Regarding the population of interest, i.e. teens from divorced parents, I would think that one of the issues that had potential bias would be whether or not the parents themselves were substance abusers. In other words, if the teenagers came from a household where by the parents used alcohol or drugs, I would think that that would increase their likelihood of also using alcohol or drugs to self medicate. Your sample would have to take into account that factor, i.e. substance abuse history of parents.

In regards to the random sampling, here is a potential bias: would you be comparing adolescence from parents who have divorced parents against adolescence from households without divorce? In other words, when adolescence from divorced homes are compared with adolescence from intact families, is there substance use or abuse rate different? The issue here is whether or not you would be using a control group to make such a comparison. It would be a lot more complicated to do so, I realize, so you would have to decide what your research question really is. Is it “are children of divorced parents more likely to use substances than children of not divorced parents?” Or is it, “are adolescents more likely to use substances if their parents divorce?”

Finally, I would have some questions about your data collection. For example, you say that you are going to look at divorce records. Are you talking about public records such as those that exist in City Hall? I’m not sure that those kind of records would include the ages of the children. You might have to find your population from contacting several divorce attorneys, and asking them if they might get permission from some of their clients with teenagers to have you contact those adolescents for the research purposes. Then again, if your study is retroactive, you say that the data could be taken from counseling sessions. How could that be arranged, when the content of sessions are completely confidential?

My understanding, from your order instructions, is that you wanted feedback about your proposal. That is what I have tried to do. Please let me know if that is not what you had in mind. If you do pursue the subject, it is very interesting. Good luck!