Short Notes - APA PsycNET - American Psychological Association

6 downloads 1115 Views 414KB Size Report
College of Business Administration, University of Oklahoma ... Pfeffer's conclusion that a graduate degree in business is particularly useful for persons not ...
Copyright 1985 by the American Psychological Association, Inc. 0021-9010/85/S00.75

Journal of Applied Psychology 1985, Vol. 70, No. 4, 769-773

Short Notes

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Generalizability of MBA Degree and Socioeconomic Effects on Business School Graduates' Salaries George F. Dreher

Thomas W. Dougherty

School of Business University of Kansas

Department of Management University of Missouri-Columbia

Bill Whitely College of Business Administration, University of Oklahoma

This study tested the generalizability of findings reported by Pfeffer (1977a). The determinants of starting and current salaries for 314 graduates from the business schools of three large state universities were examined. The results support Pfeffer's conclusion that a graduate degree in business is particularly useful for persons not coming from the highest socioeconomic backgrounds. A variety of factors identified by Pfeffer were controlled for, and the possession of the MBA degree was positively related to starting salary irrespective of socioeconomic origin but was positively related to current salary only for those not coming from upper-middle and upper-class backgrounds. Finally, current salary sex differences only were observed for persons from upper-class socioeconomic backgrounds. This was attributable to a salary advantage possessed by upper middle- and upperclass men. In this note we present findings from a replicaStudies of the managerial and professional staffing process have been largely based on the principle tion of a study conducted by Pfeffer (1977a) that that accession to organizational positions is deter- focused on the joint effects of a graduate education mined by ability, achievement, and contribution in business and socioeconomic origin on business to the organization. However, organizational re- school graduates' salaries. Whereas other studies searchers have long recognized that positions and have followed the careers of graduates from Master's their associated rewards are often allocated ac- in business administration (MBA) programs (e.g., cording to ascriptive criteria (e.g., Coates & Pel- Harrell, 1969; Weinstein & Srinivasan, 1974) only legrin, 1957; Dalton, 1951; Quinn, Tabor, & Gor- Pfeffer (1977a) has simultaneously studied the don, 1968). Such criteria as sex, age, social back- effects of the MBA and socioeconomic origin on ground, personal connections, lodge membership, salary attainment. His sample was drawn from a and marital status have been found useful in listing of business school alumni from the MBA understanding managerial stratification. If criteria and bachelor's programs of a large, prestigious such as social origin or social connections are state university. A random sample of 20 graduates central to the managerial succession process, failure per year (from each degree program) for the years to take these criteria into account when designing 1960-1974 was drawn, with data collected from or evaluating selection systems or training programs both a mail survey and student records. There was for managers diverts attention from the processes a significant effect for the MBA degree on starting actually operating within organizations (Pfeffer, but not current salaries and no effects on either starting or current salaries for such variables as 1977b). grade-point average or admission test scores. In This investigation was supported by the University of addition, Pfeffer observed that the current salary Kansas General Research allocation #3127-20-0038 and effect for socioeconomic origin was larger for the by the University of Kansas School of Business Research bachelor's subsample and concluded that a graduate Fund provided by the Joyce C. Hall Faculty Development education in business "is an important avenue for Fund. Additional support was provided by the Ponder social mobility, particularly for those persons not Faculty Development and Research Fund, College of coming from the highest socioeconomic backBusiness and Public Administration, University of Missouri-Columbia. The ideas and opinions expressed herein grounds" (p. 704). Pfeffer's (1977a) results are important because are solely those of the authors. Requests for reprints should be sent to George F. they suggest that the graduate business school Dreher, School of Business, University of Kansas, Law- experience substitutes for membership in a social group. Although Pfeffer's findings have equal emrence, Kansas 66045. 769

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

770

SHORT NOTES

ployment opportunity, promotion system, and individual career planning implications, they require considerable interpretive caution. Two issues require particular attention. First, Pfeffer's conclusions regarding social class were based on observed subgroup regression weight differences. He noted that the effect of social class on current compensation was larger for those with bachelor's degrees, but the statistical significance of the difference between standardized regression weights of .14 and .28 was not reported. Essentially, Pfeffer was interpreting an instance of single group validity. A more appropriate approach would include a moderated regression analysis testing for the unique contribution of the social class by degree level interaction term. The second issue worthy of additional attention concerns the generalizability of Pfeffer's (1977a) results. His study was conducted with a group of graduates from a highly ranked, prestigious state university. In addition, his sample contained, on average, graduates of the late 1960s and consisted almost completely of men. Given the unique reputation of the university's graduate programs, the quality of admitted students, and the substantial labor market changes characterizing the decade of the 1970s, there is need for caution when generalizing to other, more recent business school graduates. The purpose of the present study is to test the generalizability of the Pfeffer (1977a) findings. The study reports on a recent group of graduates from three large state universities. Two are located in the Midwest, and the third is located in the Southwest. The central research question addresses whether educational level (an undergraduate vs. graduate degree in business) and socioeconomic origin interact in accounting for variance in both starting and current salaries. Pfeffer argued that the effects of socioeconomic origin should be most observable as the individual's career progresses. Therefore, following Pfeffer, it is hypothesized that socioeconomic origin and educational level will interact in accounting for current salary variance but not starting salary variance. The positive effects of a graduate education in business should be greatest for individuals not coming from the highest socioeconomic backgrounds. Also, given a sufficiently large number of female graduates, it was possible to examine the data for salary differences between men and women. Method Criterion

starting salary was measured in different years, price inflation for this measure was a problem. To remove the inflation effect the starting salary figure was divided by the gross national product price deflator, with 1972 = 1.00 (Economic Report of the President, 1983, p. 166). Sample Using alumni records at each university, business school graduates from the bachelor's programs were randomly sampled from the classes of 1978 and 1979. All MBA program graduates from the same years were included in the initial sample. Sample members received a questionnaire and a stamped return envelope. Questionnaires were coded so that survey and student record information could be merged. Respondent anonymity was guaranteed, and respondents were informed that they would receive comparative salary data on graduates from the three universities. Using a follow-up procedure, 486 questionnaires were returned along with permission to access student files. This represented a response rate of 52%. The returns were essentially the same between universities with equal response rates from the bachelor's and MBA samples. The data were collected during July and August of 1983. The entire respondent group averaged 29 years of age at the time of response and included 126 women. The analysis sample consisted of 314 individuals (34% of the full mail sample) who were employed on a full time basis at the time of data collection and who had not completed an additional degree subsequent to earning either their MBA or bachelor's degree. All regression analyses were performed on individuals with complete data on all analysis variables. Predictors The predictor and control variables used in the study are the same used by Pfeffer (1977a). The questions and scales were taken directly from the appropriate sections of his questionnaire. He provides sufficient arguments as to why certain control variables are required and why variables are operationalized as follows: (a) previous work experience was the number of years of work experience completed before entering either degree program; (b) socioeconomic origin was the respondent's selfrated family social class and included the following coding categories: 1 = lower class, 2 = lower middle class, 3 = middle class, 4 = upper middle class, and 5 = upper class;1 (c) degree earned was coded

1 Following Pfeffer (1977a), annual salary, both As discussed by Pfeffer (1977a), father's educational earned currently and starting salary upon grad- level should be related to self-reported socioeconomic uation, was used as a criterion measure. Because origins. However, parental education is not regarded as

771

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

SHORT NOTES

0 for bachelor's degree graduates and 1 for the MBA degree; (d) -high school was coded with a value of 0 if the respondent went to a public school and 1 if a graduate of a private or parochial school; (e) number of years worked was computed by summing the number of years worked prior to entering the degree program and the elapsed number of years since obtaining the degree; (f) selfemployment was coded a value of 0 if the respondent was not self-employed and 1 if the respondent was self-employed; (g) line or staff was coded as a 0 for line positions and 1 for staff positions; (h) year was the year graduated.2 Also, sex was coded with a 0 when the respondent was a man and with a 1 when the respondent was a women. Finally, grade-point averages at both the undergraduate and graduate level were obtained from student files as were Graduate Management Admission Test scores as available.

Table 1 Regression Results for Determinants of Starting Salary Variable Previous work experience Year Socioeconomic origin High school

M

SD

weights

2.15 78.42

3.70

.13* -.09

3.20

.84 .33 .45 .50

.13 .28 .48

Sex Degree earned n R Adjusted R2

.73

.02 .04 -.15** .40**

314 .49 .23

*p