# How are svymean, svyby, etc useful to understand a causal relationship?

#### blattla

##### New Member
I am sitting on a data project and got totally stuck on following question:

Justify which population quantities of the variables included in your project could be of interest (e.g. svymean(), svytotal(), svyby(), svyciprop()) How could they provide you with a better understanding about the variables included in your causal relationship? (at least one per variable, 2-3 sentences per quantity)

My assumed relationship (based on respective dataset) is that people who volunteer are satisfied with life. I really don't understand how these quantities could improve my understanding other than that I get to know the mean value or population total. Especially since I do not even have to calculate the values at this point (that's the following task). The required 2-3 sentences seems to ask for more. Can anyone provide food for thought?

#### AngleWyrm

##### Active Member
Another approach is to measure the proportions within sets, which doesn't rely on proving or asserting causality. Consider a set of apples, where some of them are rotten, and some apples are in wooden baskets.

When the proportion of rotten apples in wooden baskets is greater than the proportion of rotten apples in the population, then the category Apples in Wooden Baskets becomes a predictor of membership in the category Rotten Apples greater than random chance.