Each year, more high schools across the United States are integrating Advanced Placement (AP) classes into their course offerings, providing students with a cost-effective alternative to general college courses. AP scores are seen to indicate college readiness and a student’s ability to succeed in a specific subject. According to a study recently published ahead-of-print in the Journal of Human Resources, more than 70% of US high schools have adopted AP courses into their curriculum, some even requiring students to take them. In the first-ever experimental study of the AP program, authors Dylan Conger, Alec I. Kennedy, Mark C. Long, and Raymond McGhee Jr. found that the classes succeed in improving students’ skills but conversely may also lead to reduced confidence.
The study offered enrollment in AP biology or chemistry courses to randomly selected students from twenty-three US high schools. At the end of the semester, researchers measured AP- and non-AP students’ ability to analyze and develop arguments about science and participants were surveyed to assess their confidence in the subject, their interest in a future STEM degree, and their levels of stress. As many high schoolers could attest, students in AP science classes were found to have increased stress levels and lower grades due to pressure and rigor. Additionally, taking AP classes decreased students’ confidence that they could succeed in STEM courses in college. But despite the negative impacts, the study found that these classes do successfully increase students’ scientific skill levels and prepare them for college-level coursework, as well as increasing high schoolers’ interest in majoring in STEM in college.
Study author Dylan Conger discussed with us these surprising findings, as well as the origins of tracking AP performance effects. To learn more, read the full Journal of Human Resources preprint article, “The Effect of Advanced Placement Science on Students’ Skills, Confidence and Stress.”
How did you decide to pursue this topic?
My collaborator, Mark Long, and I had been studying advanced high school courses for a few years and we were struck by the lack of causal evidence in support of the AP program despite its near-ubiquitous presence in US schools. Determining how the AP program affects students is difficult because students self-select into the program and teachers often decide which students are allowed entry. We started to brainstorm about how we might successfully randomize access to AP courses. We landed on a research design that was ethical and that would minimize concerns among parents, students, and educators. We decided to focus on AP science courses in particular because these courses were being promoted by policymakers and educators as a key tool for improving the STEM workforce in the US.
Though AP courses are often seen as a tool for college preparedness, how would you explain the low confidence that AP science students have for achieving success college STEM courses?
We spent some time reading the literature from psychology and learned, perhaps somewhat intuitively, that reduced confidence doesn’t necessarily have a negative effect on student performance. In fact, some of the literature suggests that overconfidence can lead to academic failure. AP courses are very challenging and they cause some students to lower their estimation of their own ability. In our study, we found suggestive evidence that this loss of confidence did not interfere with their learning in the AP class itself. We found that the AP students gained more knowledge in science than the students in other honors and regular courses. How that affects learning in college is an open question.
What is one takeaway from your article that you’d like to communicate to a non-specialist?
The schools that participated in our study tend to have above-average shares of low-income students. Many of the students in our study were eligible for subsidized school meals. For these types of schools and students, our findings suggest that the AP program has both benefits and costs. For instance, we find that students appear to have learned more about science in the AP course than they would have learned in other regular and honors courses. At the same time, the AP courses led to worse grades, losses in confidence, and higher levels of stress.
Why did it make sense to publish in the Journal of Human Resources?
As one of the leading journals focusing on policies that promote human capital, the JHR was a natural fit for our paper. The JHR has also been intentional about disseminating the research to a broad audience.
What’s one question that emerged from your research that you’d like to follow up on, or that you hope someone else looks into in the future?
This paper focuses on the short-run impacts of the AP program on students’ cognitive and socioemotional outcomes. Ultimately, everyone wants to know how the AP program influences students’ college-going, and in particular, college-going at selective institutions. We are currently working on a follow-up paper that estimates the effect of AP science on these important life outcomes. We also plan to follow our cohorts for a few more years to determine whether and how the AP science program influences their college graduation.
Dylan Conger is a Professor of Public Policy at the George Washington University and a research affiliate at New York University’s Institute for Education and Social Policy. Conger’s research focuses on explaining disparities in achievement between social groups and evaluating policies aimed at reducing those disparities.
