For the second year, students' scores on Minnesota's statewide standardized tests remained flat across south metro, a trend that persisted across the state as well.

Despite the steady scores, some local school officials and leaders from the statewide teachers' union are questioning the impact a series of computer problems, including a cyberattack, had on results.

In Dakota and Scott County, scores varied widely among districts, from a high of 79 percent math and reading proficiency in Prior Lake-Savage to a low of 50 percent reading proficiency in South St. Paul.

For some districts, like Rosemount-Apple Valley-Eagan, results were a mixed bag of slight increases, decreases and scores that stayed the same as last year.

South St. Paul and Inver Grove Heights saw small declines in both math and reading scores, while Lakeville and West St. Paul-Mendota Heights-Eagan showed increases in both areas.

The Minnesota Comprehensive Assessments (MCAs) are math, science and reading tests taken annually by students in grades three through eight and high school sophomores and juniors. The tests, administered online, measure students' ability to meet benchmarks in each subject area and are also used to track the progress of schools and districts.

"We always try to improve on [MCAs] as well as everything else, but we try to keep it all in context," said Jay Haugen, Farmington's superintendent.

Haugen said there's much more to a student's education than how they score on the MCAs.

Inver Grove Heights and Farmington posted four-point decreases in math scores, the largest drops among districts in Dakota and Scott counties.

The biggest south-metro leap was a five-point increase in math scores in the West St. Paul district.

For Shakopee, both math and reading scores stayed exactly the same from 2014 to 2015, after the district posted a five-point increase in reading last year.

Dave Orlowsky, Shakopee's data and testing administrator, said he was proud of his district's scores. He attributed the steadiness to improved collaboration among teachers and new kinds of teacher training, like bringing in experts to coach teachers in reading instruction.

Five years ago, things were different — the district was "really at state level across the board," he said.

Now the district scores six to eight percentage points above statewide averages.

"We kind of put a stake in the ground and said, 'That's not acceptable to us,'" he said.

Computer concerns

This year's MCA testing season was marked by a series of online problems, from cyberattacks to students getting booted offline to problems with the online calculator used by students.

Cassellius said an independent analysis that examined disruptions on April 21, a day when there was a cyber attack, found no evidence the problem impacted students who were testing.

But officials from Education Minnesota, the state teacher's union, questioned that finding, saying a survey they conducted this spring showed that many teachers thought the glitches affected results.

"Parents expect these scores to show their child's academic progress, but no one knows if the 2015 scores of an individual student show what she learned or what she endured on testing day," said Denise Specht, Education Minnesota's president.

Some administrators, including Orlowsky, said they believe the problems adversely affected students.

"The stamina that's actually required to be totally engaged in throughout [testing sessions] is really surprising," Orlowsky said. "So I have a hard time believing it doesn't have some impact when kids have to be in there repeatedly."

At local levels, educators can decide for themselves whether to trust the data, said state education commissioner Brenda Cassellius.