Myth: Transient students bring down test scores

February 24, 2004

Kimberly Swygert over at Number 2 Pencil has an interesting post on the threatened closure of a Harlem charter school. She took them pretty well to task, poking holes in all their arguments and excuses except one: The school had claimed that it did poorly on the standardized test because of a new crop of 8th grade students.

Kimberly writes, "the argument that some kids had been at the school only five months before being tested isn't ridiculous."

Ah, they slipped that one past her quantitative eye, so we gotta give them some credit.

In the world of statistics, the most useless adjective (or useful, if you like to use statistics to mislead people) is the quantifier "some," as it covers anywhere from 1% to 99% of a population.

Some kids? Exactly how many? What percentage of the test-taking population? More importantly, what was the average number of months that children attended that school before taking this damning test?

It's quite possible a good portion of the 8th grade class in question had been at that school for years, as we shall explain.

We've had a fair bit of experience with charter schools, and a common way for a charter school to get on its feet is for it to start small, then "vertically" expand, adding one grade level per year for one or more years. While it's possible for a school to expand toward the younger grades, by far the most common method of vertical expansion is to grow with the students.

For example, if a school's highest grade is 6th grade, when the current 6th graders graduate, the school can expand to 7th, hiring more staff as necessary.

Reisenbach's parents and staff tried a number of arguments to explain away their failing scores, including the fact that "2002 was the first year the school had an eighth-grade class."

It would logically seem that the school only had higher grades, and expanded downward to 8th grade. Thus, when these students were tested for the first time, they got slammed unfairly.

The facts are the precise opposite. The school has served younger students all along, and has expanded vertically as their children graduate each grade. (A quick Google search revealed a pdf report showing the school with 75 seventh graders in 2001. If this school is so universally well-loved by parents, you can bet that a huge chunk of that 75 became the very eighth grade class that tanked in reading and math.)

The "transient students are pulling our scores down" appeal is largely a myth, but makes for a handy excuse.

On a side note, one parent claimed in defense of Reisenbach, "All schools in Harlem are failing." (So why should they pick on Reisenbach?) Apparently this parent isn't aware of the huge number of private and parochial schools in Harlem, which, by the very nature of the tuition-based educational free market, cannot all be failing.

Our recommendation for Reisenbach? Throw them a bone, and grant them a conditional renewal. Hand all the reading teachers a copy of Why Johnny Can't Read, and tell the math teachers they can earn professional development credits if they read fifty articles (any fifty!) at MathematicallyCorrect. Finally, to build synergy, all staff members should be required to study the high-performing high-poverty schools profiled in No Excuses (125-page pdf).

Then, in three years, see where their scores are. If they don't have at least half their kids passing the reading and math tests, then they don't deserve to stay open, not with taxpayer money they don't. Tell them to reincorporate as a private school and see just how many parents would pay out-of-pocket tuition for that kind of failure!

Let's keep those #2 pencils sharpened!