---- Original message ---- >Date: Sun, 29 Apr 2007 20:32:34 -0400 >From: "Pollak, Edward " <[EMAIL PROTECTED]>
> 2) Your findings on 2 vs,. 3 distracters is > counterintuitive and conflicts with my experience. > Did you delete the 4th distracter at random or did > you make a point of deleting the most > difficult/confusing distracter? IMO, removing the > right distracter is more important than removing a > distracter at random. I deleted the most obvious one to delete--the one students had least chosen in the past--when trying to come up with "enough" distractors to satisfy a criterion I find that people tend to come up with one or two really really good ones that challenge the student to consider them but beyond that it's very hard to come up with good additional distractors. Students can readily dismiss one of them out of hand. So for the first year I just looked at the items and nad noted all of the responses and noted which ones the students didn't pick at all. It was a waste of my time to have come up with it; it was a waste of student time to read it and immediately cross it out. I kept the ones that students most often 'fell' for. In addition I did item analyses to see which items best discriminated between the 'better' students and the ones who didn't do so well. I then used those item the next year with two versions of the test randomly distributed in all sections of intro that I taught; one version had 4 answers and one had 3. I did not tell students until the end of the semester (at which time I explained what I had been doing and asked for consent and used the data from the students who gave consent--all expect 3) what had been going on other than that there were two versions--something we do routinely in larger sections such as intro, to minimize cheating. Sometimes I am frustrated that space can get very cramped and it's nearly impossible not have one's eyes wander during an exam right to another desk and test.... Anyway, this made more sense to me than the random rejection because my purpose was two fold: to save time writing options that aren't options at all; and allow me to sample more information because it takes less time to read each item and process it. I find that I can add about 10 items to a 55-minute test. > > 3) As for your statement that "most students study," > I agree. I disagree in that I believe they don't > study nearly enough. Well, true, I guess it's like my statement in #1 relative to your statement (that I deleted) about learning names... Annette Annette Kujawski Taylor, Ph.D. Professor of Psychology University of San Diego 5998 Alcala Park San Diego, CA 92110 619-260-4006 [EMAIL PROTECTED] --- To make changes to your subscription go to: http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english
