Giving Compass' Take:

• A report by Dan Cullinan and Erika B. Lewy concludes that multiple measures assessments more accurately predict student capabilities and result in better achievement outcomes than standard college-course placement tests.

• What are the equity implications of high-pressure exams that catapult some students into advanced classes while requiring others to take non-credit-bearing developmental lessons? How can you support assessment reform that provides a more nuanced picture of students' capabilities and college readiness?

• Read about American high-schoolers' college preparedness.

Colleges have many choices of measures and typically select them based on their predictive validity, availability, ease of use, and cost. There is strong empirical evidence that the high school grade point average (GPA) is one of the best available predictors of student success in college. Unlike a one-time assessment, the GPA is an aggregate measure of performance over multiple years and reflects not only students’ content knowledge, but also the types of behavior that influence success in college, such as attendance and participation. Box 1 presents various options for determining course placement using multiple measures of college readiness.

MDRC and CCRC collaborated to evaluate the impact of two MMA models—one based on placement formulas at seven 2-year state colleges in New York and another based on decision bands at four 2-year state colleges in Minnesota. The research team conducted a randomized controlled trial at the study colleges in each state, assigning incoming students in the study sample to either a control group, whose members were placed using traditional placement testing, or a program group, whose members were placed using MMA.

The study found that students who are placed into college-level courses using MMA are more likely to complete gatekeeper courses—basic introductory or prerequisite college-level courses—than their counterparts who are placed into developmental courses using placement tests. These findings held for placements in math and English courses in the first semester and after three semesters. The study also found that student success rates can improve when MMA is applied and students who would have otherwise been placed into developmental education courses are instead referred to college-level courses. Other key findings from the study include the following:

  • In the first semester, students who were “bumped up” using MMA—that is, placed into a college-level course when they would have otherwise been referred to developmental education using a traditional placement method—were more likely to enroll in and complete this college-level course than similar students in the control group.
  • By the end of the third semester, students who were bumped up to college-level courses using MMA were more likely to have completed their gatekeeper courses than their counterparts in the control group.
  • However, students who were “bumped down” using MMA—that is, placed into a developmental education course when they would have otherwise been referred to a college-level one—experienced a negative effect similar in magnitude to the positive effect experienced by students who were bumped up.