Friday, February 6, 2009
Interaction Assignment
Using the dummy variable indicating that the child lives in a non-two-parent family (SINGPAR, 1=single-parent or other adult guardian family; 0=two-parent family) and the continuous SES measure (WKSESL), create a SINGPAR*SES interaction term. Employ multiple regression to explore whether the relationship between SES and entering literacy ability (C1RSCALE) depends on family structure (and whether the relationship between family structure and entering literacy ability depends on SES). Create an appropriate regression table.
Subscribe to:
Post Comments (Atom)
2 comments:
Hi - One question that I thought might be useful for others:
I'm a little hung up on how to talk about the second and third models in regards to the main effects....i don't know how to speak about the single parent x1 or the SES x2 main effects on their own.
And my response:
So, I think generally, it doesn't make much sense to talk about the main effect independently of the interaction effect, so you can just talk about the SES effect and the single parent effect as a whole, without talking about each coefficient -- interpreting the main effect for when the other variable=0 doesn't always make substantive sense. (e.g. if it was IQ or something similar).
But, in this case, looking at the main effect on its own does make some sense. you could think of the main effect for singpar in the model that includes the interaction effect as the difference between predicted scores for children of singpar vs non-singpar for families with SES=0 (which is average SES, because you've z-scored it...).
Hi -
Another comment on some similar questions I've been getting...
In general, I think it is difficult to talk about the interaction effect on its own. You can think about the interaction effect as the difference in the slope for your relationship of interest. But then, it seems helpful to me to interpret the interaction effect and the main effects together. So, I think the most useful way to figure this out is to make the Mr. Potato-heads that Doug always mentions. After you talk about/quantify the difference in slopes, maybe add in some sentences about *predicted* scores to illustrate your point.
(Keep in mind the equation from the regression, and then just plug in some values...)
Post a Comment