A rough draft of my analysis can be found here, although it is a little barren. I included most of what I wanted to write and most of the graphs I want to use, but there are none of the tables I want to include. I decided to analyze UIUC transfers from Illinois Community Colleges, and the graduation rate, transfer rate, and graduation plus transfer rate of Illinois community colleges. I decided to leave out the Pathways sections for now, as the data for that section was not as good. I can always include it later if I feel the paper needs more.
My plan is to retype my whole paper in Latex for the final draft, so hopefully it will look a lot nicer and more organized. If you want some more updates on my progress, I'd be happy to give them, but otherwise I don't plan on any blog posts until I retype the whole thing. As always, any suggestions would be appreciated. Thanks.
I made a copy of your document and marked it up, mainly with comments. A few corrections are there, in red. (I hope you are not color blind.)
ReplyDeletehttp://goo.gl/RhNFp2
There are two big issues/questions for me. The first is on definitions. Do students who transfer from a community college to UIUC also graduate from the Community College? Is that always true, meaning the transfer students typically enter as juniors, sometimes true, meaning sometimes they enter before that, or never true. It really would be good to understand that.
The other is on the model(s) you are estimating. It really would be good to write out the equation for each model you have - say how the variables are constructed whether raw data or log of raw data or something else - and something on why you did what you did. This needs to precede your results.
Just as a for instance, it looks like you tried to predict raw transfers as opposed to transfer per enrollment - rates of transfer. Did it occur to you to possibly try transfer rates instead? Might you do it both ways? These are the sort of questions that occurred to me reading this.
For this rough draft, I purposely did not put in my equations, because they would be hard to do in Word. But for the final draft, I will definitely show estimates/ standard errors for all of my models.
DeleteThe graduating community college then transferring is a big problem that I have not thought of before. I can't think of any way to find this out, But I will definitely list it as a flaw in the data.
I didn't think of doing transfer/enrollment, I'll play around with it and see if I find anything interesting. Thanks.
Okay - you have to consider this from the U of I's point of view and then from the student's point of view. If I understand the pathways program, those people are guaranteed admittance if they get their associates degree. There may be others who aren't in pathways. Why admit them? I'd think that would be, again, because they got their degree.
DeleteThe only time I can imagine the U of I admitting somebody who has yet to earn an Associates Degree is a student who actually got admitted to the U of I out of high school, but chose Community College (say because the tuition is less) but after a year decided that CC wasn't stimulating enough and wanted to then transfer to the U of I. That sort of thing is probably rare, but I'm sure it happens once in a while.