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: lab final exam



On Friday, November 28, 2003 4:49 PM, Aaron Titus asks:

For the first time, I'm going to give a final exam in the lab
that assess students' "lab skills". Does anyone have a final
exam they use or does anyone have good ideas for questions
and exercises. I want the lab to be "hands-on".

We use a lab final for both our first and second semesters of
algebra-based physics. Neither one is perfect but we think they've
worked reasonably well. I'd love to hear ideas for improving them.

The second semester, we give them unknowns (last semester we gave them a
circuit contained in a box with wires hanging out and an unknown optical
material) and ask them to identify them (e.g., the components in the
circuit and the index of refraction of the material).*

In comparison, for the first semester, the "lab final" assesses "lab
skills" more so than "physics". Students are assigned one "station" in
which they are asked to make a measurement (mass, density, acceleration,
etc.) using only the materials at the station. The measurement is one
they were asked to complete during the semester as part of one or more
of their labs. For example, they might be asked to measure the
acceleration of an object connected to an Atwood's machine set-up, or
they may be asked to measure the density of a cylinder by using a mass
balance and a caliper.

In any event, students are graded on three things: they are given 100
points for clarity (i.e., is it clear what they did), 100 points for
appropriateness (i.e., did they go about the measurement appropriately)
and 100 points for their estimate of uncertainty. For the last 100
points, we compare their uncertainty to our estimate. If their estimate
is less than twice ours, we give them the full 100 points.

Finally, we compare their answer to OUR measurement. If the difference
is greater than twice their uncertainty, we take off points (up 60% of
their score, depending on how far off they are).

It seems easy but our average score is only about 190 (out of 300).

*During the "lecture" exam, I typically include a "lab" question. For
example, I could have a circuit with, say, two light bulbs, one of which
is not lit. I'd then ask them to identify a possible reason why the
light bulb is not lit.

____________________________________________________
Robert Cohen; 570-422-3428; www.esu.edu/~bbq
East Stroudsburg University; E. Stroudsburg, PA 18301


-----Original Message-----
From: Forum for Physics Educators

AT