Rising Test Scores in Connecticut School Districts Related to the Exclusion of Students with Disabilities

Ω

This report finds that the exclusion of thousands of students with disabilities from reported Connecticut Mastery Test results has distorted reported trends in test scores. Following test scores from year to year in the same grade, the study finds that statewide improvements in standard Connecticut Mastery Test (CMT) scores reported by the Connecticut State Department of Education (SDE) between 2008 and 2009 -- the period of the largest reported gains -- were largely the result of the exclusion of students with disabilities from these standard test results, rather than overall improvements in performance. For example, 84% of the reported improvement in 4th grade math proficiency between 2008 and 2009 and 69% of the improvement in 8th grade reading proficiency could be attributed to the exclusion of these students. Much of the reported improvements in later years could also be attributed to this exclusion, though there were some modest overall gains as well.

In 2009, state and federal policy changes enabled school districts to offer a modified assessment (MAS) to students with disabilities that the districts determined would not have passed the CMT in math and/or reading. As a result of these policy changes, the share of students taking the regular CMT declined substantially. Prior to 2009, students who did not reach the proficient level on the CMT because of their disabilities were included in statewide CMT results. In 2009, thousands of low-scoring students were assigned to take the MAS test instead of the standard CMT, and these students were not included in the CMT results. Thus, CMT scores reported by the State Department of Education appeared to improve in large part because these low-scoring students were no longer included in the calculations.


The analysis by Connecticut Voices for Children, a research-based think tank, focused on 4th and 8th grade CMT scores and also finds:

· Across school districts, there was a very strong correlation between a declining percentage of students taking the standard CMT and increasing percentages of students reported as reaching proficient scores – ie, as more students with disabilities were excluded from score results, more students were reported as reaching the proficiency level.

· If students with disabilities that took the modified assessment in reading and math in 2009 to 2011 were included in the total sample of 4th and 8th grade test takers, then the revised percentages of students at or above the proficient level would be two to three percentage points lower than the state reported. There were some modest improvements in scores, but not to the degree that the state initially reported.

· The percentage of students who were assigned to take the MAS rather than the standard CMT varied substantially across school districts – from 0% to 12.8%. Most districts had some participation on the modified assessment that affected their test score data.

To ensure more valid, “apples to apples” comparisons of trends over time, Connecticut Voices recommends that state officials clarify the impact of the exclusion of students with disabilities when reporting on changes in CMT scores over time. Further, the organization suggests that policymakers:
· use a variety of indicators, not just standardized test scores, to evaluate improvements in public education; and
· reconsider policies that assign rewards and punishments based on these test scores.

The report includes district-level data on 4th and 8th grade CMT scores, the percentage of local students who took the CMT and MAS, and a recalculation of district test scores that includes students with disabilities (see appendices F and G for local 2011 data).

Ω