An examination of the admission criteria for the MBA programs: A case study

Admission to the Masters of Business Administration (MBA) programs at most universities which are accredited by the American Assembly of Collegiate Schools of Business (AACSB) is largely based on prior undergraduate grade point averages (GPA) and Graduate Management Admissions Test (GMAT) scores. Is GMAT score an adequate measure of potential success in these programs and, therefore, an acceptable criterion for admission? The purpose of this paper is to determine whether GMAT scores and MBA success (measured by graduate GPA) are correlated. In addition, other factors which may affect the success of MBA students are discussed in this paper.

To investigate these research questions, an MBA program at an accredited business school in the southeast of the United States was chosen for in-depth case study. Bivariate regression models were developed. Data analysis was performed on all variables to determine their significance in predicting graduate GPA and MBA program success. Results are presented and suggestions for consideration of other relevant factors are made.

The current admissions process at the University of Tennessee at Chattanooga (UTC) like that of many other AACSB accredited graduate business programs, relies heavily on Graduate Management Admission Test (GMAT) scores as predictor of success. To be admitted into the MBA program at UTC, prospective students must complete standard admission forms and submit their GMAT scores. Students are admitted into the MBA program by one of two methods, the index system Dr the petition process.

In the index system of acceptance, a point index for each prospective student is computed. To calculate the index, the applicant's undergraduate GPA (based on a 4.0 scale) is multiplied by a factor of 200. This number is added to the applicant' GMAT score producing a total index score. All applicants meeting the minimum score of 950 are admitted. A second index is calculated if the prospective student's score is less than 950. This second index is based on the GPA of the last 60 undergraduate credit hours (typically representing major business courses). This 60 hour based GPA is then multiplied by 200 and added to the GMAT score. If the total index is at least 1000 points the applicant is admitted.

If the prospective student meets neither requirement, he or she may appeal the admissions process by petitioning the Graduate Council. The Graduate Council, at their discretion, accepts or denies the petition. Only a very small percentage (less than 10%) are admitted to the program through petition.

The benefits of this study will allow the directors of MBA programs to reassess their admission requirements and make policy changes if necessary. By improving admission criteria and standards, more promising candidates will be admitted to the program, leading to a higher quality of MBA graduates.

Background of Standardized Testing

Standardized testing has been utilized for over a century. Providing many functions, testing can be used to sort people into groups, classify and rank employees, or admit students into educational programs. Tests can be designed to measure aptitude, personality, achievement, or even competency. There are many advantages to standardized testing; however, standardized testing does have drawbacks.

Some business schools accept the Miller's Analogies Test (MAT). In a Pepperdine study, Graham (1991) found a significant relationship between MAT scores and graduate GPA, but the correlation was not as strong as that found between GMAT scores and graduate GPA, particularly when these scores were combined with undergraduate GPA. The Graduate Management Admission Test was first administered in 1954 by the Educational Testing Service in conjunction with the Graduate Management Admission Council. The GMAT scores range from 200 to 800 points, with a mean of 500 and a standard deviation of 100.

The GMAT and MBA Admission

The GMAT is designed to measure the ability and knowledge of the student. The test is all multiple-choice questions covering two sections, verbal and quantitative. The score a prospective student earns usually determines whether he or she is admitted into a school or university. Other factors such as grade point average may also be considered. Each school or university has its own policy and requirements.

In a study (Edwards, 1990) done to gather information concerning MBA programs, 657 accredited institutions were surveyed. Out of the 333 responses received, all but one of the responding schools used some sort of pre-admission testing. Johns Hopkins does not require candidates to submit a pre-admission test score such as GMAT. Some other graduate programs including Standford, Boston College, and University of Indiana at Bloomington require submission of GMAT score for consideration, but do not mandate a minimum acceptable score. The most popular decision method for most MBA programs is an index system based on GMAT scores and undergraduate GPA's. From the responding schools, 177 respondents used an index, 81 imposed a minimum test score in addition to applying an index, and the rest used a minimum test score without applying an index.

Biases of Standardized Testing

Standardized test are not without their problems. Since the tests are objective, multiple-choice type questions, they do not accurately measure an individual's ability, critical thinking skills, or competency. Because of the way the tests are constructed, they ignore some knowledge and performance expected from students, and they place test takers in a passive, reactive role, rather than engage their capacities to structure tasks, generate ideas, and solve problems (Darling-Hammond, 1991).

Another problem arises from the fact that these tests are not constructed by the educational system. Testing in the U.S. is primarily controlled by commercial publishers and non-scholastic agencies (Darling-Hammond, 1991). Their tests are designed to assess and rank students inexpensively and efficiently. Furthermore, these tests are not set to schools' curriculums and, therefore, do not necessarily measure students' potential success in a specific MBA program. Some critics argue that the nonprofit Educational Testing Service, which administers well known test such as the SAT and GMAT, perpetuate the use of invalid and biased tests to assure that only those from wealthy families are admitted to the most prestigious colleges (Kaplan, 1982).

Admission Criteria for the MBA

Programs

A study by Sobol (1984) used multiple-regression analysis to establish a scale for admitting graduate-school candidates. The scale included the following factors, in addition to GPA and GMAT scores; involvement in campus activities (as an undergraduate), work experience, letters of recommendation, and personal goals. Such scoring is necessarily subjective, however.

Benson (1983) used the Pearson Correlation to evaluate the relationship between GMAT scores and graduate level GPA. He found no significant relationship between them. Newman (1986) conducted a study of those graduate students initially accepted into the Dowling College MBA program: GMAT score plus Undergraduate GPA times 200 > 950 and those provisionally accepted, total score

There are some studies, however, that claim GMAT is a better indicator of academic success than other variables (Graham, 1991; Youngblood and Martin, 1982; Deckro and Woundenberg, 1977). A multiple-regression study by Fisher and Rensik (1990) indicated strong support for the value of GMAT scores in predicting firstyear grades at the graduate level. Youngblood and Martin (1982), Paolillo (1982), and Sobol (1984) support the predictive ability of GMAT scores when combined with undergraduate GPA.

Methodology

The sample for this study consists of 279 students currently enrolled in the accredited MBA program at UTC. The dependent variable is the graduate GPA (on a 4.0 scale) earned in the MBA program. Seven independent variables are tested for their ability to predict graduate student success in the MBA program: (1) GMAT score, (2) age of the student, (3) gender, (4) marital status, (5) race of the student, (6) undergraduate GPA, and (7) undergraduate major.

To verify a correlation between the GPA and any of the seven variables listed above, a bivariate linear regression was developed for each of the independent predictor variables. Analysis on each independent variable was performed to determine if one variable was a better indicator of success than another. Findings

A general summary of the sample characteristics is shown in Table 1. The average age of the MBA student at UT is 29.8 years. Of the 279 MBA students sampled, 61.3% fall within the 25 to 35 year old age category. Of the students samples, 55.9% were male. Approximately 85% of the students were Caucasian. Only 2.5% were African American and the remaining 12.2% represented other ethnicities.

From the students surveyed, 60.2% had undergraduate GPA's under 3.000. The average undergraduate GPA was 2.77. The majority of students (71.9%) hold a B.S. degree in business administration. Approximately 86% of them had GMAT scores over 400. Forty-two percent had scores higher than 500. However, 14% had GMAT scores under 400. The average GMAT score was 483.

Bivariate Regression Analysis

Bivariate linear regression models were developed to analyze the correlation between the dependent variable, graduate GPA, and each independent variable to determine whether or not he independent variable was significant in predicting success. Table 2 shows the results of seven regression analysis in the descending order of Pearson correlation coeffient.

Undergraduate GPA

Undergraduate GPA had the highest correlation coefficient, 0.521, meaning 27.1% (R2) of the variability in graduate GPA is explained by the undergraduate GPA. The T-test shows that undergraduate GPA is a significant variable to predict success.

GMAT Score

The correlation coefficient for GMAT score is 0.433. This explains 18.75% of the variability in graduate GPA. Statistically, this correlation is significant at 0.05 level. Index Score

The correlation coefficient for index is 0.356, meaning 12.67% of the variability in graduate GPA is explained by the index score. The T-test shows that the index score is significant, indicating that combination of undergraduate GPA and GMAT scores are statistically significant variables to predict graduate GPA.

Age

The correlation coefficient for age is 0.276. This means that 7.62% of the variability in graduate GPA is explained by age. Even though the correlation coefficient was very small, age was a statistically significant variable in the model for predicting success.

Race

The correlation coefficient for race is 0.111 and the T-test shows that race is not a significant variable in predicting GPA.

Graduate Major

The T-test shows that graduate major is not significant in predicting success and has a correlation coefficient of only .071. Gender The correlation coefficient for gender is -0.041. The T-test shows that gender is not a significant variable to predict success. Even though for of the independent variables were statistically significant, the correlation between the graduate GPA and the independent variables is small and has little predictive ability. Although the GMAT score showed a statistically significant correlation with the graduate GPA, its practical value as a predictor variable is not so great. Our findings have some similarities to the past studies. In Graham's study (1991), he found that GMAT scores had the highest correlation coefficient of 0.41 and undergraduate GPA the second highest with 0.22 at a significance level of 0.05.

Paolillo's studies (1982) found correlation coefficients of undergraduate GPA of 0.35 and GMAT score of 0.26 at a significance level of 0.001. He also found hours attended to be significant at a level of 0.01 with a correlation coefficient of 0.18. However, the three variables only accounted for 19% of the variability in the dependent variable, graduate GPA.

Our results, along with past studies, suggest that other additional criteria may ned to be used in selecting prospective students. It appears that undergraduate GPA and GMAT scores are significant variables in predicting success, but due to their low predictive ability other forms of assessment such as writing samples, interviews, work experience, or other non-quantitative measures or assessments would be useful.

Conclusions and Recommendations

The current admissions process at most AACSB accredited business schools relies heavily on GMAT scores and undergraduate GPA's as indicators of future success in the MBA programs. However, the results of our study suggest that GMAT scores and previous GPA's may not adequately predict success and may unfairly deny admission to some qualified students. Other studies agree that admissions and Allina, using a bench marking format, examined the admissions and testing policies at seven top graduate schools, including Harvard, MIT, and Johns Hopkins. Each of the schools examined in this study was found to be moving away from heavy reliance on standardized tests as sa primary admission criterion. Since a good portion of the variance is left unexplained by the quantitative variables used in the studies, we feel that qualitative measures are needed to more accurately predict academic success. Schools should reconsider their admissions policies, incorporating such requirements as writing samples, personal interviews, or preferences. This would allow schools to admit more qualified students into the MBA programs and raise the school's overall academic standing.

References

Allina, A. (1987). Beyond Standardized Tests: Admissions Alternatives that Work. Research paper. Cambridge: National Center for Fair and Open Testing.

Benson, G. (1983). GMAT: Fact or Fiction - A Look at the Validity of the Exam Research paper presented at the Rocky Mountain Educational Research Association; Tucson, AZ, November 2-5.

Darling-Hammond, L. (1991). The Implications of Testing Policy for Quality and Equality. Phi Delta Kappa, 73, 220-225. Deckro, R.F. and Woudensberg, H.W. (1977). MBA Admissions Criteria and Academic Success. Decision Sciences, 8, 765-769. Edwards, D.E. and Brannen, D.E. (1990). Standardized Testing and Graduate Business School Admission: A Review of Issues and an analysis of a Baruch College MBA Cohort. College and University, 65:2, 721-727. Graham, L.D. (1991). Predicting Academic Success of Students in a Masters of Business Administration Program. Educational and Psychological Measurement 51, 721-727. Hudis, P. and Stolzenberg, R. (1987). Recent Trends in Characteristics of Graduate Management Admission Test Takers. Princeton: Graduate Management Admission Council.

Madaus, G.F. (1991). The Effects of Important Tests on Students. Phi Delta Kappa, 73, 226231.

MacGlaughlin, S.D. (1992). Closing the Gender Gap. The College Board Review, 36, 18-25. Newman, J. (1986). A Comparison of the Rate of Success in the Dowling Master of Business Administration Program. Practicum. New York: Nova University.

Paolillo, J. (1982). The Predictive Validity of Selected Admissions Variables Relative to Grade Point Average Earned in a Master of Business Administration Program. Education and Psychological Measurement, 42, 11631167.

Sobol, M. (1984). GPA, GMAT, and Scale: A Method for Quantification of Admission Criteria. Research in Higher Education. 20, 77-88.

Youngblood, S.A. and Martin, B.J. (1982). Ability Testing and Graduate Admissions: Decision Process Modeling and Validation. Educational and Psychological Measurement, 42, 1153-1161.

MOHAMMAD AHMADI, FARHAD RAISZADEH, MARILYN HELMS Professors School of Business Administration, University of Tennessee at Chattanooga Chattanooga, TN 37403

Copyright Project Innovation Summer 1997
Provided by ProQuest Information and Learning Company. All rights Reserved