Next steps logo

New evidence in the field of cherry-picking

In a Florida House committee last week, a dutiful Democratic representative objected to a bill expanding Tax Credit Scholarships by raising a common objection. “I could run a successful school as well if I could cherry-pick my students, ’’ said Rep. Scott Randolph, “unlike the public schools who take everyone, both low-performing and high-performing.”

Leave aside the fact that the scholarship to which he referred is available only to students whose household income qualifies for free or reduced-price lunch and that the actual average income this year is only 12 percent above poverty. Randolph’s question persists in the national debate, and a new, as-yet-unpublished academic study gives us one of the most complete answers to date about which students choose these private options and why.

The short answer is this: In Florida, the students who choose these scholarships are struggling academically and come from school districts that don’t give them many other options.

The research is titled “Selection in means-tested school voucher program” and was conducted by University of California-Davis education professor Cassandra Hart with help from respected Northwestern University researcher David Figlio. It takes the measure of the Florida Tax Credit Scholarship, which is now 10 years old and has 38,000 students in nearly 1,200 private schools. Figlio has previously determined, with four straight years of consistent data, that students choosing the scholarship are among the lowest performers at the public schools they leave behind. This report goes further in providing context.

The students who choose the scholarship:

  • Were more likely to be black and were poorer than other students on free or reduced-price lunch in public schools.
  • Had lower standardized test scores than other students on free or reduced-price lunch in public schools before entering the program.
  • Were previously in public schools with low aggregate academic performance.
  • Were previously in public schools with higher rates of violent incidents and out-of-school suspensions.
  • Were previously in public schools in which principals were more likely to report that student disruptions interfered with learning.
  • Were previously in public schools in which teachers were more likely to report they spent too much time on discipline and too little time working with low-performing students.
  • Were more likely to live near a wide variety of private schools.
  • Were less likely to have public school options beyond their assigned neighborhood school.

“From a policy perspective,” Dr. Hart wrote, “examining public school contexts is important because these policies are often justified as a way to help students exit underperforming, unsafe schools. Knowing that these policies do in fact attract applicants in poor schools is useful in gauging the success of the voucher policies in opening up choices to families in low-quality schools.”

Just as important, the research suggests that, at least in Florida, there is no skimming the educational cream. These are students who are among the lowest academic performers and their parents go through an extensive application process to verify their income and usually pay hundreds of dollars out of pocket to make up the difference between the $4,011 scholarship and actual tuition and fees. As Hart noted, “These time and financial costs suggest that parents are unlikely to apply on a whim. Rather, some specific factors are likely impelling them to apply.”

The people who administer the program in Florida and work with the parents who apply for scholarships typically describe them as desperate. Their children are on a downward path academically, and they’re just looking for an option that might work. In other words, these are not students that most educators would associate with the words “cherry pick.”

Avatar photo

BY Jon East

Jon East is special projects director for Step Up For Students. Previously, he was a member of the editorial board and the Sunday commentary editor at the St. Petersburg Times, Florida’s largest daily newspaper, where he wrote about education issues for most of his 28 years at the paper. He was also a reporter and editor at the Evening Independent and Ocala Star-Banner. He earned a journalism degree from the University of North Carolina at Chapel Hill.


Are you saying that there aren’t smart kids in lower socio-economic areas? This seems to be based on that assumption given the title. Is it not possible to “Cherry Pick” from under-privileges areas? It would seem to me the author is trying to twists facts into an agenda.

Hi. You are right. It is indeed possible to cherry-pick from lower socio-economic areas, but the evidence is that the students who choose the scholarship are the lowest performers among the students on free or reduced-priced lunch.

“Had lower standardized test scores than other students on free or reduced-price lunch in public schools before entering the program.”

This is the only one of the bullet points above that pertains to the performance of the individual students participaticing in the program. The rest simply illustrate that the schools these students are leaving are lousy, and we hardly need a study like this one to tell us that.

The problem with the bullet point I referenced above, if I am reading the study correctly, is that the stats are flawed. This is because they do not compare the participants to the non-participants within their OWN poorly-performing school. Therefore, it may very well be that the participants are in fact the “cream” at the schools they are leaving. (In other words, they score lower amidst the overall sample because the overall sample includes a large number of kids from better schools.) So, these students, or rather their parents, are “self-deporting,” to borrow a term from Mitt Romney. The others, most of whom we can guess have less conscientious parents who are unwilling or unable to go through the application process, are left behind at an increasingly bad school.

Looking at bullet points 4, 5, and 6, points made are violence, student disruptions, and discipline. A parent can infuse good behavior or no behavior at all. Now, when it comes to the child going out in the world walking to school, stuff can happen and the child may have to prove something or get beat up or worse. A good behavior does not work with teen peers in rough neighborhoods. The school needs to work with these students in mediation that they all agree to leave everything outside the school grounds and the school is neutral territory. This worked when the new Akins High School in Austin, TX opened. The new school put rival gangs members from 2 other schools together and the principal got hit with a broke bottle, spent the rest of the semester out on medical leave. They had to do something so they got the gang leaders and such together in mediation. This was for the good of all, having neutral territory.

Comments are closed.