COVER STORY: Making The Cut
Photo Courtesy Long Beach City College
C O V E R S T O R Y
Making The Cut
Colleges, State Re-Examine Placement Tests
By Paul Bradley
Consider the cut score, that all-encompassing, essential number that determines the future of legions of community college students.
Score above the cut score on a standardized placement test and proceed to college-level course work, greatly enhancing the chances of eventually earning a college degree.
Earn a score below the cut line and get a ticket to one or more developmental courses, a place sometimes dubbed the Bermuda Triangle of higher education — the place where students go in, but never come out. Only a tiny percentage of students who take remedial courses ever finish college.
That high-stakes nature of placement tests employed by community colleges across the country is among the factors driving a fundamental re-examination of the exams, raising questions about whether they create a serious impediment to the college completion agenda.
An emerging body of research indicates that standardized placement tests are poor predictors of college success and that a student’s high school transcript does a far better job of telling a college where a beginning student belongs.
That hypothesis is now being tested by Long Beach City College in California, which just admitted a cohort of nearly 1,000 students whose placements were determined not by a placement test, but by their high school grades.
“I don’t think that tests are the evil here,” said college President Eloy Oakley. “The way we have used the tests are the problem. We have leaned on the placement tests almost totally to place students. I don’t think you can rely just on a test to judge a student’s capacity to succeed.”
Last month, Oakley was among the college officials who greeted the new cohort of students, who are graduates of schools in the Long Beach Unified School District, which has had a long partnership with the college. The students, part of a project called Promise Pathways, enjoyed a picnic lunch, met with counselors and learned about the expectations of college life.
For LBCC, the challenge is particularly acute. More than 90 percent of incoming students typically test into one or more developmental courses. Worse, a study conducted by the college found that “the average number of semesters of development coursework our students are required to complete reached 5.6 semesters in fall 2011. The amount of remediation required significantly increases the time needed to complete educational goals and for far too many students becomes an insurmountable barrier, dissuading many from completing.”
The study, which examined a five-year cohort of more than 6,000 students, also found a wide gap between the results of placement tests and the performance of students in high school classrooms. More than 60 percent of students placed in remedial English classes had earned an A or B in their senior year English class. At the same time, one-third of students placed in college-level English courses had received a D or an F in high school English.
“Taken together, these results suggest the possibility of a dramatic misalignment between the measures we commonly used to assess and place students and those most likely to predict performance in our classrooms,” the study found. “Moreover, increasing the alignment between how we assess students and how they are likely to perform in our courses may have dramatic potential for both our students, by aligning their placements more closely with their zone of proximal development, and for our instructors in the classroom.”
The findings buttress those of the Community College Research Center at Teachers College, Columbia University. A study the CCRC released earlier this year found that up to one-third of students assigned to remedial courses through placement tests are improperly assigned, and that high school transcripts are a better indicator of college readiness.
Under LBCC’s new initiative, 52 percent of cohort students will be taking college-level English, compared to just 11 percent during the first semester last year. The college predicts that 571 Promise Pathways students will complete transfer-level English in their first year of college, compared to 101 Long Beach high school graduates who were admitted in fall 2010.
The idea is to start students at a higher level and boost their chances at success by preventing them from languishing in developmental education classes. Hispanics and blacks, who are disproportionately assigned to developmental education classes, are expected to benefit from the initiative.
Oakley said the approach is not a pilot program, but a permanent change. It is being closely watched in California; the office of California’s community college chancellor is studying whether high school transcripts and grade point averages should be made part of placement decisions at all of the state’s 112 community colleges. A report is due in October.
“The program will be reviewed, and we’ll take a close look at the effect to see if we can scale it up to all freshmen,” Oakley said. “We hope to find many thousands of students are succeeding in their first year, getting through college English and math. And I hope we can close some the achievement gaps that affect students of color.”
LBCC’s work is the most vivid example of efforts to boost student outcomes by improving assessment and placement at community colleges. But it’s not the only one. A broader movement is taking hold across the country to embrace a more holistic approach to determining student placements in developmental education courses.
Some states are rewriting tests to better align with curriculum. Others are keeping the tests, but combining them with other data such as high school grades.
In Connecticut, for example, a new state law will drastically reduce remedial placements in 2014. Only the most severely unprepared students will be able to take a remedial course, and it will only last one semester. For other students, developmental education will be embedded into college-level coursework.
Austin Community College in Texas is evaluating student essays to refine placements for students at or near the cutoff score. Both Florida and Virginia are developing customized placement tests that are closely tied to curriculum.
The movement is described in a new report by Jobs for the Future, a Boston-based nonprofit that studies education and workforce issues. Titled “Where to Begin? The Evolving Role of Placement Exams for Students Starting College.”
The report describes a “new narrative” that questions many of the assumptions that have long shaped thinking on student readiness. This new narrative is based on five key elements, the report said:
Placement tests are high-stakes exams that could affect a student’s entire educational trajectory.
The effectiveness of traditional developmental courses is unclear. They may hinder rather than help a student’s educational progress.
Accelerating students through and out of development classes, condensing their time there, leads to better outcomes.
Placement exams are weak predictors of success in gateway courses. High school transcripts do a better job.
Math and English assessments provide a narrow picture of student readiness, not measuring things such as motivation, persistence and critical thinking.
While states and colleges are looking at different approaches, one thing is clear: the one-size-fits-all approach to assess and place students, as embodied on standardized tests, is quickly falling out of favor.
“The states are at the forefront of this movement,” said Gretchen Schmitt, program director for post-secondary state policy at Jobs for the Future. “Different states and different colleges are taking different approaches. It’s part of a more holistic approach to the student success initiative.”
Schmitt said the movement is being fueled by technological advances which have made data more accessible and useful than ever before.
“It’s not just the access to data,” she said. “It’s also the ability to use it, to get it into the hands of practitioners. It’s not just the institutional research people who have the data now. The data is no good unless you can use it.”
Oakley hopes the data that LBCC will develop will advance the completion agenda.
“Our goal is to get more students to finish,” he said. “We want to close those achievement gaps that have been so stubbornly persistent. Developmental English is such a huge barrier to our students.”
Even as colleges implement new assessments, more research is needed, Schmitt said. Researchers should examine whether the customized assessments are better in predicting student success than off-the-shelf diagnostics. They also need to study whether informing students of the high-stakes nature of the test makes a difference in test scores and predictive value.
She foresees a day, perhaps within five years, when tests are just part of a larger assessment of a student’s readiness.
“Some kind of assessment will be needed to determine where a student has deficiencies,” she said. “But the tests won’t be all or nothing. They’ll be part of a larger process.”
It’s YOUR TURN: CCW wants to hear from you!
Q: Do standardized placement tests represent a serious impediment to the college completion agenda?
Share your Comments: ccweekblog