There is no solid consensus on the inputs that go into educational achievement or attainment. The list of variables is vast and varied.
One of the most common applications of this is attempting to calculate the returns to education - i.e. the average increase in income that results from one additional year of schooling.
I myself have performed research on this topic in grad school, and a good reference is:
In my research there were many variables that were statistically significant:
- marital status
- union membership
- year at time of measurement
- number of siblings
- religious beliefs
- opinion on sex ed in schools
- opinion on national educational spending
- opinion on education as catalyst for racial/class divides
Most researchers in this area agree that parental involvement is one of the main determining factors for a child's educational outcome. The difficulty is in quantifying this and other subjective data such as teacher quality and peer effects. This is one of the main issues in the field now - figuring out how to statistically measure non-numerical variables. The opinion variables I mentioned above are from national social surveys and are an attempt to quantify these subjective things. Specifically, my data came from the General Social Survey (GSS) available here: http://www3.norc.org/GSS+Website/