Dyadic analysis - what are the key assumptions

[Maggie Yu]

Hi all,

We want to perform dyadic analysis for a cross-sectional dataset of about 800 families with father, mother and children.

A wide range of information was collected, e.g., mental health, physical health, relationship quality. We would like to study how father's trauma experience relate to mother and children's mental health, using dyadic analysis.

I want to know whether I can perform dyadic analysis using the current dataset, but I don't know what criteria I should check against. For example, sample size, do the data (e.g., depression measure) has to be collected at the same time from all family members?



Any insight is highly appreciated!

Maggie

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Also, assume we're not from your area. I'm not sure exactly what you mean by dyadic analysis. What data structure and statistical procedure would you use? Without such information, we can't really help you.

If your dv is continuous, I could see using health status of family members as the dependent variable and father's trauma as an independent variable in a regression. This would probably need clustered errors by family. If the dv is ordinal or binary, the appropriate logit or probit (or ologit or oprobit) might be appropriate. You might even be able to treat the family as a panel and do fixed or random effects.