Mail Time: Should I give a pretest?

This is Michael Gray. I was the new teacher who attended your AP Physics workshop in Kennesaw last week. I am preparing for August and I have a few questions:

Hi, Michael... great to hear from you.  Yeah, it's crazy how youall in Georgia start so danged early.

I am teaching a second year class. Would you recommend giving them some sort of prior knowledge test? And if so, how would you use it?

First, I'll answer this for one of your SPECIFIC classes -- you are picking up a second-year course taught by your competent colleague who left the school on good terms.  There's no need for a pretest.  You have much better uses for your class time. 

If you wonder what was covered last year, ask the departing teacher.  If you wonder how much the class remembers or retains, then a combination of a deep conversation with the departing teacher and observation over the course of the course should answer that question just fine. 

Try starting the year with a topic completely new to them.  After a week or so of that, you should have some idea of how much you need to discuss the things that were covered last year... and you won't have inadvertantly made bad assumptions about what they retained.  The combination of conversation and observation will be more accurate and less classtime-consuming than a pretest.

BUT FOR THE TYPICAL INTRODUCTORY PHYSICS COURSE:  You've put your finger on the crux of the question by asking how you would use the pretest.  I once had a long argument with the gentleman charged with evaluating my school's science department.  He contended that I'd do a better job if I were to give a pre-test.  I asked, "To what end?  For what purpose?  How would the results change or inform my teaching?"

"Then you'd know where the students are, and can tailor the course to their needs," he said.

I had three responses, two of which are relevant to your purposes, Michael:

1.  In a large sense it doesn't MATTER where they are, especially mathematically.  I only need basic algebra I level skills.  I don't care what they already know about, say, Newton's Laws, because much of what they know will be misconceptions born of misinformation.  In a first year course, I'm teaching everything from scratch; for a second year course, I'd strongly recommend the "conversation" approach above rather than a  pretest.

2.  The students will be at many DIFFERENT places, both in terms of mathematics, and in terms of prior knowledge of physics concepts.  Sure, maybe 1/3 of your class is mathematically astute, or 1/4 of the class had a really good 8th grade physical science teacher who taught them properly about Newton's Third Law.  Big deal.  You still have to find a way to make everyone successful and interested, whether it's 5 or 15 students at the top end of the class. 

3.  I already know where my students are, or at least I can figure it out very, very quickly.  I've been at the same school for 11 years now.  I'm getting similar classes every year.  With each new class, I can tell with reasonable precision after two problem sets who is smart, dumb, lazy, clever, or whatever.  How?  By talking to them, reading their eyes during class, reading their problem sets, and if I'm totally confused, talking to teachers who had them last year. 

Moral: I don't need no stinkin' pretest.  And neither do you.

GCJ