Key Facts and Findings:
Minnesota schools spend significant time and resources on state standardized tests, but their usefulness is limited.
Standardized test scores are the state’s primary measure of school performance and student achievement. Although test scores have limitations, they enable comparisons of student performance across schools and school districts.
Federal law drives the use of standardized tests in Minnesota. The state must meet federal testing requirements in order for state and local entities to receive various federal grants. In 2016, Minnesota used $325 million in federal education funding tied to these requirements.
The Minnesota Department of Education (MDE) primarily uses two tests to meet federal requirements. The Minnesota Comprehensive Assessments (MCAs) assess math and reading skills in grades 3-8, reading in grade 10, and math in grade 11. Students also take a science MCA in grades 5 and 8 and one high school grade.
The Minnesota Department of Education has effectively managed its outside testing vendors.
The ACCESS for English Language Learners assesses students identified as English learners on English proficiency from grades K-12. Students take four ACCESS tests: listening, speaking, reading, and writing. Schools may use alternate tests instead of the MCAs and the ACCESS tests for students with severe cognitive disabilities.
MDE funds its testing work using a combination of state and federal sources. Federal funds constitute a little more than one-third of revenue in most years. MDE spent $19.2 million developing, distributing, and maintaining tests in Fiscal Year 2016. For Fiscal Year 2016, the Legislature appropriated $11.2 million for statewide testing that meets federal requirements, compared with $16.9 million in Fiscal Year 2015 and $16 million in Fiscal Year 2014.
Federal legislation passed in 2015 altered some testing requirements, but left others unchanged.
The Every Student Succeeds Act (ESSA) was passed by Congress in 2015. It requires states to set statewide academic standards and assess students’ performance in meeting those standards.
Compared to previous law, ESSA gives states greater discretion to intervene when a school’s students do not perform well on standardized tests. Additionally, ESSA provides states with some new options for student testing.
MDE is currently developing a state plan to meet ESSA’s requirements. Some of ESSA’s changes will be challenging to implement. For example, schools may incur penalties for not testing 95 percent of eligible students, but they must also allow parents and guardians to refuse testing for their children if permitted by state law. Minnesota allows parents to refuse tests for their children.
Overall, MDE has appropriately selected and monitored its outside testing vendors.
MDE uses outside vendors to develop, distribute, and maintain its standardized tests. MDE carefully selected its current MCA vendor using a competitive process and monitors the company’s performance. MDE does not competitively select a vendor for the ACCESS tests because Minnesota belongs to a consortium of states and territories that collaborate on English language proficiency tests.
Although MDE’s vendor selection and oversight process was sound, the department does not systematically assess how well its vendors serve local stakeholders. MDE can do a better job gathering information from school districts and charter schools about their experiences with the state’s vendors.
Administering statewide tests creates challenges for school districts and charter schools.
School districts and charter schools must administer the state’s standardized tests. Doing so can create logistical, staffing, and equipment problems that affect instruction and cost money.
Students take the tests on computers, but some schools have limited computer resources. Some must shuttle students in and out of computer labs for weeks in order to complete testing. Students not being tested are often unable to use computers for learning on testing days.
Schools and districts may have to divert staff from other duties to assist with testing. Students receiving special education or English language instruction are often particularly affected while specialist teachers are managing testing for other students.
These impacts can occur for long periods of time. Over half of Minnesota’s schools spent more than 15 days (or three weeks) on MCA testing in 2016. Over 300 schools spent 25 or more days (five weeks). Schools with many English learners spent additional days administering the ACCESS.
More than half of principals and teachers responding to a survey felt unprepared to interpret key test score data.
Students varied widely in the amount of time they spent taking standardized tests, in part because some tests take longer than others. For example, students spent much longer taking the seventh- and eighth-grade math MCAs than the fifth-grade science MCA. English learners spent more time completing the MCAs than other students, and they had to take ACCESS tests as well.
Testing also costs schools money. In a survey, 83 percent of local testing administrators who responded said their school districts or charter schools had bought computing equipment in the last three years to administer state-required tests. Nearly one in five reported hiring extra staff to assist with test administration or test score analysis.
MDE does not collect data about the local impacts of testing that would allow decision makers to consider the effects of proposed policy changes. To provide better information for MDE’s own decision making and valuable context for the Legislature, MDE should work with local stakeholders to develop reporting mechanisms that track local costs and impacts.
Many local administrators and teachers do not feel confident interpreting test score data.
MDE reports several scores for each of Minnesota’s statewide tests. For example, a seventh-grade reading MCA score report includes, in part, (1) a proficiency score indicating whether the student met state standards; (2) a growth score indicating whether the student improved over the past year at the same rate as other students; and (3) a career and college readiness progress score, showing whether the student’s current performance puts the student “on track” to eventually be ready for college-level work.
We surveyed teachers and principals across the state. Many said they found standardized test scores at least somewhat useful. For example, 85 percent of principals and 77 percent of teachers offering an opinion said they found MCA scores very useful or somewhat useful for identifying achievement gaps between groups of students.
However, many also reported that they did not feel prepared to interpret the scores provided by MDE. Over half of the principals and teachers who responded to our survey said that they did not feel prepared to analyze the MCA growth scores MDE uses most frequently. Even more felt unprepared to use the career and college readiness progress scores. Nearly one-third of teachers said they did not feel prepared to interpret MCA scores overall.
Many teachers and administrators also expressed a lack of familiarity with ACCESS scores, even those who worked with English learners. Nearly 60 percent of teachers who reported having English learners in their classrooms said they did not receive ACCESS scores for their students or did not recall receiving them.
MDE provides some assistance to local educators to improve their understanding and use of test scores, and the department has recently added a position to do further outreach. MDE also targets additional training resources to schools with the lowest-performing students.
Nonetheless, our conversations with administrators and teachers indicate a statewide need for more support. MDE should further increase outreach and training regarding the use of test scores at the local level.
Many principals and teachers prefer locally adopted tests to Minnesota’s statewide tests.
Most Minnesota school districts and charter schools administer both statewide standardized tests and other tests adopted locally. The locally adopted tests are frequently designed to provide immediate information to assist teachers in adjusting classroom instruction to fit student needs.
Some legislative requirements intended to improve testing have had unintended consequences.
Legislators have required MDE to add components to the MCAs to make them more like the popular locally adopted tests. However, teachers and principals still find locally adopted tests useful more often than they find the MCAs and the ACCESS tests useful.
At present, it is probably not possible to use a single test that provides both helpful ongoing information to educators and meets federal requirements promoting school and district accountability. Tests designed for one purpose do not necessarily serve other purposes equally well.
Some standardized testing laws have lengthened tests and required MDE to report scores that have a high level of uncertainty.
The Legislature has required MDE to develop tests and report test scores in certain ways. Some of these requirements are ill-advised.
State law requires that the MCAs include questions above and below a student’s grade level. However, due to federal requirements, MDE has been unable to use these questions in calculating most of the test scores it reports. As a result, statewide tests have been lengthened for all students without much benefit.
State law also requires MDE to report a score based on the MCA describing each student’s progress toward career and college readiness. But such scores for elementary and middle school students are methodologically problematic. Projections extending far into the future have a high level of uncertainty, and some of them are likely to be wrong.
The Legislature should remove or reconsider these requirements and instead focus on setting priorities for MDE’s testing program.
Summary of Agency Response
In a letter dated March 2, 2017, Minnesota Department of Education Commissioner Brenda Cassellius agreed with the report’s key recommendations and called the report “thorough and fair.” She stated that collecting more information on testing costs and impacts from school districts and charter schools “should not only inform MDE implementation but also help inform decisions made by policy makers as well.” Commissioner Cassellius also said the department would increase its outreach to school districts and charter schools, writing that “in order for schools to successfully use data analytics to inform instruction and interventions, ongoing professional development is critical.” She committed to working with the Legislature to “identify and decrease areas of statute that may prescribe test designs and reporting formats.”
The Program Evaluation Division was directed to conduct this study by the Legislative Audit Commission in March 2016. For a copy of the full report, entitled "Standardized Student Testing," 110 pp., published in March 2017, please call 651/296-4708, e-mail Legislative.Auditor@state.mn.us, write to Office of the Legislative Auditor, Room 140, 658 Cedar St., St. Paul, MN 55155, or go to the Web page featuring the report. Staff who worked on this project were David Kirchner (project manager), Caitlin Badger, and Catherine Reed.