Higher Ed Survey
What Tests and Cut Scores are Used for Entry-Level College Course Placement?
The higher education survey was conducted to gather information from a nationally representative sample of postsecondary education institutions regarding the tests and cut scores used to determine entry-level students' need for remedial/developmental instruction in reading and mathematics. This information is intended to provide a source of nationally representative data as validity evidence for statements in NAEP reports about 12th grade students’ academic preparedness for college.
Several of the tests most widely used for admissions and placement, such as the ACT, ACCUPLACER, COMPASS and SAT, are compared with NAEP in other parts of the preparedness research program. For example, the results from this survey are examined in light of the statistical linking studies between NAEP and the other tests. Ultimately, the goal is to identify the score on NAEP that represents the knowledge and skills needed to qualify for freshman credit-bearing courses without remediation.
- Key Conclusions
Higher Ed Survey Methodology
An important premise for the design of this survey study is that the cut scores used by postsecondary education institutions to determine student need for remedial or developmental instruction represent each sampled institution’s conception of "just academically prepared." Survey respondents were asked to report the tests used and the score at or above which entry-level students would be deemed "just academically prepared" for credit-bearing courses, and below which students would be deemed to need remedial instruction in reading and mathematics.
Between August and January 2012, survey questionnaires were completed by a nationally representative sample of 2-year and 4-year postsecondary education institutions. The weighted response rate was 81 percent (about 1,340 institutions) of the 1,560 eligible institutions from the original sample of 1,670. To be included in the study, institutions had to provide data on the cut-scores used for placement of:
- Students enrolled in an undergraduate degree program in the liberal arts and sciences (applies to 4-year institutions).
- Students in a degree program designed for transfer to a 4-year institution (applies to 2-year institutions).
To produce national estimates, analyses included procedures for evaluating nonresponse bias and appropriately weighting the data.
Higher Ed Survey Key Conclusions
- Overall, 71 percent of postsecondary education institutions reported using some mathematics test and 53 percent some reading test in evaluating student need for remediation in those two subject domains.
- Other than test scores from national standardized admissions and placement tests, no other criterion (e.g., high school grade point average, high school courses taken, end-of-course tests, etc.) was used by more than 11 percent of institutions overall. This relatively low level of use aligns with the study design premise that cut scores on the national standardized admissions and placement tests represent postsecondary education institutions’ conception of “just academically prepared.”
- The study examined variability in the cut scores reported for the tests used by institutions for determining student need for remedial/developmental instruction, looking at the distance (in standard deviation units) between the 25th percentile and the 75th percentile in reported cut scores on the respective tests. Variability ranged from 0.15 standard deviation units to 1.35 standard deviation units. This variability suggests that postsecondary education institutions across the nation do not hold a common conception of “just academically prepared.”
- The most frequently used standardized national mathematics tests were the ACT (23%); COMPASS Algebra test (20%); SAT (17%); and ACCUPLACER Elementary Algebra test (16%). The remaining ACCUPLACER and COMPASS mathematics tests were each used by 5% or less of the institutions. Twenty-two percent of respondents reported using other mathematics tests than the standardized national tests for which data were collected.
- Comparing 2-year and 4-year institutions, the mean cut scores for the ACT, SAT, and COMPASS Algebra tests were higher for 2-year institutions while the mean for the ACCUPLACER Elementary Algebra test was higher for 4-year institutions.
- The most frequently used standardized national reading tests were the COMPASS Reading (22%); ACCUPLACER Reading Comprehension (19%); ACT Reading (16%); SAT Critical Reading (11%); and ASSET Reading Skills (9%). Ten percent of postsecondary education institutions reported using other reading tests than the standardized national tests for which data were collected.
- Comparing 2-year and 4-year institutions, the mean cut scores for the ACT and SAT reading tests were higher for 2-year institutions. For the other reading tests, there were no statistically significant differences when comparing the mean cut-scores set by 2-year and by 4-year institutions.
The overall mean cut scores reported for national mathematics tests were:
- 19 on ACT mathematics on a scale of 1 to 36
- 471 on SAT mathematics on a scale of 200 to 800
- 70 on ACCUPLACER Elementary Algebra on a scale of 20 to 120
- 57 on ACCUPLACER College-Level Mathematics on a scale of 20 to 120
- 49 on COMPASS Algebra on a scale of 1 to 99
- 43 on COMPASS College Algebra on a scale of 1 to 99
The overall mean cut scores reported for national reading tests were:
- 18 on ACT Reading on a scale of 1 to 36
- 456 on SAT Critical Reading on a scale of 200 to 800
- 76 on ACCUPLACER Reading Comprehension on a scale of 20 to 120
- 41 on ASSET Reading Skills on a scale of 23 to 55
- 76 on COMPASS Reading on a scale of 1 to 99