Preliminary data overestimated test results in Perham schools
Monthly evaluations taken throughout last school year at the elementary and middle schools in Perham had teachers anticipating better results on state standardized tests.
District leaders are now looking into the discrepancies between those preliminary evaluations and actual student test performance.
In a written report presented to the school board at a recent meeting, middle school Principal Scott Bjerke said the preliminary data was off-base in all grades at the school - five through eight - with student scores generally 14 to 22 percent lower than expected.
The school had been considered to be on a 'pathway to proficiency,' meaning students showed an understanding of the skills they were to be tested on.
In fifth grade, for example, 80 percent of students showed a strong understanding of necessary math skills on the preliminary evaluations. However, only 58 percent of those students actually met standards on the Minnesota Comprehensive Assessment II math tests.
The Perham-Dent School District uses an evaluation program called STAR, offered through the company Renaissance Learning, to conduct the monthly evaluations. Last year was the first year the middle and elementary schools utilized STAR assessments. The high school does not use the assessments.
Bjerke's report stated that STAR reading and math assessments were not meant to be exact indicators of the MCA test results: "However, the STAR assessments are based on content from the state standards... One would think it would be a fairly accurate indicator of success on MCA tests."
Elementary Principal Kari Yates stated in an email that the correlation between STAR assessments and MCA results in the elementary school was not as high as she had hoped for.
The elementary school may not use STAR assessments at all next year, Yates said, "because the state assessments, which are required, are evolving in ways that make them increasingly meaningful tools for shaping instruction."
Thanks to new features of the state standardized system - such as online testing with immediate results - "we know a student's strengths and needs the moment they complete the online MCA math assessment," Yates said.
In a follow-up interview, Superintendent Mitch Anderson said the district is looking into the inconsistencies with the STAR data.
"I do believe that testing anxiety can be an issue for some students, but I don't feel that is the sole reason for the wide gap," he said.
Yet the overly optimistic STAR results don't mean the program didn't serve in other helpful capacities, Anderson said.
He explained that the STAR reading and math assessment programs are not just used to prepare for MCA tests, but they also help teachers and administrators monitor student progress and measure individual growth.
Bjerke also commented later on the benefits of STAR, stating that, despite the discrepancies, the STAR assessments provided the middle school with a common form of data that teachers can use to continue to monitor student growth.