5984.2	Assessment
Research Question (Clear, unambiguous)	The reasoning leading to "that as the level of education increases, so does the quality of interpersonal relationships." is obscure.  The implication that the intelligentsia live in a world of outstanding personal relationships while the uneducated flounder in a sea of truncated personal ties is a bit bizarre.  As is the historical implication, that everyone must have suffered from poor personal ties in centuries past and we must now be in a relative paradise of personal connections.
Relevant Literature	The appropriate literature would seem to be about variations in the quality of people's social networks, and perhaps on the quality of marriages.  Work on education and psychology seem less relevant.
Causal Interpretations (clear, all parts defined, mechanisms, controls, plausible)	The causal discussion seems all over the place.  Certainly, we can posit that people vary in the quality of their relationships to those with whom they are closest.  Whether or not education has any effect is a legitimate question, although why we would expect its effect to be a monotonic change in the average quality of relationships is obscure.  As described, the question would seem most simply to imply that the quality of marriages is a function of spouses' education levels.  This seems dubious, but it at least would be a clearer idea.
Data (variables, sample, comparison)	The standard GSS seems a questionable data set for this issue.  A respondent's report on the quality of her or his relationships seems a poor indicator of whatever a disinterested observer might characterize as a good or poor relationship (the meaning of which is unclear here).
Research Value	Learning about some obscure possibility that education influences people's personal relationships seems insufficient justification.  Perhaps something that shows clearer policy relevance.
Overall	What causes some individuals to have better and some worse personal relationships is a worthy question.  Upscaling this question to categories of people, here defined by education level, seems to start with an undemonstrated assumption about the empirical distribution.  Combining that with unclear strategies to pursue the question leaves this project unconvincing.