This December will mark the beginning of my 24th year writing about the Advanced Placement (AP) program in high schools. I haven't had much competition. Relatively few people know or care about these college-level courses, sponsored by the College Board, that energize high schools and prepare students for higher education. In the early years I wondered if the AP teachers I followed were really as important as I thought they were.

Seventeen years passed before I saw the first major effort to ascertain the impact of AP courses and test-taking on high school students. That 1999 report, "Answers in the Tool Box" by U.S. Education Department senior researcher Clifford Adelman, was not really about Advanced Placement. But Adelman included AP among the intense academic experiences that his data indicated gave high school students a better chance of graduating from college.

So after more than two decades of underwhelming scholarly interest in this topic, I am delighted to report a surge of serious AP research. (By "surge," I mean four studies in the past year. That is not exactly a deluge, but to me it looks like a lot.)

Education Next | , a quarterly that often explores the most newsworthy school issues, is about to publish in its winter issue Andrew Mollison's detailed and nicely balanced piece on the 50th anniversary of AP. Mollison mentions the four new AP studies. I have reported on two of them in this column, and have my doubts about the third, but the fourth one, "The Relationship Between Advanced Placement and College Graduation" by Chrys Dougherty, Lynn Mellor and Shuling Jian of the National Center for Educational Accountability, is a gem and deserves close attention.

I know these academic reports with long titles and many co-authors sound like something you would only read to cure insomnia, but this paper -- let's save space and call it the Dougherty-Mellor report after the lead authors -- is too good to miss, and I think I can sum it up without putting too many people to sleep.

The National Center for Educational Accountability is in Texas, a hotbed of educational research these days. Pushed by activists like Tom Luce, a lawyer turned education guru who just got a big job in the U.S. Education Department, the Lone Star state is following the progress of individual students in a more extensive and sophisticated way than most of the other states. Luce, in his role as chairman of the center in Austin, along with Lee Thompson, deputy director of the education-oriented O'Donnell Foundation in Dallas, wrote a book "Do What Works" earlier this year that uses much of this new student data. I wrote about it in my Nov. 23 column |

Almost all the studies so far show that students who get a good score on an AP test in high school do better in college than those who get a bad score or don't take AP at all. I am interested in how those students with bad scores did in college compared to students who did not take AP at all. Many AP teachers have shown me examples of students who did poorly on the exam but did well in college in part, they think, because struggling with AP gave them a useful dose of thick-reading-list-and-long-final-exam trauma.

The Dougherty-Mellor report takes that discussion to a new level. Luce and Thompson used a study of 78,079 Texas students who entered college in 1998 to show that low income and minority students who took and FLUNKED an AP exam in high school still had higher college graduation rates than students who did not take an AP exam. Although correlation is not the same thing as causation, that result indicated that high schools were doing the right thing when they urged students to try AP even if the course might be difficult for them and lead to a low grade on the AP exam. I have said more than once that educators who see those results and still bar average students from AP classes are committing malpractice.

Dougherty-Mellor considered an alternative explanation for that result that has been suggested to me by several readers. Maybe the students who took and flunked an AP exam had higher college graduation rates than those who never took an AP exam not because taking AP helped them, but because they were different people -- more motivated, more persistent, more likely to keep going even if they flunked an AP test or struggled in their freshman year at college.

Dougherty and Mellor, whose paper can be requested by e-mailing, found no way to measure how motivated those who flunked AP tests were in comparison to those who did not bother to take the tests, or the courses. But they were able to compare those who flunked the AP test to those students who did not take it and were of similar economic status and academic achievement, and attended high schools of similar average incomes and academic achievement.

The paper is full of cautious statements, since it sometimes puts great weight on a relatively small sample of students. But the results indicate, in most cases, that students who take and fail an AP test are not much more likely to graduate than students who do not take an AP test. Among non-low-income students, there was no advantage. Low income students who took and failed an AP test were 5 percent more likely to graduate from college in five years than similar students who did not take an AP test. White students in that category were 3 percent more likely and Hispanic students, 1 percent more likely.

African American students who took and failed an AP test in high school had a much larger advantage over African American students who didn't take an AP test -- a graduation rate that was 18 percent higher. But the numbers of African American students in the study was so small that Dougherty and Mellor warned against making much of that result.

The study authors say they have not settled the argument. "We have not shown that there is no advantage to taking AP courses for students who struggle with the material," they said. Dougherty emphasized to me that motivation is a powerful force, hard to measure with his numbers and high schools at the very least should not discourage students who want to work hard in a college-level course like AP or International Baccalaureate.

Another new study, by Kristin Klopfenstein of Texas Christian University and M. Kathleen Thomas of Mississippi State University, is less favorable to AP. They looked at a sample of 28,000 Texas students and concluded that "with the lone exception of Hispanic students taking AP science, AP experience has no impact on first semester college GPA."

Klopfenstein and Thomas are economists, as are many of the most interesting researchers in education today. They are good with numbers, but I think once they spend more time inside American high schools they will define the term "AP student" less broadly. To them, anyone who has taken an AP course is an AP student. Their results show that this group does no better than non-AP students in college.

I think they would be better to focus, as Dougherty and Mellor did, just on those students who took an AP exam. During these 23 years of visiting high schools, I have found many with courses labeled AP that are not actually taught at the college level. At those schools, few AP students are encouraged to take the AP exam -- which is written and scored by outside experts -- and so there is no sure way to tell that they are getting an inferior product.

The test -- three hours of difficult questions, including 90 minutes of free-response questions that must be graded by human beings -- defines the rigor of the course. Counting students who do not take the test corrupts the sample, and Klopfenstein says she hopes one result of her study will be for colleges to no longer give extra grade point credit to students who take the AP course but avoid the AP exams. (The Klopfenstein-Thomas study, "The Link Between Advanced Placement Experience and College Success," can be found at Klopfenstein's Web site |{tilde}kklopfenstein/ , with a great bonus -- a photo of Klopfenstein's 11-month-old son Collin.)

I have my own study I would like to do. If you know of a foundation with the necessary grant money, send them this column. One of the other major new AP studies, by University of California researchers Saul Geiser and Veronica Santelices, shows no college advantage for high school students who take AP tests but do not get good grades on them. That makes sense, since they are looking only at UC students who are almost all in the top 12 percent of their high school classes and have academic talent whether or not they have taken AP. I would like somebody to do a similar student of California State University students who are less likely to be academic stars and for whom AP experience may be more of a help in getting through college.

I am not worried about high schools discouraging average students from taking AP as a consequence of these research findings. Dougherty and Klopfenstein have told me they do not want that to happen. Good grades on AP tests correlate with better performance and higher graduation rates in college. An average student has little or no chance of getting a good grade on an AP test unless the high school lets the student take an AP course. Similarly, there appears to be no indication that taking and flunking an AP test hurts you. It may give you little or no advantage in college, but at the very least, it does not appear to be a disadvantage. So why not give it a try? Any AP teacher will tell you of kids that surprised them with good grades on the test.

Maybe next year, there will be even more AP studies, and we can get closer to the truth. I have had at least three high school students tell me this year of AP studies they are doing for credit. There must be many doctoral candidates thinking of exploring this booming high school trend. The whole issue may be a bore to many of you, but it is as thrilling as reading the Redskins game results is for me, and the more I see of them, the better I will feel.