New Studies Say AP Works

4. Extra grade points for AP

Geiser shares with me, and most of the AP teachers I know, the view that AP students should be encouraged to take the AP exams. This is sometimes difficult because the seniors who often take AP have their exams scheduled in May, after they have gotten into college, after the weather has turned warm, and just as the season of prom/senior-cut-day/senior-prank-day/senioritis extremis reaches its peak.

The Texas researchers, like Geiser and others who have looked at this, have convincing evidence that it is working hard to understand the material and getting good grades on the AP exams that correlates with college success, not just taking the course. Geiser sees this as an education policy issue of great importance:

"Though the authors of the Texas studies interpret their findings as showing that the rapid expansion of the AP program has not harmed program quality, some of the Texas data would seem to suggest a different conclusion. Among the tables appended at the end of the larger statewide study, data show that the number of students in the 'AP Course Only' category -- students who enrolled in AP courses but failed to take the AP exams -- increased both overall and in each AP subject area studied, without exception, over the four years studied.

"College Board and AP officials will be understandably thrilled, as they should be, with the results of the new Texas studies and the association between AP exam scores and college outcomes that both studies demonstrate. But those officials should also emphasize to their constituents in U.S. schools and colleges what the studies do NOT show. The new studies are consistent with a growing body of evidence that, while mastering the material taught in AP classes and performing well on the AP exams is correlated with later success in college, mere enrollment in AP classes is not. The widespread practice of 'weighting' students' high-school GPAs simply for taking AP courses is not justified by the research evidence and has had perverse consequences for both high schools and colleges."

5. AP vs. Dual Enrollment

Dual enrollment -- a common term for courses taken at local colleges, or conducted by local colleges, for high school students -- takes a major hit in the Keng-Dodd study. The researchers say students who took dual enrollment courses in high school did not do as well in college as those who took AP courses.

This is a very important issue, worthy of another column as soon as dual enrollment's defenders get a chance to read the full report. I have been talking about the relative rigor of AP, IB and dual enrollment with many educators who think Newsweek's America's Best High Schools list should count dual enrollment courses, as it now counts AP and IB exams. The Keng-Dodd data convinces me Newsweek needs a lot more information before it takes that step.

Former U.S. Education Department researcher Cliff Adelman, the guru of college completion data, had these thoughts:

"I am not overly surprised that dual-enrollment courses don't have the same impact as AP. I'll put good money on the table that dual-enrollment courses are capturing a somewhat different population. What would be interesting would be to compare the education histories of students who took dual-enrollment course X at a local community college while in high school with those who entered the community college and took it there, and then to divide the latter group into those who transferred to the 4-year sector and those who didn't."

6. AP and students who speak Spanish at home

I occasionally get e-mails from readers who think it is unfair that Spanish-speaking students are allowed to take the AP Spanish exam designed for English-speaking students struggling to master that language in high school. My standard answer is that being bilingual, as these students are, is a valuable academic skill and they should get some credit for it, just as I would not begrudge the 5 on AP biology credited to a girl whose mother the DNA expert has been letting her help in the lab since she was five.

But there is more interesting data in the Texas AP studies that touches on this, since there were plenty of Spanish speaking AP students in their large sample. Many students -- almost 2,000 in one cohort -- took the AP exam but not the AP course, extremely rare in AP courses other than Spanish.

Chrys Dougherty, director of research at the National Center for Educational Accountability in Austin, noticed something else. In the larger of the two studies, "the correlation between AP exam grades and college outcomes seems to go away for students with lower prior SAT scores. This may well be due to the large number of native Spanish speakers in Texas who earn credit on the AP Spanish exam, but who may not be well prepared in or take AP exams in other subjects, and who thus would populate the statewide group of high AP exam scorers with low SAT scores."

7. What does it all mean? It will take some time to figure that out. There is too much data here to absorb all at once, and we are likely to see an acceleration of research on AP and IB, given their growing importance, that is going to make it hard for even us devotees to keep up.

One of my favorite fellow AP addicts is Hawkins, the senior study director I quoted in my news article yesterday. He is an educational activist who has made a study of AP results among minority students. His view on these numbers has strongly influenced my thinking. He believes that it is not so much what is learned in AP, but the act of struggling with a difficult course that adds the most value.

"So taking AP courses and exams is highly associated with college outcomes down the road," he said, summing up this research. "But I think this happens not just because the kids who take these courses learn stuff and master knowledge (college level stuff), but because these kids also in the process become better learners, students, scholars. They end up mastering the importance of working hard, clearly something that will serve them well in college and later in life."

That seems right to me, but I await the next deluge of data to see if Hawkins can be proved right.