Gneezy and List take the book in a completely different direction in Chapter 4 when they use evidence from field experiments to suggest methods to reduce the infamous Achievement Gap in public schools.
The authors begin by laying out the facts of the achievement gap, particularly between high income families and those living in inner cities. The U.S. spends the 5th most amount of money per capita on students – $11,467 – yet there is a 9% drop out rate for “low-income” students and drop-out rates of above 50% for inner city kids, compared to a low 2% rate for “high-income” families.
Similarly striking, the average ninth grader in New York and Chicago public schools reads at the same level of a third or fourth grader in “better endowed” school systems. The U.S. has fallen outside the top ten in reading, math, and science scores and its graduation rates mirror those of Mexico and Turkey.
Yet the U.S. isn’t getting worse across the board. Gneezy and List imply our worse students are bringing U.S. scores down, which must mean that the achievement of top students rivals those of Eastern Asian and Northern European countries. Simply put, our education system works great for some and very bad for others. The authors wanted to know how we can get it to work for everyone.
To do so Gneezy, List, and their team of researchers conducted several field experiments in the public schools of Chicago Heights; a town of 30,000 located thirty miles south of Chicago where the average per capita income is below the poverty line. Using a large donation from Kenneth and Anne Griffin they created several monetary incentive laden programs for Chicago Heights students aimed at improving their test scores and drop-out rates.
The motivation behind providing monetary incentives to students is that young children and even more so, teenagers, fail to see the advantages of an education because at this point in their development their brain is wired to respond much better to immediate gratification, say playing a new video game, than gratification that is long delayed like studying hard to get into your dream college.
This, the authors say, explains why students drop out, failing to recognize the long term advantages of staying in school over the short term advantages of dropping out, like being able to earn money immediately by working – a very popular alternative in the public high school I attended.
You can read about all the specific ways the researchers incentivized students yourself but let’s highlight some of the findings:
- Offering 9th graders $50 at the end of each month and entrance into a monthly lottery of $500 for grades all above C or higher and no unexcused absences or daylong suspensions led to 50 of 400 “borderline” students meeting 9th grade achievement standards. The incentives also increased achievement of students “on the brink of failing” by 40% and these students continued to outperform un-incentivized students (the control group) after the program ended.
- Paying students $20 for improving on their standardized test scores (presumably reading and math tests) increased scores by 5-10 percentile points on a 100 point scale.
Gneezy and List also discuss how they incentivized teachers and what results that produced:
- Giving each teacher in a team of two a 4,000 bonus for improved scores helped to increase scores (Gneezy and List don’t provide numbers here).
- But not as much as giving the teachers the bonus at the beginning of the year with a stipulation that it would be taken back if certain standards were not met. This improved scores by 4-6 percentile points, more than the former treatment.
The authors highlight this last result because framing an incentive as a loss was more effective than framing it as a gain. This was true with teachers as well as with the students who were given $20 for improved test scores.
Though this result is grounded in a lot of theory including loss aversion, Gneezy and List are brief in discussing the particulars in this chapter. If you’re curious Amos Tversky and Daniel Kahneman have a seminal paper about loss aversion and framing effects.
Next comes an interesting discussion about whether incentivizing the three major inputs (student, parent, teacher) of a student education at the same time would increase scores more than incentivizing each party in isolation as they had done in previous experiments. What their experiments found was counterintuitive; motivating one input with a high reward was more effective in improving scores and drop-out rates than splitting the reward equally among all three.
Gneezy and List conclude the chapter by mentioning a solemn pattern they had encountered. High school students responded worse to the incentives than younger students. A possible reason for this is that by the time they are fourteen students are “tilting one way or the other.”
That is to say, their early experiences are so influential that if they are massively underachieving by the time they are in high school there may not be anything that can be done to help them catch up.
Or is there? To find out you’ll have to read chapter 5.