AEA365 | A Tip-a-Day by and for Evaluators

TAG | datasets

My name is Brad Coverdale and I am a doctoral student at the University of Maryland, College Park. I am interested in researching post-secondary access as well as success initiatives for students, particularly first-generation and/or low income. One of these initiatives that was very dear to me was Upward Bound. As such, I conducted a program evaluation for my Master’s thesis using data from The National Educational Longitudinal Survey of 1988-2000 (NELS 88:2000).

Rad Resource: Because NELS 88:2000 is a longitudinal study, it met my data needs perfectly. This survey started with a cohort of 8th graders in 1988 and attempted to track their academic pursuits through 2000. By asking the students many questions including whether or not the students participated in pre-college programs like Gear Up and Upward Bound, I was able to create a treatment group and comparison group by matching similar characteristics through propensity score matching. This dataset has also been useful for analyzing psychological responses and educational objectives, finding the highest predictors for particular subjects, among other research questions. Best of all, the dataset is FREE to use.  All you have to do is send an email to Peggy Quinn, the Publication Disseminator (peggy.quinn@ed.gov) with your request for an unrestricted copy of the Data and the electronic codebook.  NCES is in the process of putting together an online application for analysis but for now you can just use Data Analysis System, a product developed for NELS analysis, if you are familiar with the program by going to this link http://nces.ed.gov/dasol/ and selecting the NELS 88/2000 data.

Hot Tip: Remember to use either the panel weights if you are tracking students over time or cross-section weights if you are only interested in a particular study (1988, 1990, 1992, or 2000). Also, be wary as to what students are included as well as excluded from your analysis. Data from students that drop out of school or are removed from the study are not included in the overall results. You may want to consider appending them specifically to your data source.

Want to learn more from Brad? He’ll be presenting as part of the Evaluation 2010 Conference Program, November 10-13 in San Antonio, Texas.

· · · ·

Archives

To top