For the first time since 1999, the state does not have a standardized-testing contractor, forcing the Education Department to manually check and refine 640,000 scores on the new, controversial Common Core tests given this spring.
Parents can expect individual student reports sometime after Nov. 9, later than usual.* But first, officials will release statewide results Oct. 12, and then school- and district-level results the following week — days before the primary election for the state school board.
State Superintendent John White said the state simply didn’t have enough money to ask its contractor to do everything it typically does because of a compromise in the Legislature this spring.*
A timeline released Friday by the department said the massive process would take only two weeks, and then another two weeks to verify that the initial raw score was correctly converted into a scaled score. That’s a number used for comparisons to other states and sets a benchmark for measuring growth.
In a Sept. 17 story in The Advocate, White said a vendor was doing the conversion. But department spokesman Barry Landry later said staffers would be doing the work.
Ken Pastorek, another department spokesman, declined to say precisely who is doing the work or how.
About 320,000 third- through eighth-graders statewide took two tests, one in math and the other in English.
White said typical procedural changes associated with a new test have slowed the release of scores this year, such as the test-maker’s review of student answers to verify the expected difficulty of the questions. He said the same delayed release happened in 1999, when a new LEAP test was introduced. However, those results were issued at the end of May that year, the same period when the 1998 and 1997 results were released.*
The scores are necessary to compute a school’s accountability letter grade, and those designations won’t be made until December, about two months later than in the past several years. That’s very close to the end of New Orleans’ OneApp process for applying to the city’s schools with admissions requirements.
White has been criticized by the state school superintendent’s organization for saying that the scores would be “sobering,” but then not providing the information on which he based his assessment. He relented this week and released raw scores to superintendents who asked for them, but cautioned that the results will mean little to them at this point.
“All that a raw score is a number of points a kid got on the test out of the points available,” White said, adding that local districts won’t be able to discern what particular questions or concepts students may have had trouble with.
White has dismissed accusations that he’s delaying the process because of the election. He said the schedule for releasing results has been set for at least a year. However, information he released to back up that assertion does not include specific dates for release, merely that the results will be out in the fall. Even so, he stressed that most information down to the school level will indeed be released before voters go to the polls.
All eight elected state Board of Elementary and Secondary Education seats are on the Oct. 24 ballot.
Like Jindal and White, the candidates are lining up on either side of the Common Core issue. The results of the first administration of the tests aligned with these national standards likely will be used selectively by both sides to support their contentions that the tests should be kept or scrapped.
The state contracted with Data Recognition Corp., a testing company based in Minnesota, to create and provide the state standardized test for the 2014-15 school year. But that contract ended June 30.
Landry said the company was to provide the state with the information necessary for it to convert raw scores to scaled scores.
“I do think that it is probably true that this year there is slightly more in the production of reports that we are doing than the vendor is doing,” White said in an interview with The Lens.
That’s a significant difference from years past, White said. “And the reason for that it is just money.”
After the scores are put on a standard scale, the state board needs to set the cutoff scores for each of the five categories. White said the Board of Elementary and Secondary Education will consider those at its Oct. 13 meeting.
In a call with reporters Monday, White said he would recommend the same cut scores to BESE as several other states are using. Like 10 other states and the District of Columbia, Louisiana students took a test devised by the Partnership for Assessment of Readiness for College and Careers, or PARCC.
Students across Louisiana “took the exact same form as did kids across the country,” White said. “Same questions. Same order. Nothing different.”
The state also can set different cutoff levels for its own purposes, aside from the national comparisons for which PARCC was established.
For instance, in Ohio, categories don’t line up. If students score in the third category of five, they will be labeled “Approached Expectations” by PARCC. Ohio is categorizing them as “Proficient.”
White said that is all a matter of interpretation and that Louisiana is in the middle of a 10-year process to raise standards.
“It will be over time that we’ll migrate away from calling a Level Three good to calling a Level Four good.”
*Corrections and clarifications: The original story and headline mistakenly said student-level results will be issued in December. The school letter grades will be issued in December, but individual student reports will be out the second week of November. White called after the story was published to clarify that the state had money for a contractor, but not to pay for all the work it usually did. He also said he made a mistake in saying LEAP scores were delayed in 1999; he clarified that it was the iLEAP test.