Since I started my research fellowship, I’ve gotten many more questions about what types of research I’m working on and what, if anything, I’m learning. As we near the end of the fall semester, and the halfway point through the first year of my fellowship, I thought it might be a good time to take stock of what I’ve learned. Perhaps others will find the following few bullets of results more enlightening and interesting than I’m sure many of my in-person responses to such questions have been in the past.
And this figure tells us it doesn’t really matter which type of public college you look at – prices were on the climb everywhere. Underlying these trends are huge year-to-year increases in tuition and fees (often in the ballpark of 15 to 20 percent!) at public institutions across the country (e.g., California riots!?). Such increases were especially prevalent after the year 2000.
With a colleague, I’ve learned that large increases in the cost of attending 4-year public institutions of higher learning are causing college-going students to do at least two things: (1) Substitute down within state systems of higher education – they become less likely to go to the top, flagship schools, and more likely to go to less selective 4-years or 2-year institutions; (2) Go out of state to college, rather than attend local 4-year schools that have become relatively more expensive. Further, these effects are largest among low-income student populations. Both findings hold potentially deleterious implications for college completion rates and the robustness of local labor markets.
2. For a while now, economists and policymakers have been concerned about the effects mandatory high-stakes high school exit exams might have on student dropout. Using district-level dropout data from 1998 to 2008, I find that the existence of an exit exam is associated with an increase in the high school dropout rate of about 0.55 percentage points, relative to if no consequential exit exam policy were in place.
Turns out, this effect is concentrated among 12th grade students (1.44 percentage point increase), minority groups, and is almost exclusively driven by states in which students do not have access to alternative graduation pathways (i.e., replacement educational experiences for those unable to pass some or all of the state’s exit exam).
To put these findings in context, you might ask what average dropout rates looked like over this period of time… and if you did… I could happily show you:
Perhaps the silver lining to such findings is that, if alternate ways, beyond retest opportunities, for students to demonstrate mastery of state standards mitigate the increases in dropout rates otherwise observed when students are exposed to high-stakes, rigorous exit exams, this suggests that such state-level policies may be able to reduce some of the social costs associated with high school exit exams. Nevertheless, this assumes at least two important things: First, the alternate graduation pathways crafted by states are indeed adequate proxies for the educational experience and skills students would have received if the goal were still to help them successfully pass the exit exam; and second, teachers, schools, and administrators are not strategically funneling underprepared or minority students into alternate graduation pathways and thereby sub-par high school educational experiences. These two points are worthy targets for future research.
3. For those of you interested in K-12 research… I discovered an interesting impact of the main sanction embedded in the No Child Left Behind (NCLB) law on school-level student performance: In terms of math and reading proficiency, schools performed worse the year after being labeled as “failing to meet adequate yearly progress” IF the reason for failure was that the school missed the aggregate student achievement goal (e.g., only 53% of the school’s students were proficient in math, when the goal was 62%). But, if the school met this aggregate benchmark, but failed to make progress solely because of a particular student subgroup (such as special education students, African American students, Asian students, Limited English Proficient students, etc.) – students within the failing subgroup performed better the following year.
So, schools that meet achievement targets for the aggregate student group, yet fail to meet at least one demographic subgroup’s target see between 3 and 6 percent more students in the failing subgroup score proficiently the following year, compared to if no accountability pressure were in place. I see this as a story of scope: Broad failure leads to little improvement, whereas focused failure allows schools to more effectively direct scarce resources. For all its negative aspects, NCLB did shine a spotlight on traditionally underserved student populations – and it looks like it increased their performance (at least in one state).
-----
If you’ve made it to this point, congratulations. Thinking about some of the major things I’ve learned over the past year or so, as a function of being engaged in research concerning the economics of education and education policy was helpful for me – and I hope that some of you loyal readers found this write-up interesting, provocative, and policy-relevant at best… and an acceptable substitute for Nyquil at worst. I’d love to hear thoughts, impressions, or questions that this raised for you. Also, I promise more riveting posts from my better half soon.
