Statistical procedures for binomial data

I have 2 questions and would be grateful for any advice I can get about
which statistical procedures would be appropriate:

1) I am administering a survey to students to see what teacher
behaviors they consider important for good student-teacher
relationships. I also want to administer the same survey (rephrased) to
teachers to see which behaviors they think students would rate as
important. The questionnaire will be in the form of agree/disagree
items. What is the best way to analyze this data? Would it be possible
to treat student and teacher responses as dependent variables, coded 1
and 0, and run a logistic regression analysis? Would a generalized
linear model be appropriate?

2) I am also going to give teachers their own survey asking what
student behaviors they consider important for good student-teacher
relationships. I would like to combine student and teacher results and
run a factor analysis. However, as I am working with dichotomous
categories, what would the appropriate statistical procedure be? Could
I use structural equation modeling? What about multi-way frequency
analysis?

Thanks in advance.

Julia

.

Relevant Pages
Loading