How to statistically test whether parents and children share similar humor styles?

Question

I have data from children and their parents on their humor styles (using the Humor Styles Questionnaire by Martin (2003)).

For the adult version: It's a 32-item measure, with 4 subscales for each humor style (Affiliative, Aggressive, Self-Defeating, and Self-Enhancing) so there's 8 items for each subscale.

For the children's version: There are 24 items, with 4 subscales so 6 items per humor style. The items for the children are on a 4-point scale and the parents are on a 7-point scale.

Research Question: I want to assess whether children share humor styles with their parents.

What kind of test would help me find out if children have the same humor styles as their parents?

References

Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J., & Weir, K. (2003). Individual differences in uses of humor and their relation to psychological well-being: Development of the Humor Styles Questionnaire. Journal Of Research In Personality, 37(1), 48-75. doi:10.1016/S0092-6566(02)00534-2

Answer 1

Assuming the data are dyadic (i.e., one child per parent), you would probably focus on correlating child scores with parent scores.

Hopefully, you have a sense of how the child and parent scales align. Hopefully, they are designed to measure equivalent constructs albeit with different items.

The fact that you've used different scales is a potential confound. Presumably, the are good reasons for this given that the humour of children and adults is expressed quite differently. But that's something, you'll have to reason through in terms of measurement and theory.

The fact that the scales have different numbers of items and response options is mostly irrelevant. As soon as you start using different items, then you've essentially got different variables.

Although not relevant to you, more commonly, data involving dyads involves using the same instruments. e.g., husbands and wive both completing the same measure of big 5 personality. In that case, you can do lots of things, like comparing descriptive statistics, ICCs, and other values. David Kenny has lots of resources in that case on dyadic data analysis: http://www.davidakenny.net/dyad.htm

Answer 2

I'm not sure about my answer but first of all, to have comparable variables, I think you need to divide children's scores to 4 and their parent's to 7. As far as I understood you want to understand if two groups (parents and children) have similar humor styles. I think you can do 2*4 mixed anova to solve this problem. Your parent and child variables can be within factor because they are paired with each other. Your sub-categories can be between factor. In the results, first, you need to look if age influences the humor score. If it did you can say that they have at least some diffirences in some categories. Then you can check the interaction. If there is no interaction between age and style, parents and their children are certainly different. If there is a interaction, you can look in which categories age doesn't differ significantly from each other. And you can say in ... category parents and children have similar humor styles and in ... they differ from each other significantly. I hope that I could express myself clearly.