The Reliability of Shorter Assessments in New Jersey for Group-Level Inferences

Education policymakers must balance the reliability of assessments to measure academic knowledge and skills with the burdens that assessments place upon students, teachers, and schools. In 2019, New Jersey began using the New Jersey Student Learning Assessments (NJSLA), shorter assessments based on the Partnership for Assessment of Readiness for College and Careers (PARCC). Regional Educational Laboratory researchers examined the reliability of test results for the NJSLA by comparing results at the school, test, and subgroup levels from 2016 to 2019. The findings indicated a high degree of reliability across most measures the researchers examined; during the transition to the NJSLA, the reliability did not decrease for any test results--except the Algebra 2 test--reported by the New Jersey Department of Education. The instability of the Algebra 2 results was most likely not attributable to changes in the assessment but instead to changes in the student population that was required to take the test following a change in the state's testing requirements.