Hello research experts,
I'm currently working on a project that involves assessing the validity/reliability of a four-option MCQ designed to assess trainee performance as well as a training tool. Each assessment will be vocation specific. In other words, we would be collectng a few assessments from a wide variety of fields of employment.
I was thinking of starting with calculating districtor efficiency, discrimination/difficulty indices, p-values, means, etc...
However I was wondering whether there could be any other analyses (supplemental or primary) that could be used to assess a trainee training/performance review MCQ, something I may be overlooking. I feel as though I have a basic starting point, but feel as though there are other analyses I could be implementing. I also realize the sample size might be a limitation, as we could potentially be analyzing data from 3 or 30 assessments from a given industry.
Any suggestions would be greatly appreciated.
Thank you!
Hi Marko, In France, to do this type of validation we use the psychometric method. This method is described in many books. One of our reference books is "Introduction aux théories des tests en psychologie et sciences de l'éducation" (Laveault et Grégoire, 2014 - http://www.deboecksuperieur.com/ouvrage/9782804170752-introduction-aux-theories-de-tests-en-psychologie-et-sciences-de-leducation ) But I don't know if you have the equivalent in English. Fabien
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size