The analysis looked at each district in the state and compared the percentage of AP tests given a low grade - 1 or 2 out of 5 - in 2009-10 with the percentage of students districtwide who receive a free or reduced-price lunch, which the government funds for low-income students.
Low test scores often prevent students from receiving college credit for their AP coursework. Although it is up to each college to decide how well a student must score to receive credit, many require students to receive a 3 or higher.
There are limitations to the data available from the state Department of Education and the conclusions that can be drawn from a statistical analysis, but an outside researcher who reviewed California Watch's analysis said the correlation is robust enough and consistent enough with other evidence that these small discrepancies would not change the overall finding.
Sean Reardon, director of the Stanford University interdisciplinary doctoral training program in quantitative education policy analysis, who reviewed the analysis, said the findings raise - but do not answer - the question of whether all districts are adequately preparing students for the final exams.
"Although the correlation between free and reduced lunch rate in a district and the proportion of kids passing the AP test is clear, it is not clear whether this pattern is due to students being less prepared for AP classes prior to enrolling in them in high-poverty districts, or to the fact that they have less access to high-quality AP courses and instruction in those districts," he said.
Access to classes is more important than scores on the final test, said Joe Radding, administrator of the state Department of Education's Intersegmental Relations Office, which monitors California students' performance on AP tests.
"The tests are only one measure of knowledge," he said. "It may not be a sufficient way to measure the student's experience."
The state Department of Education tracks enrollment trends and administers roughly $4 million in federal money to help low-income students afford the test. The department relies on reports from the College Board, the nonprofit that oversees the AP curriculum and test, to assess the performance of low-income students. It does not conduct its own independent research.
The differences across the state are often stark. At Inglewood High School, where students filed a statewide class-action lawsuit in 1999 charging that they had been denied equal opportunity to take AP courses, more than 85 percent of the roughly 200 tests taken in 2009-10 were given a 1 or a 2, too low to receive credit at any of the California State University or University of California campuses. By contrast, at Beverly Hills High School, which serves a more affluent community, only about 20 percent of the tests received low marks.
Districts point to a number of factors to explain low scores, including a discrepancy between the rigor of their courses and AP final exam standards.
At the Cutler-Orosi Joint Unified School District in the Central Valley, almost every student in the district is eligible for free or reduced-price lunch. Despite the fact that many received high grades in AP classes, more than 80 percent of tests were given low scores.
"I think part of this issue in the past has been that those classes have not been rigorous enough," said Scott Norvell, an instructional data specialist for the district. "You look at the coursework, and you look at the grades. They're getting high marks, but the tests don't reflect how well they're doing in the class."
Under the direction of the new assistant superintendent of curriculum instruction, the district is working to help teachers better understand the AP curriculum and making sure that students have strong prerequisites before taking AP courses.
Betsye Steele, principal of Leadership Preparatory High School, a small school in Oakland, says scores have historically been low due to a lack of teacher training. Of the 51 tests scored, all but one scored too low to receive credit at many colleges. The district is partnering with the College Board this year to increase teacher training and introduce a special curriculum in ninth grade designed to give students a foundation for advanced courses in 11th and 12th grades.
As districts work to expand access to AP courses to low-income students, low test scores are not surprising, said Sergio Flores, a former school administrator, and Martin Gomez, a vice principal in the Fontana Unified School District, who gave a presentation in July about expanding AP access at the College Board's AP Annual Conference in San Francisco. They worked together to revamp the AP program at Fontana Unified. While the percentage of students receiving high scores isn't as good as it was a few years ago, they said they have a larger number of students passing than before, due to the expanded enrollment, and have a strong support system to help students who are struggling.
A recent ProPublica study examined whether states are providing equal opportunities for students to take advanced courses. It found that some states are leveling the playing field better than others. The number of California students enrolled in at least one AP course is close to the national average. (You can search for your school and compare it with others here.)
Some education researchers note that by the time a student is a junior or senior in high school, many of the educational inequalities already are ingrained.
"The best way to close these achievement gaps is to get children better prepared before kindergarten and in the early grades. By the time they get to high school, it might be too late," said Eric Larsen, a policy fellow at the Public Policy Institute of California.
The opportunity to take AP courses contributes to the likelihood of completing a bachelor's degree, according to the U.S. Department of Education. However, researchers at UC Berkeley found that the number of AP and honors courses taken in high school bears little or no relationship to students' later performance in college. Based on a survey of about 80,000 freshmen entering the University of California between 1998 and 2001, researchers Saul Geiser and Veronica Santelices found that exam scores, not just participation in AP courses, are a fairly good indicator of how students will do in college.
Read more California investigative reports at CaliforniaWatch.com.