Hello
What do you want to test ? The fact that people choose 3 options ? Your problem seems to be a simple probabilistic exercise, that is the probability to select 3 terms into 10 possibles ? In that case, the population has 120 possible outputs. So any choice of three terms in 10 has an equal probability of 1/120.
Vincent
The question actually was what studing style do students prefer. They could select three options (i.e. books, www, libraries, studing groups, chat rooms, leacture notes etc).
What's the null hypothesis you want to test? If it is: "the students do not prefer any possible triple" than you can make a chi-squared test (again the uniform distribution of the 120 possibilities). If you want to know how options are preferred by the students, it calls for a multinomial logit model.
Indeed Jochen I am interesting in investigating how options are preferred. Thanks!
Configural Frequency Analysis (CFA) was designed for this kind of problem:
"Configural Frequency Analysis (CFA; Lienert, 1968, 1971a) allows researchers to identify those patterns of categories that were observed more often or less often than expected based on chance...If d variables are under study, each configuration consists of d elements. The configurations differ from each other in at least one and maximally in all d elements."
(von Eye's Configural Frequency Analysis: Methods, Models, and Applications).\
I think this is just the binomial test
http://en.wikipedia.org/wiki/Binomial_test
Clopper, C. J. & Pearson, E. S. (1934). The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika, 26, 404–413.
and some illustrations (but in German language)
http://de.wikipedia.org/wiki/Binomialtest
I don't believe binomial test is the appropriate one. You should consider multinomial logistic regression, since the answers are interrelated
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics