Thoughts on social science research

Social science research findings seem to me to break down into several different levels of question:

1) What is going on? (Essentially, descriptive research.)
2) Why is it going on? (This level seeks to identify the causal chains that lead to particular outcomes, or sometimes, interpreted a little more loosely, what the critical variables are that differentiate different classes of states or outcomes.)
3) How is it effected? (This level seeks to identify the levers by which action can affect outcomes and what the affects of different actions are; it's the only one of the three levels that is directly useful from a policy or "social engineering" standpoint.)

The line between 2 and 3 can be subtle but can also be really important, because misinterpreting which your author is trying to do can lead to a lot of confusion and unnecessary mental critique. It's also really easy to misinterpret - partly, I think, because authors often either aren't clear about or aren't themselves aware of which level they're working on.

I don't think these are theoretical distinctions so much as they are practical ones. It's not that level 2 and 3 aren't, in fact, closely related - they are - but it matters for understanding your author which level he thinks he's working at.