Work Experience As A Predictor Of Mba Performance
Arthur J. Adams
Students bringing post-baccalaureate work in business to an MBA program may have gained experiences that will help them succeed in their course work. This study investigates the relation between years of such work experience and success in the MBA program, as measured by grade point average. Using a sample of 269 recent graduates from an urban university’s MBA program, we use ANOVA and correlation analysis to show that amount of work experience prior to enrolling in the program is a predictor of success. Surprisingly, work experience is found to be more associated with success than GMAT score or undergraduate grade point average, the typical strong correlates of graduate grade point average found in the literature. A possible admissions policy change based on this finding is discussed.
Quite a few years ago, far too many of the students beginning our university’s MBA program had the same profile: male, undergraduate degree in business, a product of our own bachelor’s degree program, and no work experience. Fortunately, we now have a much more diverse group of MBA students. Although more than half of current MBAs do bring a business background to the program, there is gender balance, and about two-thirds of the students get their undergraduate degree elsewhere. In fact, over ten percent are of foreign nationality. The proportion of MBA students with meaningful work experience has risen as well.
As professors, we have often had the occasion to appreciate having MBA students in class with work experience who can relate concepts and situations discussed in class to their current or past place of employment. This type of exchange clearly benefits the entire class, perhaps especially those students who may be wondering why class time is being devoted to a particular topic. We take it for granted that work experience will have a positive effect on a student’s MBA program experience. Indeed, many highly respected MBA programs encourage or require interested students to get work experience prior to applying for admission. Accordingly, we should view work experience as a possible predictor of success in the MBA program, not unlike undergraduate grade point average (UGPA) or score obtained on the GMAT exam.
There have been a few studies that in one way or another try to deal with the perceived importance of prior work experience for MBA students. For instance Sobol (1984) develops a variable called “scale” that is used along with UGPA and GMAT to forecast graduate grade point average (GGPA) at Southern Methodist University. Students can get points on the scale variable in six different ways, one of which is work experience. (Others include such things as recommendations and undergraduate activities.) The scale variable is found to be significant ([Alpha] = .05) in several models. However, given the half dozen factors that make up the variable, the effect of work experience per se is unknowable; it is confounded with the other factors. Graham (1991) finds that only GMAT of ten predictors is significantly related to GGPA. Noting that since the R: is low (.1677; typical for such studies), Graham observes that the door is open for admission committees to factor in qualitative assessments of a student’s potential, such as writing samples, interviews, and work experience. In similar studies (Paolillo, 1982; Deckro and Woundenburg, 1977) the authors obtain a comparable value for [R.sup.2] and likewise suggest using a variety of qualitative variables in the admission process.
Some researchers have downplayed the merits of work experience between the bachelor’s degree and the MBA program as an admission criterion. For instance, a study (Fisher and Resnick, 1990) of first year MBA students at Baruch College considers only UGPA and GMAT as predictors of GGPA, noting that other factors such as work experience lack evidence of validity. Cooksey and Rindfuss (1994) note that MBA programs offer a contrast to the wisdom that prevails in other fields about the benefits of “continuous” schooling (See, for example, Fitzgerald, Jurs, & Hudson; 1996). While acknowledging that business experience gained after a bachelor’s program can make for a better subsequent MBA outcome, the authors also note that such students may also have taken on other responsibilities that will compete for their time and financial resources. Using the National Longitudinal Study of the High School Class of 1972 as a data base, they define success as completing t. he MBA program enrolled in; no information was available on quality of school or grades attained. They report that the interruption of schooling between the BA and the MBA leads to a lower probability of completing the degree. This finding is interesting in that it would not be detected by the great majority of studies that investigate correlates of GGPA. This is because such studies are typically done after the fact; i.e., the subjects have already completed the MBA degree.
Given the lack of research into the work experience — grade point relation, the main purposes of the present study are: (1) To investigate the role of work experience as it relates to GGPA in our MBA program and (2) To assess its relative explanatory power compared to more traditional predictors of GGPA.
Subjects and Procedure
A test sample of 269 (120 women and 149 men) was drawn from those recently receiving MBAs at the University of Louisville College of Business and Public Administration. The program has roughly 400 students in it at any one time, and is internationally accredited by the AACSB. About 40% of the sample were full-time students while in the program; about 60% were part-time.
GMAT score is the primary admissions determinant for the program although UGPA is considered for marginal applicants. UGPA plays a secondary role in the admission decision since no standard exists for meaningful comparisons across undergraduate institutions and majors. Nonetheless UGPA is viewed as a measure of motivation and discipline.
For each subject, GGPA — the MBA grade point average at graduation — was recorded along with UGPA, GMAT, and other potential predictors of GGPA. The work experience variable was actually measured by a proxy — years between the completion of the bachelor’s degree and the first semester in the MBA program.
One-factor analysis of variance (ANOVA) was used to investigate the relationship between GGPA and length of work experience. Three categories of work experience were defined; none or one year; two or three years; and four or more years. Correlations with GGPA were also utilized to see the relative strengths of our predictors.
Figure 1 shows the distribution of the work experience variable for our data set. The most common value was two years of experience; about sixty percent of the students began their MBA program with three or less years of work experience while about forty percent had four or more years.
[Figure 1 ILLUSTRATION OMITTED]
Table 1 gives sample statistics for GGPA broken out according to the three categories of work experience mentioned above. The table shows that those with one or no years of work experience have the lowest GGPAs; about a tenth of a grade point below the group with a moderate amount of work experience. The most experienced group is about another tenth of a grade point higher than the middle group. These differences may not appear dramatic, but given the compressed nature of grades given in a typical MBA program (mostly As and Bs), they are noticeable differences. The difference in GGPA between the least and most experienced groups is almost a full standard deviation.
Table 1 Sample GGPA Statistics According to Years of Experience
Years of Experience
Statistic 0 and 1 2 and 3 4 or more
Sample Average 3.45 3.55 3.64
Standard Deviation .28 .21 .23
Number of Cases 79 77 113
In Table 2 we show the results of conducting an ANOVA on the GGPAs for the three groups. The resulting value of F has a very small p-value, which denotes that the differences observed are extremely unlikely to have occurred by chance. The average effect of four or more years experience is an increase in GGPA of about .10 relative to the mean; the average effect of no experience or one year is a decrease of about .10 relative to the mean.
Table 2 ANOVA Summary Table for Years of Experience
Source SS df MS F p-value
Experience 1.30 2 .65 11.30 <.001
Error 15.3 266 0.575
Total 16.6 268
We now look at correlations to measure the association of GGPA with potential predictors used in our study, and to assess the relative strength of work experience as an explanatory variable. Table 3 reports the correlations of GGPA with a set of ten variables. In general, these values are similar in magnitude to those in many other studies reported in the literature. (Given our sample size, correlations above about .13 are significant at the .05 level.) What is perhaps most interesting in Table 3 is that the correlation for years between completion of the undergraduate degree and start of the MBA program is the highest value in the table, at r = .244. The two variables most commonly found in the literature to strongly relate to GGPA — score on the GMAT and undergraduate grade point average — are significantly correlated but less so than the work experience variable.
Table 3 Correlation Results for GGPA
Variable with GGPA
Undergraduate Grade Point Average (UGPA) .181
Score on the GMAT .208
Score on the Verbal Section of the GMAT .146
Score on the Quantitative Section of the GMAT .171
Years Between Graduation and Start of MBA Program .244
Gender (Female = 0, Male = 1) -.046
Status (Part-time = 0, Full-time = 1) -.119
Undergraduate Major (Nonbusiness = 0, Business = 1) -.147
Origin of Bachelor’s Degree (Internal = 0, External = 1) .127
Country of Origin (Foreign = 0, U.S. = 1) -.064
The data also suggest that work or life experience prior to an undergraduate degree is quite different from experience gained afterward. Many non-traditional students matriculate into MBA programs relatively late in their careers but perhaps just shortly after receiving undergraduate degrees. They may bring a long history of pre-bachelor graduation work experience. However the correlation between GGPA and student age upon starting the MBA is negligible (r = 0.08). The significant relation between GGPA and the years experience variable appears less a function of general experience than the nature of opportunities that only arise after receiving a bachelor’s degree.
These findings have led to a proposed policy revision in MBA admissions. Some faculty have been making the argument that our program is best suited for students who do have some business experience, and that we should admit fewer students coming to the MBA program directly from their undergraduate degree program. The proposed change is to require a minimum 3.5 UGPA for a student wanting to enter the program immediately after the undergraduate degree; a 3.4 UGPA if one year of business experience; a 3.3 if two years experience, etc. The previous conditions all assume a suitable GMAT score.
Having found a significant relation between GGPA and work experience, we then investigated the possibility of interacting variables, and found that an interaction variable for gender with years of experience had a correlation with GGPA of .156, which is significant at ([Alpha] = .05). Given the coding for our gender variable, this means that while the years of experience variable is significant for both genders collectively, the effect is stronger for males than it is for females.
This article identifies years of work experience between the BA and MBA as a potential predictor of success in an MBA program. For a sample of recent graduates we find this factor to be more strongly related to GGPA than more commonly seen variables such as UGPA and GMAT.
As with any study, care should be taken in extrapolating the results. The sample was drawn exclusively from a population of students that had successfully completed the MBA degree. Accordingly it makes no suggestion as to the possible performance of those denied admission because of a low GMAT score or those who started but did not complete the program. Further, the results were obtained at an urban institution that has perhaps more of an emphasis on entrepreneurship than many other schools. Further research may investigate whether work experience is a factor at similar and dissimilar institutions.
Cooksey, E. C., & Rindfuss, R. R. (1994). Prior activities and progress in MBA programs. Research in Higher Education 35, 647-667.
Deckro, R. F., & Woundenberg, H. W. (1977). M.B.A. admission criteria and academic success. Decision Sciences 8, 765-769.
Fisher, J. B., & Resnick, D. A. (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, 137-148.
Fitzgerald, S. M. , Jurs, S. J., & Hudson, L. M. (1996). A model predicting statistics achievement among graduate students. College Student Journal 30, 361-366.
Graham, L. D. (1991). Predicting academic success of students in a master of business administration program. Educational and Psychological Measurement 51,721-727.
Paolillo, J. G. P. (1982). The predictive validity of selected admissions variables relative to grade point average earned in a master of business administration program. Educational and Psychological Measurement 42, 1163-1167.
Sobol, M. G. (1984). GPA, GMAT, AND SCALE: A method for quantification of admissions criteria. Research in Higher Education 20, 77-88.
ARTHUR J. ADAMS TERENCE HANCOCK College of Business & Public Administration University of Louisville Louisville, KY 40292
COPYRIGHT 2000 Project Innovation (Alabama)
COPYRIGHT 2000 Gale Group