Steamboat Pilot & Today

Hayden School District officials are trying to sort out what caused 24 students in fourth and fifth grades to receive no score on the CSAP reading and writing tests.

Meanwhile, they are working to get the actual scores for those students and release them to parents and the community.

The false no-score reports were the result of a coding error, Superintendent Mike Luppes said. The biggest repercussions from the coding error likely will be the public perception that the school scored poorly for those two grades, Luppes said.

Students given a "no score" are not considered proficient or advanced. That means only 42 percent of fourth-graders and 30 percent of fifth-graders had a reading score of proficient or advanced, and only 23 percent of fourth-graders and 10 percent of fifth-graders had a writing score of proficient or advanced.

The only effect on school financing would be if parents see the results, do not know a coding mistake was made and decide not to move into the district because of the low scores, Luppes said.

The district might consider appealing the scores, but Luppes was unsure whether officials could or want to do so.

"Obviously, an error happened someplace, and we're working to get that corrected," he said.

Beth Celva, director of the unit of student assessment for the Colorado Department of Education, was confident the error could not have been caused by a computer glitch at the state level.

"No one ... ever touches a district's data, but we check it for purpose of comparison to see if there were any errors," Celva said.

The district is completely re--sponsible for completing and checking biographical data for each student.

"Nothing appears magically or disappears magically," Celva said. The scanners pick up the graphite left by a pencil mark, and it's unlikely or impossible for a small number of tests to show something weird.

Luppes has gone through the district's submissions of biographical data and pinpointed where the data went wrong.

The data is sent back and forth a couple-dozen times -- each time it is sent back, errors are highlighted so they can be corrected. Between the 25th and 26th submission, the coding changed, Luppes said.

There are 42 fields of biographical data that have to be filled out. Somehow, 12 students in the fourth grade and 12 in the fifth grade were coded with an "8" in the field for non-approved accommodation or modification, instead of a "0." Those "8s" signaled to the scanners that the students' test scores should not be counted.

Luppes said the error could have been the fault of the district or the CDE. But, he said, he approved the final submission of the biographical data, which included those errors.

The errors should not have a serious effect on school accountability ratings, Luppes said. The ratings look at a school's performance during several years, so a dip in two grades for one year should not have a huge effect, he said. Also, high scores in other grades should help buoy the district's overall rating.

However, whether the district had "high scores" in other grades also is open to interpretation.

In the reading test, excluding fourth- and fifth-grade results, the district scored lower than the state average in all grade levels except third and scored lower than last year at every grade level. On the writing test, only third- and sixth-graders beat the state average. In math, third- and fourth-graders, who were tested in the subject for the first time last year, did significantly better than the state average and were on par with Steamboat Springs students.

At all other grade levels, except seventh, the number of students scoring advanced or proficient dropped from last year. Seventh- and eighth-graders did better than statewide averages, but the number of fifth-, sixth-, ninth- and 10th-graders at the advanced or proficient level missed the state mark.