New Ideas...but are they too narrow?

So, after reading Richard Haswell's breif synopsis of Post-Secondary Entry Writing Placement I now have a better idea of where I want to go next. He has quite a few sources in his bibliographic information that I am currently trying to find, since he only mentions them briefly in his paper.

The ones I am most interested in finding are:

• Matzen and Hoyt's 2004 research on holistically scored essays from the Journal of Developmental Education (unable to find on the shelf)
• Royer and Gilles 2003 Directed self-placement: principles and practices
• Blakesley's Directed self-placement from WPA
• Willard-Traub's Portfolio assessment at U of M in Assessing Writing (possibly available through ILL)

I still want to find a way to see if students "accurately" place themselves in the appropriate course through the guided self-placement "quiz" that EMU plans to administer to incoming freshman for the Fall 2005 school year during FastTrack. The only way I can think of to measure this "accuracy" (and who knows if it would be statistically significant...I may have to rack my brain for the SPSS knowledge I retained from undergrad.) is to do a somewhat logitudinal study.

Asking students for permission to follow them throughout the placement process, and their first semester of freshman comp. at EMU. I could request to see the results of the placement "quiz" and follow the students who take the corresponding course. At the end of the course they have chosen...I will seek permission to see the final grades of these same students. If the students pass the course that they placed themselves into, then the "quiz" worked for those students. I want to use a C or higher cut-off (or whatever is considered passing for university standards).

The only problem I foresee is the issue of ethics (I would have to use students not enrolled in my course). Would it be a problem if students knew that I would be looking at their grades at the end of the semester? Would that itself skew my results? Is it necessary to have a data pool that is statistically significant? I guess I can't worry too much about that right now.

The other idea I have is to look at the success rate of students who were placed based on ACT scores prior to the implementation of the guided placement program at EMU. This would determine if there is actually a difference between the two methods. Unfortunately, this may be too big.