State Testing and Suspensions: Is There a Connection?
by Tiffany Pakkala
If they have to cancel recess, offer pricey incentives or start school
early in order to improve student's state test scores, many school
principals will do it.
But what about preventing the worst-performing students from taking the
tests at all? Principals are doing that, too, according to a University
of Florida study that examines out-of-school suspension rates for low-
and high-performing students in 504 Florida schools.
UF economist David Figlio reviewed the schools' test scores and
discipline records and found that students who traditionally scored low
on state tests faced longer suspensions during test time than the rest
of the school year. They often missed state tests and test make-up days
during their punishments, and their schools' test scores, theoretically,
were higher because they weren't there.
At the same time, high-performing students involved in the same
incidents had shorter punishments at test time than they did during the
rest of the school year, Figlio found.
The average low-performing student was suspended an average of 2.35
days, while high-performing students had a 1.91-day average. About 23
percent of the low-performing students were suspended for a week or
longer, while only 18 percent of high-performing students received the
The study conducted from the 1996-97 school year to the 1999-2000 school
year could just be a coincidence stemming from the pressures of
test-time, Figlio said. But that doesn't explain why the change appeared
strictly in the fourth, fifth, eighth and 10th grades, the same grade
levels that carried a schoolwide ranking dependent on test scores.
"You can interpret that in a number of different ways. One is that we
have some cool, calculating principals trying to game the system,"
Figlio said. "Another could be that the principals are really stressed
out during test time and coming down harder on the low-performing
students without tying it back to the accountability pressure. But if
that were true, I think it would have occurred for all grade levels."
When he brought his findings to principals in Florida and other states,
Figlio said no one outright admitted to suspending students with the
hopes of improving test scores. But some did say testing season brings
extra pressure to maintain order and discipline in their schools.
Ironically, Florida is the least-likely place a researcher would find
such a trend today. While Figlio said he's "absolutely sure" the
discipline discrepancies are continuing in other states throughout the
nation, he said Florida did the most to close the loophole when it began
judging schools by the percentage of students whose scores improve from
year-to-year, rather than simply high scores, in 2002.
The change means even the lowest-scoring children have the potential of
succeeding, as long as their scores showed some improvement. Since
there's no way to predict who would improve from year to year, the
incentive to suspend low-scoring students no longer exists in Florida,
The researcher conducted his study in several of Florida's larger
elementary, middle and high schools under the condition that he keep
their identities confidential.
Once he gathered the data, he analyzed it using formulas so complex he
worried economists like himself would have a hard time understanding it.
Once he felt the numbers were clear enough for those in his profession,
he turned his study over to the Journal of Public Economics, which
published his findings in its May edition.
Figlio said the formula is too complicated to break down into simple
statistics, but that it was clear that - while students who get good
grades generally face lighter discipline than poorly performing students
- the gap becomes much wider during test time.
"If schools are selectively disciplining students, this raises some very
interesting questions about accountability," he said. "Getting rid of
the tests altogether would be short-sighted, but there needs to be an
accountability system that can't be gamed."
In Alachua County's public school system, research and testing director
Mel Lucas said he hasn't heard of such practices happening in Florida's
schools, "and I'd be very surprised if it happened at all in the schools
of Alachua County."
While Figlio's study is the first of its kind, at least one test
researcher finds the results "consistent with a very disturbing pattern."
"There's kids who are dropped out, as well as forced out, of schools
(because of low scores)," said FairTest.org's Public Education Director
Robert Schaeffer. "When test scores are all that matters, educators feel
the pressure to boost scores by hook or by crook. It's yet another way
standardized testing overkill leads schools to play games that hurt the
very kids that need help
INDEX OF NCLB OUTRAGES