Pre/Post Multiple Choice Quiz to Assess Learning from an Educational Program

adewan

New member
Joined
Aug 30, 2016
Messages
1
I have been given a data set that I must analyze and I'm having some trouble. An educational panel was held for a group of students. Before the panel, students were given a 14 question test (mainly multiple choice, a couple T/F, and one select all that apply). After the panel students were given the same exact test again. Now my original analysis plan was turned down because I was not utilizing all of the information. I recommended calculating a grade for each student (#correct/14) for both the pre and post-test and then taking the difference between the two. If the difference was normally distributed I would use a paired t-test and if not I would use the wilcoxon signed-rank test. In addition to this, I would look at each question individually. I would do this by testing marginal homogeneity (either with McNemar’s test for T/F or a more general test for marginal homogeneity in the case of multiple choice). I don’t think testing for marginal homogeneity is what they are looking for, but I am pretty sure that I am supposed to look at each question individually. I have been reading for days on end and every study I can find that is similar to this they use the same methodology as I have. Any recommendations on how to approach this dataset differently would be very much appreciated. The topic is cultural awareness in medicine and the structure of one of the multiple choice questions would be like **Whichgroup of people are least likely to have a normal primary care physician? A)African Americans B)Latinos C)American Indians D) White ….etc. Thank you so much!
 
Top