Indiana third-grade reading results: big splash, bad data

Indiana Superintendent of Public Instruction Tony Bennett made a splash Tuesday with the release of 2012 IREAD-3 test results for the state’s third-graders. But maybe he should have waited a bit longer to leap into the pool.

No sooner had Bennett announced the numbers than his Department of Education sent out a notice that some of the numbers were being corrected. And two days later, there are apparently still inaccuracies in the data the state has made available.

It’s impossible to know how far off the reported data are from the actual test results. It’s probably not a lot, but it’s enough to matter to the affected schools.

“It’s unfortunate the data was not right before the state sent it out,” said Cameron Rains, director of elementary education for the Monroe County Community School Corp. in Bloomington.

IREAD-3 is the new reading test that Indiana third-graders must pass to be promoted to fourth grade. Three categories of students can be promoted if they don’t pass: special-needs students, English language learners and students who have already been retained two or more times.

Third-graders who didn’t pass in March can try again in July. In the MCCSC, those students have been receiving intervention and will be asked to take part in a seven-week summer reading camp, Rains said.

Part of the data problem involved students who attend private schools or are home-schooled but receive special-education services from public schools. Initially, those students were mistakenly coded as attending public schools. DOE spokeswoman Stephanie Sample said the error involved 700 students statewide.

But even with that issue addressed, discrepancies remain, Rains said. As of Thursday, at least nine students had their scores attributed to MCCSC schools even though they hadn’t been enrolled in the schools when the test was given in March. A couple of the students had never attended MCCSC schools.

The verified passing rates for MCCSC schools were published last month by the Bloomington Herald-Times (subscription required). Rains stands by those numbers.

The Department of Education sent initial IREAD-3 results to schools and news media on Monday afternoon, giving notice that Bennett would report the results at a news conference at 11 a.m. Tuesday in Indianapolis. Schools officials noticed problems but weren’t able to contact the DOE until Tuesday morning. By that time, the announcement was just hours away.

Sanders sent news media a follow-up notice Tuesday afternoon, after the news conference, adjusting the results by removing the private-school and home-schooled students who didn’t belong in the data. She said news media will receive “regular updates as first-year results continue to be reconciled.”

Why does all this matter if only a few students are affected?

For one thing, if 90 percent of third-graders in a school pass the test, the school is exempt from certain requirements regarding how they carry out reading instruction. Templeton Elementary School in Bloomington is on that bubble. The district says 91.7 percent of its students passed, but the state says its pass rate was 87.5 percent.

Another issue is that teachers and students worked hard to improve reading skills, and they deserve credit for success. At Bloomington’s high-poverty Fairview Elementary School, just over 80 percent of third-graders passed IREAD-3, close to the state average. But the DOE figures show only 75 percent passing. “It was sort of a kick in the stomach for people who worked so hard,” Rains said.

And while the DOE will no doubt keep updating the test scores until it gets them right, those minor adjustments won’t make the news.

At Bloomington’s Marlin Elementary School, for example, the initial report from the state on Monday said that only 70.7 percent of third-graders passed. By Tuesday afternoon, the state had revised the figure to 80.6 percent – which MCCSC officials agree is correct.

But look up Marlin in the searchable IREAD-3 databases created by both the Indianapolis Star and NPR’s StateImpact Indiana, and you’ll still be told that the school’s pass rate is 70.7 percent.


5 thoughts on “Indiana third-grade reading results: big splash, bad data

  1. Am wondering if the state DOE needs to go back to summer school as well, to learn how to “read” results. I know how hard the Marlin School teachers worked to get students ready. How does one explain a discrepancy of 10 points?

    • Yeah, 10 points is huge. It looks like the state initially coded five students to Marlin who don’t attend there.

  2. getting reports from several school that the ESL and Special Education students who received waivers as well as those students who were absent (all Zero scores) are included in IDOE’s calculation # of possible students taking the test (the denominator). This statistically incorrectly calculates the passing rate of this test and nullifies the validity of the test.

    • I think it’s fairly certain that the students who will receive waivers were included in the calculations — not sure about students who were absent. You’re right that including the waiver students undercuts the validity of the percentages, at least as a measure of how well the schools are teaching reading. You can imagine a small school with self-contained special-ed classrooms for, say, the entire school district — that could skew the school’s passing rate quite a bit.

  3. Pingback: Standardized testing in Indiana « One Teacher's Voice

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s