if they understood the material § That’s what exams are for § Oftentimes problems are way too difficult • We mostly should refrain from “cool” physics • We should refrain from homework that just shows off how smart we are • We should not assign homework that requires a “trick”

Why do students copy homework? • They don’t have enough time? § Our course is not their only course § Students make surprisingly sophisticated choices how to allocate their time budget § Don’t assign too much!

Why do students copy homework? • They don’t see the value of doing it? § Why should the students pay attention to something that you ignore? • Don’t just assign “Problems 3 through 16” • Discuss homework difficulties in lecture § Occasionally put variations of homework on exams

Why do students copy homework? • They are not aware of the consequences? § Homework is seen as a chore • rather than a way to succeed in the course • they will seek out the most “efficient” method of getting it done § Very hard to refute • students don’t believe us that it’s good for them • useful data by Bauer et al. and Pritchard et al. , which can be shared with the students (later in this session)

Why do students copy homework? • It’s considered an acceptable form of “cheating”? § “Everybody does it” § Well, whom are they cheating? • They are cheating themselves § Usually, the vast majority of the grade is based on exams § Should we really care? • Go ahead and shoot yourself in the foot? § But: demoralization of the whole course can happen quickly • Good students who invest a lot of time should not feel betrayed

Why do students copy homework? • It’s so easy to do § Remainder of this talk: making it harder § “Making homework easier to do than to copy” • while fostering constructive collaboration § Key: the right types of randomizing problems

The right types of randomizing problems • Problem types: § § § Multiple choice Numerical Graphical Symbolic etc. • Randomization: § Different students get different versions of the same problem § The same student always gets the same version of the problem

The right types of randomizing problems Numerical Multiple choice

Randomization Spectrum: low to high energy No Randomization Different order of options in multiple choice Different numbers in numerical problems Different options Different images, graphs, formulas Different scenarios with similar physics Different scenarios with different physics Completely different problems

Randomization Do all work independently Too Good? Different order of options in multiple choice Good? Gray area Bad? No Randomization Productive collaboration, give and take Different numbers in numerical problems Collaboration, mostly one-way Different images, graphs, formulas Different options Different scenarios with similar physics Ugly! Blind copying, no understanding Different scenarios with different physics Completely different problems

Randomization? • Next section: Show-and-Tell • Random examples of randomization in random order

Different Numbers • Simple numerical response

Different Formulas • Different formulas, same concept

Different Formulas • Different formulas, infinitely many correct answers

Different Questions Higher/lower normal/frictional force

Different Scenarios, Same Answer Choices

Different Questions Lifting/lowering, speeding up/slowing down, different numbers

$Different Scenario Construction • Carefully construct the problem • Egyptian fraction: 1/n 1+1/n 2+1/n$

Different Scenario Construction • Carefully construct the problem • Egyptian fraction: 1/n 1+1/n 2+1/n 3+… • Want limited number of small denominators

Different Graphs

Different Graphs Something big changed here

Different Graphs and Scenarios Need to change this, too Same options, but different one is correct

Different Scenarios, Graphical Input Diff ere nt s tori es • Graphical input • Infinitely many correct answers

Different Line Graphics

Using Learner Answer

“Hidden” Differences Two ways how the paper could slide off the fridge: • Magnet slides off paper • Paper and magnet slide off fridge Depending on values, one or the other decides.

“Hidden” Differences Most course discussion ever! Panic.

Collaboration Too Good? Do all work independently Good? Productive collaboration, give and take Gray area Bad. Collaboration, mostly one-way Ugly! Blind copying, no understanding

Online Discussions Encouraged, since all students have different versions. Peer-Instruction.

Online Discussions • Classifying homework discussion contributions • Green: desirable collaboration Yellow: unproductive Red: undesirable (“copying”)

Online Discussions • More successful students exhibit less “copying” during discussions

Online Discussions • More successful Post. FCI=5. 486+0. 922 • Pre. FCI+0. 24 • %Physics students exhibit less Post. FCI=7. 606+0. 857 • Pre. FCI-0. 042 • %Solution • Meaning what? “copying” • Students who contribute 100% solution-oriented during discussions on the average achieve 4. 2 points (out of 30) less on the post-test, controlling for pre-test

Online Discussions • Reducing copying: don’t use simple problem types • Multiple Choice: § highest percentage of solution-oriented discussions (“that one is right”) § least number of physics discussions • Ranking and click-on-image problems: § Physics discussions highest • Problems with representation-translation (reading a graph, etc): § slightly less procedural discussions § more negative emotional discussion (complaints)

Online Discussions • Reducing copying: mid-range difficulty • Above mid-range: more pain, no (significant) gain

Helprooms • Staffed • with Learning Assistants in the evenings Collaborative learning space, peer instruction

Homework Do all work independently Too Good? Different order of options in multiple choice Good? Gray area Bad. No Randomization Productive collaboration, give and take Different numbers in numerical problems Collaboration, mostly one-way Different images, graphs, formulas Different options Different scenarios with similar physics Ugly! Blind copying, no understanding Different scenarios with different physics Completely different problems

Exams Do all work independently Good! No Randomization Different order of options in multiple choice Ugly! No gray area Productive collaboration, give and take Collaboration, mostly one-way Different numbers in numerical problems Different options Remain images, graphs, fair Different formulas Different scenarios with similar physics Blind copying, no understanding Different scenarios with different physics Completely different problems

Exams • Problems can also be rendered for bubble sheets • Every student gets a different version

Exams

Learning Success Do all work independently Gray area No Randomization Productive collaboration, give and take Collaboration, mostly one-way Blind copying, no understanding Success? Completely different problems

Learning Success • Intro Physics for Scientists and Engineers • Grades in years before and after online homework

Learning Success Mostly helps students who are on the brink of failing the course. Fail

How can I write these complicated problems? 450000 LON-CAPA Shared Resource Pool, Summer 2011 400000 Number 350000 300000 Total Problems Pages Shared content repository with almost 450, 000 resources Images 250000 Libraries 200000 Assemblies 150000 Movies/Sound 100000 Animations/Simula tions 50000 0 2002 2004 2006 Year 2008 Almost 200, 000 online homework problems 2010

Thank You Gerd Kortemeyer Lyman Briggs College and Department of Physics and Astronomy Michigan State University korte@lite. msu. edu http: //www. lon-capa. org/