Surviving a midlife crisis: advanced placement turns fifty.

In 1956, 1,220 college-bound juniors and seniors in 104 American high schools took the first Advanced Placement (AP) exams conducted by the Educational Testing Service for the College Board.

Only 11 subject areas were offered at the time: American history, biology, chemistry, English, French, German, Latin IV (fourth-year Vergil), Latin V (prose, comedy, lyric), mathematics, physics, and Spanish. So few students took the earliest exams that the first AP European history exam, which debuted in 1957, was scored on the second floor of a firehouse near the ETS headquarters in Princeton, New Jersey. "There were 59 exams and four of us examiners," recalled Henry Winkler, then a professor of history at Rutgers and later president of the University of Cincinnati. Each answer booklet held responses to one long essay question and the 25 short-response items on the three-hour test. They finished scoring the exams the same day.

"The AP program was unabashedly elitist when it began," says AP chronicler Eric Rothschild, who became a teacher of AP history teachers after retiring from 38 years of teaching social studies at Scarsdale (New York) High School. "Those taking the exams in the early years were largely male, largely students from private prep schools and elite public high schools, and probably mostly Protestant."

And no wonder. The idea for the program emerged from elite colleges, prep schools, and high schools in two collaborations. One was initiated by Kenyon College and the other by Harvard, Princeton, and Yale as a way of accelerating and fortifying the education of the nation's future leaders in anticipation of cold war national-security demands. "Shortening the conventional process for some students by even one year," concluded one of the group's final reports, "if it could be done with no significant educational loss, would add thousands of professional 'man years' of service to the nation's communities."

A half century later, the cold war over, the AP program is still with us, still growing (see Figure 1). Advanced Placement courses and exams now cover 35 subjects including art history, economics (macro and micro), and studio art (2-D and 3-D design), and they are taken by more than a million students (including many blacks, Hispanics, Asians, and more girls than boys) from 11,000 public and 4,000 other schools (including more than 800 in other countries and territories).

The essentials remain the same: high-school students take college-level courses, taught by teachers in their schools, and then take exams designed by the College Board to show that they have mastered the subject at college-level proficiency. Good exam results can be parlayed into college credits and permissions to advance directly to higher-level college courses. And despite the vast expansion of the number of students taking AP courses and exams, the average number of tests taken by students during their high-school careers has barely changed: 1.7 AP tests per examinee in 1956 and 1.8 per examinee in 2005.

Aside from its explosive growth, the most visible change during the program's first 50 years has been the transformation of the demographic profile of the typical AP examinee. But has the program actually increased student college readiness? Do AP courses boost student achievement? Many of the national barometers of high-school student achievement and college readiness show a flat line on proficiency scales over the past 50 years. This is true even of the top 10 percent of students taking the National Assessment of Educational Progress (NAEP) exams, many of whom, presumably, take AP exams. Their scores have not improved in 30 years. And findings from research about the impact of AP course taking have been ambiguous. The College Board is also facing an outburst of quality-control problems with some of its more popular AP courses. If the AP program is so good, why hasn't it made more inroads into high-school academic life? It seems that the much-admired program is suffering the slings and arrows, creaks and pains, of middle age.

The Advanced Placement Boom

On the surface, the AP program looks better than ever. By May of 2005, participation in AP exams was at record levels, with 1,221,016 high-school students taking one or more AP test. If recent patterns held, 83 percent of the test-takers were juniors or seniors; the rest, mostly sophomores. And despite the popular media image of an AP student as an overstressed overachiever from a high school in an upscale suburb, examinees now come from a broad swatch of U.S. high schools.

A hint of the enterprise's current problems, however, is in the fact that 300,000 additional American students were enrolled in AP courses last year but didn't take an AP exam in May, potentially embarrassing college admissions officers who had recruited some of those students because of transcripts loaded with AP courses. AP teachers say some students who skip the test feel unprepared, while others are seniors who have already been admitted to college and no longer need to use AP scores to gild their credentials.

Still, the AP program has become so widely accepted that one-fourth of the public-high-school graduates in the class of 2004 had taken an AP course. One-fifth had taken at least one AP exam, and just over 13 percent had earned a "passing" AP score of 5 (extremely well qualified), 4 (well qualified), or 3 (qualified). (A score of 5 is considered the equivalent of an A+ in a college course, a 4 is a college B, and a 3 is considered to be a C in an introductory college class.)

"This is way more than ever took rigorous, college-level courses" in high school, says Gaston Caperton, president of the College Board, but "13 percent isn't enough. Fifty-seven percent of high-school graduates enter college." In addition, exultation at the record numbers of AP participants has to be tempered somewhat by unsettling reports that nearly a third of public-high-school students drop out or fall behind before their class graduates. The bottom line: those passing one or more AP tests probably comprise less than 10 percent of the group that entered high school together.

Moreover, ACT Inc., which began measuring college readiness as the American College Testing Program in the 1950s, reports that among the college aspirants who took its admission exams last year, only 21 percent of the graduating seniors attained scores high enough in all four subjects--English, reading, math, and science--to indicate that they wouldn't need to take a no-credit remedial course when they entered college. Colleges are now requiring nearly 3 out of every 10 first-time college students to take at least one remedial course in reading, writing, or math in their freshman year. Apparently, the AP program hasn't so much raised the level of the curriculum in the American high school as it has created an escape hatch that lets a small number of ambitious students get out of the low-demand environment for a few hours.

Diversity on the Move

At this year's annual AP national conference in Houston, attended by more than 2,500 teachers, professors, and administrators, College Board officials stressed that their efforts to extend the benefits of college-level courses to previously underserved high-school students are beginning to pay off. In 1955, the typical AP examinee was a boy; today, by a margin of 56 percent to 44 percent, she is a girl. While the number of white students taking AP exams grew in recent years, it grew more slowly than the numbers in other groups. By last year, only 64.5 percent of the test-takers identified themselves as non-Latino whites.

Caperton points to Florida as "an example of what a statewide effort will do" to increase AP participation, especially in rural, small, or heavily minority schools. Like many states, Florida subsidizes examination fees. But Florida also gives faculty bonuses and extra AP funding to a public high school every time one of its students scores a 3 or better on an AP exam. The money can be used for AP materials or for professional development in AP content and techniques for teachers. Florida was also one of the first states to form a virtual school, offering free online AP courses to students who wanted to take an AP course that wasn't offered in their own schools. Between 1999 and 2005, the number of public-school students in Florida who took AP exams jumped by 95 percent. That included increases of 132 percent for African Americans and 137 percent for Hispanics.

What's the Point of Advanced Placement?

Proponents of AP courses and exams, especially the expansionists, believe that the experience of taking an AP course pays dividends for students down the road, making them more likely to do well academically in college courses and to receive their degrees on time. The most enthusiastic believe that AP courses are so good that they help students do well in college even if the students skip the exam or score only a 1 (no recommendation) or 2 (partially qualified).

Such proponents frequently cite statistics published by the National Center for Educational Accountability (NCEA), which tracked the performance of Texas AP students who entered public universities in Texas in 1996 through 1998. The center's researchers found that students who had enrolled in AP courses earned higher GPAs in college and had higher college graduation rates than those who didn't take AP courses. The AP advantage appeared even when students were separated into groups by ethnicity and income.

But other research, using more sophisticated analytic strategies to control simultaneously for the effects of multiple distinctions among students taking AP courses, casts doubt on these findings. In fact, in another study of Texas school records, presented at the 2005 meeting of the American Economic Association, Kristin Klopfenstein of Texas Christian University and Kathleen Thomas of Mississippi State University found, as Klopfenstein says, "zero effect for the average kid" of AP enrollment on college performance.

At first, researchers at the Austin-based NCEA vigorously disputed Klopfenstein and Thomas's findings. But when they reanalyzed their data using similar techniques, the NCEA researchers quietly slipped a one-paragraph announcement into their newsletter conceding that AP exam scores, not AP enrollment, predict how well a student will do in college. In other words, simply taking an AP class isn't enough in itself. Only those who do well on the exam see benefits down the road. And it could be the case, says Klopfenstein, that these students were simply smarter to begin with. "We don't know yet."

Another possibility for the discrepancy is that the results of both studies are warped by misleading data in Texas. If a significant fraction of the AP courses in Texas are, as critics charge, just preexisting courses that were relabeled by educators eager to please government and business leaders who have been crusading for more AP participation, it would not be surprising that the studies found that simply taking a course labeled AP did not boost preparedness for college.

Transcript analysts do run into cases where schools are merely "changing the marquee but showing the same movie inside," says Clifford Adelman, a veteran federal researcher at NCES.

"It looks like the labels [on some AP courses in Texas] do not necessarily match the content," said Chrys Dougherty, research director of the NCEA. "Lots of kids are getting [high school] credit for advanced placement who can't pass an AP exam."

AP Cleans Up Its Act

While the AP exams have maintained their high reputation as a mirror of typical introductory college course requirements, opinions about whether AP courses themselves are better or worse than they were 50 years ago are split. Up to now, the College Board has not monitored the courses, and any school could call any of its courses an AP course. The theory was that the exams offered sufficient validation of the rigor and focus of the course. Poor courses would lead to poor exam results.

However, now that AP course taking influences college admission decisions, which are often made before seniors take their AP exams, the use of tests as the only quality control for the courses no longer satisfies the colleges and universities that dominate the College Board. Warnings were issued in 2001 by the College Board's Commission on the Future of the Advanced Placement Program and again in 2002 by a committee of the National Research Council (NRC). "Coverage of content [in science courses] may be superficial and opportunities for inquiry-based experiences insufficient," the NRC committee said after examining AP courses in biology, chemistry, physics, and mathematics. "The committee also has learned that some teachers discourage students from taking AP ... courses or the final examinations when poor performance is anticipated."

That's a shame, because common sense tells you that taking a genuine AP course is good for students, even if they flunk the test, contends Washington Post education writer Jay Mathews. "Even if they struggle with it, they've gone one-on-one against the academic equivalent of Michael Jordan. They'll have a visceral sense of what they're going to come up against in college."

Caperton points out that the College Board has implemented or started to implement virtually every recommendation of its own commission and the NRC committee.

One reform, seen as advisable in an era of widely publicized data-driven comparisons of schools, involves a new way of measuring the success of an AP program in a school (or, for that matter, a district or state).

Until quite recently, the College Board honored schools in which a high percentage of exam-takers scored a passing score of 3 or higher on an AP test. But educators complained that this approach gave schools an incentive to discourage all but their very best test-takers from taking an AP exam.

Another metric was invented by Mathews and is published periodically by Newsweek as the "challenge index." It is based on the ratio of seniors in a school (the fewer the better) to the number of AP exams taken (the more the better) by all the school's students, regardless of performance. That perversely rewards schools where few students reach the senior class. "Technically, a school could rank high by having all its students take lots of AP exams, even if they all bombed because their AP courses weren't high quality," said Trevor Packer, executive director of the College Board's AP Program.

So the board this year unveiled a new measure that it described as "the best single measure of equity and excellence in AP": the percentage of the student body or of a graduating class that earns a passing score of 3 or higher on at least one AP exam. That way, a school can't inflate its rank by relying on a small number of polymath geniuses, by sending ill-prepared students into AP exams, or by restricting access to AP. Students who score 1 or 2 on an exam would neither raise nor lower a school's standing.

Two more substantive changes are on the way:

* The College Board is revamping its science and history courses, which now mirror typical introductory courses in colleges that receive score reports on a lot of AP students. The courses have been described as "a mile wide and an inch deep." Redesigning the courses to reflect "best practices" is expected to take the rest of this decade as the board attempts to strike a balance between those who favor the mile-wide approach and those who prefer the mile-deep. Each subject has been assigned to committees led by college professors and including experienced AP teachers, with expert advice being solicited from groups that include the congressionally chartered National Research Council and the American Historical Association.

* Starting in the 2006-07 school year, the College Board will protect the reputation of its AP trademark by refusing to let a school call a course an AP course unless the school meets two conditions. First, the course must cover a genuine AP subject (as opposed to bogus subjects, such as the AP journalism course mentioned by the NRC committee). Second, the school's principal must submit a satisfactory self audit describing such items as the course's outline, some lesson plans, available support materials, and the teacher's experience, content knowledge, and opportunities for professional development.

This move toward quality control could be the first step back toward renewing the drive for national curriculum standards that flared up and flamed out in the 1990s in disputes over ideology, pedagogy, course content, and control of the public schools, speculated Bruce Johnstone, former chancellor of the 64-campus State University of New York.

"People could start wondering who authorized the College Board to set national standards for high schools," says Johnstone, who served on the College Board's AP commission. "This question has been glossed over as long as the AP has been perceived as basically good--a lot like collegiate introductory courses and not ideologically influenced--and the rest of the high-school curriculum has been seen as so flawed."

Perhaps there could be worse fates, however, than being perceived as setting the gold standard for high-school curricula.

The Resistance to AP

In Houston this summer, Nina Shokraii Rees, Assistant Deputy Secretary for Innovation and Improvement at the Department of Education, renewed the DOE's pledge of support for the College Board's efforts to see that the AP program will "leave no child behind." But she also noted that small rural schools and schools in poor neighborhoods remain much less likely than suburban schools to offer AP courses.

In fact, low-income examinees remain few and far between, even though two-thirds of public high schools, including many in large cities, now offer at least one AP course. Though federal, state, and College Board subsidies pay all or part of their AP fee, currently $82 per exam, only 10 percent of last year's AP examinees came from low-income families, according to the DOE. That was up from 8 percent five years previously and ahead of the benchmark targets in the department's five-year performance plan. But it was also well below the official federal estimate that 16 percent of public school students are from low-income families.

However, teachers say poverty is seldom a major factor in holding down the number of students taking AP courses or AP exams. Teachers at the Houston conference said more common reasons are that students aren't offered an AP course in subjects that they like, attend a school that resists letting students who might fail an AP exam into an AP course, don't have friends who take AP courses, or haven't been taught that it takes more than a high-school diploma to be ready to succeed in college.

Such concerns don't arise among the eager beavers competing for admission to the 15 percent of colleges and universities that are highly selective, says Michael Kirst, a professor of education at Stanford University. Kirst has spent years documenting the wide gap between the relaxed class schedules of many high-school students and the set of rigorous core courses that would prepare them to succeed in college.

"The ones who get hurt [by a failure to take AP and other rigorous high-school courses] are the 80 percent of students who are not in Advanced Placement or honors classes, and who want to go to the 85 percent of colleges and universities that offer broad access," Kirst says. Many get into a less selective college, but never graduate. Too many, he explains, "won't be ready for college because they have no clue about which high-school courses they need to take."

Vertical Teaming

The influx of underprepared students into colleges has created today's strange paradox: colleges are offering more high school-level remedial courses for no credit at a time when high schools are offering more college-level courses that earn college credits.

Many high-school students who say they want to go to college also believe (sometimes, justifiably) that they haven't been taught to read and do math at the level needed to pass a college-level course. That is why the College Board uses some of the fees it gets for AP exams on an array of programs designed to get middle-school educators and high-school educators to cooperate with one another in preparing students for the rigor of AP courses.

One of the board's approaches is "vertical teaming" between AP educators in high schools and the "pre-AP" teachers in Grades 9 and 10 and in the middle schools that supply their freshmen. Yet partnerships between middle and high schools on pre-AP and AP sequencing remain rare. One reason: trying to get overscheduled educators from one school to work with those from another isn't easy, especially when tight budgets in many states lead to demands to focus spending "in the classroom" rather than on planning and professional development for teachers.

Like the Houston conference goers, members of the online discussion groups that include some 29,000 of the more than 110,000 AP teachers in the United States debate the pros and cons of the trend toward lowering barriers, such as GPA requirements, to AP courses.

"The biggest assumption by other people is that somehow AP teachers have an easier job because they have the smartest kids," says Carol Wingard, an AP English teacher in Columbus, Georgia. "First of all, the smartest kids aren't always the best motivated. Second, we also deal with more-motivated kids who aren't necessarily the smartest. I work many more hours than I ever did before."

Certainly the College Board's efforts, now some 15 years old, to simultaneously increase access, equity, and excellence in the AP program seem praiseworthy, despite the persistence of the gap that Caperton laments between the many who want to go to college and the few who take AP courses.

AP Changes from a Way Through to a Way In

The AP program has evolved quite far from its roots--winning the cold war and alleviating the boredom of superbly educated students who had been exposed to virtually identical courses during their last year of prep school and their first year of college. But it remains focused on academic excellence. That is why many schools give extra weight to AP grades in calculating GPAs and why many college admission officers see AP course taking as a sign that applicants were ambitious enough to take the most rigorous courses available in their schools.

The College Board and other AP enthusiasts still tout the program as a cost-saving tool for accelerated learning. They tell the status-conscious or cash-short that high scores could help students save time or money by rushing through college. But numerous studies have shown that hardly anybody sharp and obsessive enough to pass several AP courses wants to rush through college.

RoperASW was hired by the College Board to survey students who had scores of 3 or higher about why they took AP classes. The firm reported that 83 percent said they wanted to increase their chances of going to the college of their choice. But only 33 percent said they wanted to finish college sooner. All but 15 percent did get credits for advanced placement in college, but they used those concessions to make room for other courses, rather than to graduate ahead of schedule.

"The AP program is useful because it has made it easier for [high school] students to get higher-level, more sophisticated courses," says Marlyn McGrath Lewis, director of admissions for Harvard College. "More students who are applying to demanding and rigorous colleges are able to get a more rigorous level of preparation for college than was once the case."

In effect, ironically enough, a program designed to help speed students through college as fast as possible is now used by most students to get into college. Perhaps that fact alone answers today's most vexing question about the AP: does it improve the academic experience of those who take part? But it remains to be seen if the College Board is up for tackling a bigger problem in its next 50 years: using AP as a lever to raise the academic standards of high schools.

Andrew Mollison, a freelance education writer, has been a Washington correspondent since 1974. He is a former president of the National Press Club.

The Testing Tsunami (Figure 1)

The number of students taking AP exams and the number of exams taken
have both skyrocketed. Even so, the average AP student took the same
number of program tests in 2004 (1.8) as in 1956 (1.7).

SOURCE: College Board, "Advanced Placement Report to the Nation: 2005"