Back to Appendix A: Regression Estimates and Predicted College-Going RatesBy Billie M. Sebastian, Corrie Orthober, Jennifer Bobbitt, Tom Hampton, and Suzanne King
From Listening to Kentucky High Schools
pp. 73-75, published 2002
Based on the method described in Appendix A, we used the multiple regression analysis to predict college-going rates for 233 Kentucky high schools. We then subtracted the predicted rates from the actual rates to identify schools that were underperforming or overperforming; a negative number indicates the school is underperforming while a positive number suggests that the school is overperforming.
The Kentucky Educational Professional Standards Board provided us with the academic transcripts of teachers from the 10 schools with the largest negative residuals and the 10 with the largest positive residuals. We analyzed these transcripts to identify differences in the backgrounds of these teachers. Our purpose was to determine if differences between the two groups of teachers would help explain the performance of these schools. In the sections that follow we provide summary data about the schools and the teachers.
Table B.1 compares the entire sample of high schools to the 10 worst underperforming schools and the 10 best overperforming schools. The variables of comparison are explained in Appendix A. Table B.1 shows that the average actual college-going rate for the 10 underperforming schools is 22.1 percent, which is significantly lower than their average predicted college-going rate of 36.6 percent. On the other hand, the average actual college-going rate for the 10 overperforming schools is 62.2 percent, which is significantly higher than their predicted college-going rate of 42.1 percent.
Table B.1: Summary (Data) Averages on High Schools
We collected three types of data from the teacher transcripts: grade-point averages; high schools attended; and rank.(1) We examined the transcripts of 456 teachers from the 10 worst underperforming schools and 501 teachers from the 10 best overperforming schools. However, there is virtually no difference between the two groups when comparing these variables. We believe this highlights the importance of other, perhaps intangible factors, in accounting for the differences between the underperformers and overperformers.
Table B.2: Summary Data from Teacher Transcript Analysis
To view a list of all chapters in this book, click here. To read the chapters in sequential order, please follow the arrows below.
Back to Appendix A: Regression Estimates and Predicted College-Going Rates
A Rank III teacher is someone who has obtained a college degree or its equivalent from an approved four-year institution. A Rank II teacher has accumulated at least 30 hours of graduate work, which may include a master’s degree, or the equivalent continuing education. A Rank I teacher has earned at least 30 hours of graduate work or the equivalent continuing education beyond Rank II and in a subject field approved by the Educational Professional Standards Board, or has met requirements for Rank II and also holds current certification from the National Board for Professional Teaching Standards. Return to text.