The Good: My chair got back to me with comments on the results chapter, and overall felt that the analysis itself “looks fine” with no red flags or other signs that I messed something up. Per his email, I am still on target for an April defense.
The Bad: Despite my clearly telling him that most of my work time was on weekends, he got it back to me at 8:30pm on the Monday of a long weekend. So I took a vacation day today (my least busy work day) in order to work on it. If it’s going to take him nearly a month to respond, I want to make sure he has all the time that he needs by getting my parts done as quickly as possible.
The Ugly: While the numbers appear fine, he has some issues with the organization. Rightly so to some extent; it wasn’t until I started to look at the data that I realized what a big impact the missing data would have. (Some of my variables have as much as 35% missing data, and logistic regression by default just drops the entire record. Additionally the group who is missing is different from the group that has the variables completed, suggesting that I can’t generalize from the subset that has complete data to the broader population.)
So, now I go back through it, looking at his comments and hoping I can fix this quickly enough to not have to delay graduation.