Academic journal article The Journal of Negro Education

Assessment and Retention of Black Students in Higher Education

Academic journal article The Journal of Negro Education

Assessment and Retention of Black Students in Higher Education

Article excerpt

Over the past 60 years, considerable research has been devoted to assessing characteristics of college students. The two primary purposes that have driven these investigations are (1) a search for accurate methods to identify students who are likely to experience problems in college, and (2) the search to develop valid and powerful means to predict dropout. It is fair to say that, for both purposes, the research has largely produced ineffective methodologies and particularly questionable applications for predicting the performance of Black students. Where limited successes have been achieved, replications have failed. These research efforts are selectively condensed and reported in this article. Additionally, a summary of assessment instruments that offer data potentially useful for making admissions decisions is presented. These instruments were examined to identify devices that show sufficient predictive validity to justify their use as part of a beginning freshman assessment program. Lastly, assessment is related to retention programs.

The primary purpose of this article is to identify outcomes that an assessment program oriented to Black students in higher education can reasonably be expected to produce. Thus, the decisions about which purposes and procedures should be implemented are not addressed. The intent here is to explore components of a database upon which retention decisions and actions may be developed.

THE ASSESSMENT TREADMILL

In the mid-1970s, educational psychologists led an exciting crusade that held great promise to revolutionize educational practice. Based on the concept of learning as an individual and unique ability influenced by a constellation of aptitudes, a technology was proposed through which individualized instruction could become a reality. In many ways, this quest was the culmination of the observations of teachers and parents for thousands of years. This technology was based upon the understanding that learners differed and that these differences mattered. Thus, if teachers and parents could respond to all learners in accord with their special learning styles, achievement could be made more accessible and teaching more efficient. This idea of an interaction between learner aptitudes or characteristics and instructional treatments or methods was so powerful and intuitively appealing that it virtually consumed the intellectual candlepower of the time, generating both research and practice.

However, that which burned so brightly did not burn long. With the publication of Aptitudes and Instructional Methods (Cronbach & Snow, 1977), the evidence was clear that aptitude-treatment interactions (ATIs) would not be the beacon that would lead to education equality. Several consistent findings contributed to the diminished promise of ATIs. First, most interactions were quite difficult to establish and, when established, proved to be unstable. Second, the search for important aptitudes for which effective treatments could be developed showed little promise. In many cases, one treatment was about as effective as any other. Finally, time after time, a single aptitude emerged as the central variable that explained the vast majority of known variance. This variable was ability. Classically, the ability aptitude has been called "intelligence," and when controlled for, other aptitudes rarely account for a sufficient amount of variance to warrant developing special treatments.

The literature on predicting college success is also replete with attempts to predict academic performance. Three of these attempts will be discussed here, not because they are models to emulate, but because they illustrate consistent findings. These results, from separate and independent sources, lead to essentially a single conclusion: that prediction of future performance is technically possible but practically unfeasible.

The first is a long tradition of attempts to measure study abilities. Beginning in the 1930s, psychologists, counselors, and learning improvement specialists began developing devices purported to assess study skills in some form (Sherman, 1990, 1991). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.