Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Seeking Diagnostic Tests rides again





Maybe the case was slightly overstated, but back of the chapter problems and
questions have been used for centuries without any proof that student
thinking about physics has improved. But within two decades of the creation
of the FCI concept inventory, we have been able to show measurable progress.

A better evaluation might be to have students create a concept map. Ranking
tasks are also good pedagogical devices as well as good evaluations.
Another alternative is to ask a question which requires a written
explanation. Free response items are extremely revealing because unlike
back of the chapter questions students can't used a memorized algorithm. Or
one can supplement the back of the chapter question by having the student
explain what they would do and why they would do it, without actually
producing a numerical solution.

John M. Clement
Houston, TX



|
| Unfortunately, it is usually impossible to diagnose student
| thinking by the usual end of chapter problems. This is why
| concept inventories have been developed. A good concept
| inventory has several questions which probe the same concept,
| and can be used for diagnosis of difficulties.
|

. . .

|
| If back of the chapter questions were generally good, then
| traditional Halliday and Resnick students would achieve hig
| FCI scores. But they don't.
|

Me thinks that you are building up a bit of staw-man. The post below
merely said that you can get some idea from end-of-chapter problems.
And it has been my experience that , indeed, one can get some idea of
misconceptions from end-of-chapter problems.

Admittedly, this isn't the same as the carefully designed concept
inventory tests; but it wasn't claimed to be. Furthermore, as I read
the post, it wasn't claimed that end-of-chapter problems alone were
going to achieve large gains in measures derived from concept inventory
testing.