High-Stakes Assessment in Undergraduate and Postgraduate Studies

Abstract

High-stakes assessment is playing an increasingly important role in higher education at the undergraduate, graduate, and postgraduate levels. Such assessments are sometimes used for group purposes—to assess how well a university is doing in educating its students—and other times for purposes of evaluating individuals. High-stakes assessment at the undergraduate level generally involves assessments of learning and reasoning at the end of the college experience. Sometimes, pretests are also given to compare cognitive skills before and after the college experience. There are several different approaches to measuring learning and performance outcomes: (a) standardized instruments and inventories; (b) indirect methods that focus on students’ perceptions of learning and engagement; (c) authentic performance-based methods, such as portfolios; and (d) locally designed tests and inventories. Each of these methods of assessment has different advantages as well as disadvantages. For example, standardized tests are normed, and thus it is possible to compare the performance of students at, say, one university to those at another. But standardized tests also measure outcomes that some scholars feel are less meaningful than the outcomes measured by other kinds of assessments. Indirect measures, such as of student engagement, look at students’ level of engagement with college but tell less about cognitive gains than some other kinds of measures. Performance-based measures such as portfolios have the advantage of measuring outcomes presumably relevant to each individual student; they are harder to score than some other kinds of measures, however, and they do not lend themselves readily to comparisons across colleges and universities. Homemade tests produced by individual institutions can be tailored to the goals of those institutions but generally lack the standardization and generality of some other kinds of measures. Assessments of graduate and postgraduate students are of a different ilk. Generally, graduate, postgraduate, and hiring institutions are looking for presumed research and teaching competence. Publication records as well as letters of recommendation serve as primary bases for evaluating students going onto the job market. It is possible to entertain more sophisticated measures than just counting publications, such as various measures based on citations in the scholarly literature.