Description of Changes
Normally, growth percentiles are calculated for all students with test scores for the same subject from 1 or more prior years/grade levels. To calculate SGPs, students are grouped with their "academic peers" (i.e., students who scored similarly in previous years/grades). Students are then assigned percentile ranks describing how their current-year scores compare to those of their academic peers. This percentile rank typically characterizes students' growth relative to peers from the same cohort over a 1-year period.
Gaps in standardized test administration and participation were introduced as part of the COVID response. This in turn affects how student growth percentiles (SGPs) and adequate growth percentiles (AGPs) are calculated for years following the pandemic. At a high-level, researchers should presume that academic year (AY) 2021 growth percentiles for MSTEP aren’t usable. Growth percentiles for 2022 can be useful for research, but have a slightly different interpretation than in pre-COVID years. English Language Learner (WIDA) assessments are all usable, but should be interpreted differently in 2022 than in other years.
For MSTEP, there was no growth in 2020 because MSTEP was not administered. The student growth percentile data are present for AY 2021 but only because the US Department of Education required it. Administration of MSTEP was not compulsory and was observed at 77% participation rate. However, it is noted that non-participation is not random.
To calculate growth percentiles for AY 2021 while missing a year of assessment data (2020) the spring 2019 and spring 2021 scores were used. Specifically, an expectation of the 2020 growth was built by referencing the spring 2017 to spring 2019 growth. Because of concerns with the representativeness of 2021 assessment data the Spring 2022 growth did not reference spring 2021 data. It instead uses spring 2019 to spring 2022 growth. Therefore only have 6th-8th and 11th grades have growth because no 2019 data exists for other cohorts. Student growth percentile calculations for 2023 are largely back to normal as scores are now post-COVID. One difference is that PSAT for grade 10 will be used MSTEP grade 11 when calculating student growth percentiles. This allows for the use of post-COVID scores for this cohort and will be a permanent change going forward.
Student growth percentiles for the English Language Learner (AKA WIDA) assessment scores were less affected. In AY 2020, WIDA assessments were largely completed before the pandemic. Therefore, AY 2021 growth percentiles reference AY 2020. However, in AY 2022 growth percentiles were calculated using AY 2020 data as test taking in AY 2020 was not compulsory.
Summary by assessment & year
| Assessment | Year(s) | Grades | Base Norms | Growth Period |
| MSTEP/PSAT | Pre-Covid | 4-8 | Current Cohort | 1 year |
| MSTEP/PSAT | 2020 | None * | N/A * | N/A * |
| MSTEP/PSAT | 2021 | 5-8 * | Prior Cohort * | 2 years * |
| MSTEP/PSAT | 2022 | 6-8 * | Prior Cohort * | 3 years* |
| MSTEP/PSAT | 2023+ | 4-8 | Current Cohort | 1 year |
| SAT | Pre-Covid | 11 | Current Cohort | 3 years |
| SAT | 2020 | None * | N/A * | N/A * |
| SAT | 2021 | 11 | Prior Cohort * | 3 years |
| SAT | 2022 | 11 | Prior Cohort * | 3 years |
| SAT | 2023+ | 11 | Current Cohort | 1 year * |
| WIDA | Pre-Covid | 1-12 | Current Cohort | 1 year |
| WIDA | 2020 | 1-12 | Current Cohort | 1 year |
| WIDA | 2021 | 1-12 | Current Cohort | 1 year |
| WIDA | 2022 | 3-12 * | Prior Cohort * | 3 years * |
| WIDA | 2023+ | 1-12 | Current Cohort | 1 year |
(* = different from pre-covid)
Additional Resources
https://www.michigan.gov/mde/-/media/Project/Websites/mde/Year/2017/05/02/SGP_Policy_Brief.pdf