Perceptions of international students toward GRE

College Student Journal, June, 2005 by Emily E. Mupinga, Davison M. Mupinga

Graduate Record Examination (GRE) is an aptitude test thought to reflect intelligence or the capacity to learn and is used by colleges for admission of students into graduate school. No matter bow well written, no test can measure all aspects of intelligence, and this makes GRE inherently inadequate. But, in what respects is the test inadequate? This study sought to establish the perceptions of international students toward GRE, specifically, issues with the content & context; structure, and purpose of the exam. Data for this qualitative study were collected through interviews of seven students admitted into different graduate programs at Louisiana State University. Mainly the content & context of the exam were found to be biased against international students. Recommendations include not using GRE scores as a sole predictor of success, but other factors such as undergraduate GPA, special talents, significant accomplishments, recommendations, interviews, and other holistic judgments.

**********

The Graduate Record Examination (GRE) is an aptitude test, thought to reflect intelligence or the capacity to learn (Larsen & Buss, 2003). It is a standardized admission exam designed to predict performance in graduate school through verbal, quantitative, and analytical reasoning questions. The GRE Board encourages graduate schools, departments, and fellowship selection committees to consider GRE scores as a meaningful source of information about an applicant's chance for success in graduate school (Educational Testing Service, 2002). The exam is developed and administered by the Educational Testing Service (ETS).

Nowadays, the GRE is a computer adaptive test (CAT) and students take the test on a computer. When taking the test, students move to the next question only after responding to the one on the screen. According to American Graduate Education (2003), the questions increase in difficulty based on [one's] ability to answer the question before it correctly. The computer selects the appropriate questions based on one's answers and the test will continue this way until one has achieved the required mix of concepts and question types. Each of the three sections (verbal, quantitative, and comprehension) tested are scored within a range of 200 to 800.

The majority of universities in the United States require applicants to take at least the general GRE test for admission into graduate programs in the Humanities, Sciences, Social Sciences, and Arts (Norcross, Hanych & Terranova, 1996). Most institutions utilize the GRE in a "top-down" selection approach employing minimum standards (cut-offs) for consideration (Oltman & Hartnett, 1985). Moreover, the GRE tends to be the most heavily weighted criterion used by selection committees (Chernyshenko & Ones, 1999; Ingram 1983). According to Larsen and Buss (2004), meta-analyses reveal that GRE scores predict success in graduate school, but correlations are only modest (.15 to .40). The authors provide four arguments why GRE scores can be useful: even small inches in predictability above chance can be useful; costs of failing to select the right people into graduate school can be high; useful if used with appropriate criterion (what one wants to predict); and research has shown that without range restriction, correlations between GRE scores and success in graduate school can be as high as .30 to .70.

While no test is said to measure all aspects of intelligence, and so any admission test [SAT included], no matter how well written is inherently inadequate (CollegeDegreeGuide, 2003). What then should colleges use for admission purposes? Some form of admission testing is necessary as it would be unfair, creates problems, and misleading to accept students into graduate school purely on their grades. Hence, many graduate schools require applicants to take the GRE test, and somehow use the scores for admission purposes. While the exam is commonly used by most colleges, it is inherently inadequate. This study sought to establish problems potential graduate students have with the test, particularly international students whose cultures and education systems are different.

Problem Statement and Purpose of Study

The assessment of graduate school performance is complex, both conceptually and technically (Hartnett, 1979; Crampton, 2000), and the overall evaluation system seems to be culturally biased and unsound to most test-takers, particularly international students. Bias has been defined as a situation whereby a group of examinees always performs more poorly on a test and this group does not subsequently exhibit inferior achievement (Sternberg & Williams, 1997). Most international students do not get a good score on the verbal section of the GRE especially those with English as a second, third, or even fourth language. From an international student's perspective, it would seem unfair that the GRE Board and US graduate school committees expect international students to have the same degree of English language proficiency as native English-speakers, and to have GRE scores that are not significantly different from native speakers.