I would like to estimate the effect of one independent variable on the predicted variable in the purely observational study. On the other hand I know that there exists another independent variable, correlated with the first one, which can also affect the value of the predicted variable. Let's say that I am interested in estimating how the middle school final grad affects people earnings in their later life. I also know that both the grade and the future earnings can be affected by the income of the parents of that particular student. How can I extract and estimate the effect of the grade only, independent of the family status? I know that in the case of two groups, for example, if I divide the subjects into the "high earnings" and "low earnings" categories, I can use propensity score analysis. Is there any linear regression equivalent of that technique?