Step forward, step backward

May 15, 2020

Greetings folks!

As usual, my tidings of good health and good humor. My dog Lucky and I continue to rough and tumble our way towards figuring each other out. Much as it is with the larger world, there are good days, there are trying days, there are steps forward, and there are steps backwards. But sometimes he asks for belly rubs, and he looks so silly and cute that I know we’re gonna get there. He is stubborn, like his papa…he wants to do things his way, just like his papa…and he will jump through hoops for hot dogs as long as he believes nobody is looking…just…like…his…papa.

In case you didn’t see it, Kevin Carey wrote this for the New York Times’ Upshot blog. One thing I noted is that in the last recession, state funding cuts were largely made up through tuition hikes. At that time, student debt was around $650 billion. It now exceeds $1.6 trillion. Students and families cannot finance another contraction of state funding.

SHEEO published their State of Higher Education Finances for FY2019 this week. Nationally, funding per full-time equivalent student (FTE) rose over 2018 levels, but it remained lower than funding levels seen prior to the Great Recession. The report gives a detailed look at where different states land on a host of measures. A couple of things stood out to me in terms of the California numbers.

  1. California is home to 15% of the nation’s FTE.
  2. Over the last five years, California has seen the greatest per capita funding increase in the nation (pages 53-54).
  3. California is one of just seven states that had met or exceeded pre-Great-Recession spending levels by 2019.
  4. California’s per FTE appropriation is above the US average (fig. 3.2), but when combining appropriations and net tuition revenues, California has the third-lowest dollar/FTE ratio. (fig. 2.2)
  5. This is in part because California’s net tuition per FTE revenue is the lowest in the nation, and tuition accounts for just over 20 percent of public higher education’s total revenue – the second lowest in the country (fig 2.6).
  6. California is on the higher end of the public higher ed support per capita measure (fig. 4.3)

There is a lot here, and so much more in the report! And all of it has implications for the decisions budget-makers will be considering in the coming months. It’s hard to know exactly what to make of it all, but, once again, signs point to the need for the federal government to fill what may be massive gaps. I think this is especially true here, where you consider point #6 – California taxpayers aren’t necessarily shirking their responsibilities. We want tuition to remain low to encourage students to attend, and, as noted, family/student debt cannot finance the system this time around. There is really only one source left…

Getting even nerdier, two articles from the most recent edition of the Review of Research in Education journal caught my attention. One looks at the use of quasi-experimental research designs (QED) in education. I have linked to a few articles in this vein, but this is probably a little more comprehensive than those. Remember, the gold standard way to figure out whether a program works is to randomly assign it to some students (treatment), while the other students get the business as usual (control). In medical terms, think trial drug vs placebo. For obvious reasons, that is hard to do in education. So, we look to other methods. QEDs take advantage of things out in the world that help eliminate selection bias. Sometimes things randomly happen to one neighborhood, but not to the other one right next door which is appreciably similar. In the remedial education space, a lot of studies have looked at students who were just below a cut-off and compared them to students who were just above the cut-off, because the difference of a point or two on those exams was basically statistical noise. This article goes over types of QEDs then examines the growth of this type of study in education, times when these are appropriate, and the implications for what gets studied in a world that prizes randomized control trials and QEDs.

The second Review of Research in Education article concerns an issue that is near to our Campaign hearts – race/ethnicity categories in data collection. This article discusses ways we might do a better job of collecting and capturing this information. Some of the recommendations are smaller, some I would say are quite far-reaching. I think one thing I hear quite frequently from friends in the data analysis world, though, is how frustrating the current data collection and reporting mechanisms are. I’m not sure I’ve given sufficient thought to how I’d change collection and reporting that I can say too much about the authors’ recommendations, but this is a great piece to read if you are thinking about these issues at all.

And finally, one from this week’s National Bureau of Economic Research set. The authors look at the Fund for Wisconsin Scholars, using years for which the need-based program was allocated using a random assignment. They find that employment among recipients was lower than control group peers in the first two years following their award. Employment was comparable in the following few years, but then dipped again in years six through eight when we would expect them to have completed and returned to the labor market. The group receiving the award had overall lower earnings for the eight-year span, but higher GPAs, concurrent with the hypothesis that receiving aid tends to substitute for employment and allows students more time to study. The employment issue in years 6-8 is trickier to explain. The authors offer two hypotheses: (1) grant recipients have less loan-debt and can be pickier about employment; and (2) recipients move out of state at higher rates than non-recipients (though this might be tangled up with the idea that high-GPA students move out of state at higher rates).

Stay healthy!

Vikash Reddy

 

 

 
Vikash Reddy, Senior Director of Policy Research


Read more blog posts