Last month, 35 people, including former Superintendent Beverly Hall, were indicted for their alleged roles in a widespread test cheating scandal at Atlanta Public Schools. The educators are accused of a mix of charges involving altering -- or pressuring other educators to alter -- standardized test scores. Some of the accused then accepted bonus money based on the falsified test results.
As the long line of defendants made the perp walk into the county jail, critics of standardized testing policies noted that scandals like Atlanta’s probably resulted partially from the pressure to succeed on standardized tests.
"Standardized tests have a role in accountability, but today they dominate everything else and too often don’t even correlate to what students need to know to succeed," said Randi Weingarten, president of the American Federation of Teachers union, in a statement. "It is outrageous that schools in some states are spending up to 100 days a year doing test-prep or actual testing."
With students spending fewer than 200 days in school (Georgia requires at least 180 days in school; the national average requirement is about 178 days), the union chief’s claim could mean more than half the school year is spent on testing.
PolitiFact Georgia wondered if Weingarten’s number was correct and decided to take a closer look.
Weingarten’s and the AFT’s spokesman, Scott Stephens, told us that the union chief’s statement was based on both factual reports as well as anecdotal evidence from the organization’s 1.5 million members nationwide. "While 100 days is on the high end, it’s clear that in many states, ‘up to 100 days’ is a pretty good estimate," he said.
Stephens cited two reports to back up Weingarten’s claim. The first, from the National Center for Fair and Open Testing, includes findings from the Wisconsin Association for Supervision and Curriculum Development about the amount of time spent on testing in general and for students with special needs. The Wisconsin association found that Wisconsin teachers spent a per-district average of 976 hours administering tests. The data in that report is based on information collected in the 2005-2006 school year, and uses cumulative numbers instead of breaking down the information into the time spent per district or per school.
The second report is a 2012 white paper by the Central Florida Public School Board Coalition. The group found that in Florida, students will spend at minimum nine weeks in kindergarten and up to 21 weeks of the school year in 10th grade on testing. Twenty-one, five-day weeks equate to 105 days, which satisfies Weingarten’s claim.
Based on the data some school districts have presented, Weingarten’s claim is correct, said Robert Schaeffer, public education director for the National Center for Fair and Open Testing, also known as FairTest. That group has also been critical of standardized tests for the time and expense required by educators and school systems to administer them.
Schaeffer’s organization does not track standardized testing days nationally, but has supported legislation in states like Washington, requiring school systems to post testing schedules so parents can see how much testing must be done.
It is important to note that the testing numbers included in the studies are cumulative numbers. Not all students take the same tests at every grade level. And not all tests are administered the same number of days. Also some of the tests, like end-of-year exams, can be taken multiple times by some students, which would also impact the data.
PolitiFact Texas has examined similar claims and found that statements made about the amount of standardized testing mandated for school systems was exaggerated, as not all grades or all students took tests on each of designated testing days.
Test preparation is broad and could be considered everyday classroom standards, said Matt Cardoza, spokesman for the Georgia Department of Education. Test preparation can also differ by school.
For example, Georgia administers three basic standardized tests. Simply adding up the available testing windows for the three easily exceeds 100 days, but provides a misleading statistic. Although the test windows cover numerous days, actual test administration days are much fewer. All three tests are not administered to all students on each available day. Also key, the tests do not take all day to complete.
Unfortunately, there is no good, reliable national data involving this part of standardized testing, said Tom Loveless, director of the Brookings Institution’s Brown Center on Education Policy. (The Council of Chief State School Officers no longer tracks the data.)
Also, what is designated as test prep can vary between educators, and could be construed as almost anything taught during the school year, as opposed to skills that would not regularly be taught, Loveless said.
"And I think (Weingarten’s) statement points to an extreme example using questionable sources of evidence," he said. "The statement is quite vague and an exaggeration."
So does the union chief’s claim get a passing grade?
Weingarten said that schools in some states are spending up to 100 days a year doing test-prep or actual testing.
Upon initial review, basic testing calendars from some school districts bear out the claim. But a deeper delve into the data finds that not all students in each district take each test, tests are not administered each day during the multi-day testing windows, and in some cases, the tests do not cover an entire day.
Weingarten allowed herself some latitude by using broad phrasing, such as "some districts" and "up to 100 days" in her claim. But the thrust of her statement is misleading, according to education experts.
Weingarten’s statement contains an element of truth but ignores critical facts that would give a different impression.
We rated the claim Mostly False.
Staff writer Karishma Mehrotra contributed to this article.