Computer Disruptions Didn't Affect Minn. Test Scores, Review Finds
Recurring disruptions that affected thousands of Minnesota students taking online state standardized tests last April did not have any systematic negative effect on student test scores, according to a report released by state education officials on Thursday.
The findings mirror similar conclusions drawn by Indiana state officials after computerized test glitches there. Oklahoma and Kentucky schools also experienced online testing problems in the spring.
"While there is some evidence to suggest that there were effects from the disruption, nothing emerged in a systematic way across grades, subjects, or methodologies" in Minnesota, according to a report released by the Human Resources Research Organization, a third-party organization hired by the state to analyze the glitches' impact.
"That is, there is no statistical evidence to suggest that the disruption, on average, adversely impacted students and schools that tested on April 16 or 23," according to the report.
On those two days, many Minnesota students experienced delays while test questions loaded, and many were unexpectedly logged out of the exams mid-test. According to the review, under normal conditions, test questions took well under one second to load, but 11 percent of students experienced "latency" delays of at least 30 seconds on April 16 and over 16 percent of students experienced such delays on April 23.
All told, according to the group's report, between 15 and 23 percent of Minnesota schools that tested online in reading had known disruptions, while between 9 and 15 percent of schools who administered online math exams had known disruptions.
Minnesota's testing vendor, the American Institutes for Research, or AIR, has acknowledged that problems with its equipment led to the April 16 glitches, but said no problems were found with its system on April 23, according to an Associated Press report.
In a letter to superintendents across the state, Minnesota Department of Education Commissioner Brenda Cassellius said that although the review found no statistical evidence of statewide impact on test scores, "there is no doubt that a number of Minnesota students were affected on those days."
"I am instructing AIR to take all preventative measures to make sure this does not happen again," Cassellius wrote.
The letter makes no mention of whether Minnesota will seek financial damages from AIR. Oklahoma officials recently agreed to a settlement package with their testing vendor, CTB/McGraw-Hill, worth more than $1.2 million. Indiana officials say they are still negotiating with CTB/McGraw-Hill over problems in their state.
In her letter, Cassellius also noted that a state technical advisory committee has recommended that moving forward with plans to report individual student test scores, even for those who experienced disruptions, is "appropriate and valid." Also "appropriate and defensible," according to the committee, are plans to use the scores in Minnesota school accountability calculations.