Analysing questionnaire data

Question

I want to test if there is a rivalry among two siblings in a family. I have 15 questions in my study and I let my 100 respondents ( distributed equally to two siblings) ranked them 1 to 15.

How should I analyse this data?

Answer 1

In the spirit of an earlier response, you might be interested in David A Kenny's webpage on dyadic analysis, and models for matched pairs (See Agresti, Categorical Data Analysis, Chapter 10, or this nice handout, by C J Anderson). There's also a lot of literature on sib-pair study design, in genetics, epidemiology, and psychometrics.

As you may know, studies on sibling rivalry also suggest that parents' attitude might play a role, but also that generally sibling relationships in early adulthood might be characterized by independent dimensions (warmth, conflict, and rivalry, according to Stocker et al., 1997). So it may be interesting for you to look at what has been done in psychometrics, especially whether your items share some similarity with previous studies or not. The very first hit on Google with siblings rivalry scale statistical analysis was a study on The Effects of Working Mothers on Sibling Rivalry which offers some clues on how to handle such data (although I still think that model for matched pairs are better than the $\chi^2$-based approach used in this study).

References

Stocker, CM, Lanthier, RP, Furman, W (1997). Sibling Relationships in Early Adulthood. Journal of Family Psychology, 11(2), 210-221.

Answer 2

I agree with David that your question needs some better explanation. Perhaps you should start by giving us a snippet of your data set up so that we can see how it's structured. Also indicate what variable links the individuals into sibling groups.

It sounds like you might want to take a look at ranked order logit (type "help rologit") if your data & analysis are something like this:

*****************************! Begin Example

*-----run this code  your Stata do-file editor
 clear
inp id ranking   option   x1 x2   male
1   4  1  1 0  0 
1   2  2  0 1  0
1   3  3  0 0  0
1   1  4  1 1  0
2   1  1  3 0  0
2   3  2  0 1  0
2   3  3  2 1  0
2   4  4  1 2  0
3   1  1  3 1  1
3   3  2  1 1  1
3   4  4  0 1  1
4   2  1  1 1  1
4   1  2  1 1  1
4   0  3  0 1  1
4   0  4  1 0  1
end

rologit ranking x1 x2 male, group(id)
****************************! End Example

- Eric
eric.a.booth@gmail.com