Greek-letter membership and college graduation: does race matter?

Severtis, Ronald E., Jr. ; Christie-Mitzell, C. Andre

Research, utilizing a nationally representative sample of 3,712 Americans, revealed that Greek-letter membership increases the probability of college graduation more for African Americans than for European Americans. Conversely, father's education is a more robust predictor of educational outcomes for European Americans compared to their African American counterparts.

Keywords: race, college graduation, Greek-letter membership, African Americans, educational outcomes

**********

College graduation rates for African American students lag behind their European American counterparts (U.S Census Bureau, 2000). Of the U.S. population 25 years of age and older, a little over twenty-eight percent of non-Hispanic European Americans have earned a Bachelor's degree, compared to roughly sixteen percent of non-Hispanic African Americans (U.S. Census Bureau, 2000). Such racial disparities in educational achievement are often associated with differences in background socioeconomic status (Duncan, Brooks-Gunn, & Klebanov, 1994; Corcoran, 1995; Hochschild, 2003). Research indicates that racial minorities are disproportionately more likely to be from families with few economic resources to invest in their schooling and to live in neighborhoods with failing schools (Corcoran, 1995; Kao & Thompson, 2003). The result is that minorities are less likely than whites to perform well in secondary schooling, and hence, face greater obstacles to achieving success in college. Nevertheless, even holding constant economic indicators, African Americans are still less likely to obtain a college degree compared to their European American counterparts (Corcoran, 1995; Kalmijn & Kraaykamp, 1996; Smith, 1989). Therefore, in addition to observing socioeconomic resources, researchers have found it useful to consider other routes to educational success, including the use of social networks or social capital (Coleman, 1990; Parcel & Dufur, 2001). In this paper, we assess whether membership in Greek social organizations may be one route that African Americans may use to boost the probability that they will graduate from college.

Some research has reported that membership in Greek organizations increases graduation rates from six to nine percent above non-members (Astin, 1985). What is unknown is whether the impact of the benefits of fraternity and sorority life varies by race. Given the disadvantage that African Americans face with regard to educational attainment, pinpointing whether and how membership in Greek-letter organizations may contribute to educational success for this group is an important endeavor. There is reason to believe that such associations may have an impact on the college graduation rates of African American students. For example, Whipple, Baier and Grady (1991) found that while European American Greek members have parents with higher income and education that African American Greek members have higher levels of academic motivation. Such motivation in combination with the social networks developed through membership may very well allow African Americans to remain in college and graduate (Whipple et al., 1991).

In exploring the relationships among educational attainment, membership in Greek-letter organizations, and race, this research improves on prior research in at least two important ways. First, the majority of studies on Greek-letter organizations has focused on predominantly white samples and has been unable to investigate racial differences that may be associated with fraternity and sorority life. Second, many studies on the effects of Greek-letter organization membership on various outcomes have been restricted to small, localized samples (for an exception see Pike, 2003). While these studies with smaller samples are able to elucidate the impact of Greek life at specific institutions (see e.g., Samter, 1992), the findings may not be generalizable to the larger population of Greek-letter organizations. To address these two issues, we employed nationally representative data of the U.S. population that offers enough racial variation to make racial comparisons. Further, by virtue of the representativeness of the sample, the respondents' undergraduate institutions included a wide range of private, public, small and large educational institutions.

Background

Social capital theory guides this research. Social capital is often developed through membership in organizations and resides in relationships among actors that "... are useful for the cognitive and social development of [youth] ..." (Coleman, 1990, p.300). Coleman (1988, 1990) in his well-known theory of social capital outlines three types of resources (or capital) that may be invested in individuals and that encourage educational attainment: 1) human capital (i.e., years of schooling and cognitive ability); 2) financial capital (i.e., income and other economic resources used to pay for education); and 3) social capital (i.e., relationships developed in families and organizations that are useful for social and academic development). While this research focuses on Greek-letter organizations as sources of social capital, the analysis below controls for both human and financial capital. All three types of capital are positively correlated (Parcel & Dufur, 2001). Individuals with higher human and financial capital have greater access to higher levels of social capital.

Greek-letter fraternities and sororities are voluntary organizations. The ways in which such organizations operate and maintain their membership base builds collectively-owned social capital (Bourdieu, 1986; Brewer, 2003; Foley & Edwards, 1999). That is, the social capital intrinsic to Greek-letter organizations benefits both the individual and the collective. The individual may gain resources through this association that makes possible achievement that otherwise would not happen. The organization is then strengthened by its members' accomplishments and their ability to generate more social capital and resources that will maintain the organization. These groups close their social networks to non-members and create high levels of obligation and trust within the group. These dynamics lead to more robust connections among members and loyalty to the organization (Putnam, 1995, 2000).

Perhaps the primary expression of social capital in Greek-letter organizations is increased involvement in the college community (Lounsbury & DeNeul, 1995; Requena, 2003). For instance, research indicates that living in Greek-letter housing increases involvement in both academic and extracurricular activities (Lounsbury & DeNeul, 1995; Pike, 2003). The building of these social ties across college entities appears to play a positive role in educational progress (Astin, 1985).

Does membership in Greek-letter organizations in and of itself provide the necessary social capital to propel academic achievement and college completion? Surprisingly few contemporary research studies have addressed this specific question (for an exception see Whipple et al., 1991). Indeed, early research on the academic performance of Greek members yielded mixed findings. For example, Bradshaw and Kahoe (1967) found that Greek freshmen had a significantly lower mean grade point average the semester after they joined their organization when compared to freshman non-members (see also Pugh & Chamberlain, 1977 or Warman, 1962). Buckner (1961) and Crookston (1960) discovered that there were no significant differences between Greek members and nonmembers on grade point average, while Sherron (1969) found that Greek members had significantly higher mean freshman grade point average than non-members. Further, Magoon and Maxwell (1965) found that eighteen percent more Greeks than non-members were academically dismissed or placed on probation over a span of one semester.

Despite the mixed results of early research, there are ways that this body of research can be reconciled with more contemporary research that shows positive relationships between Greek-letter social organizations and educational attainment (e.g., Whipple et al., 1991). First, much of the research prior to the 1980s utilized fairly small non-representative samples, making it difficult to generalize to Greek and non-Greek populations in the United States. Second, the focus of research on Greek-letter organizations has often been on the negative outcomes associated with fraternity and sorority life, including hazing, racial insensitivity, alcohol abuse, sexual aggression, and academic dishonesty (Arnold & Kuh, 1992; Bryan & Schwartz, 1983; Ehrhart & Sandler, 1992; Engs & Hanson, 1988; Gonzalez & Broughton, 1986; Kuh, 1991). Therefore, those chapters of Greek-letter organizations which have engaged in these behaviors have been studied, and the academic achievement in these chapters may simply not mirror that of those chapters where such behavior is lower or absent. Third, even if members of these organizations have lower academic achievement, they may still be graduating at rates higher than the general college population. Research beginning in the early to mid-1980s indicates that membership in Greek-letter organizations is positively related to satisfaction with the college experience and college graduation, despite its equivocal relationship to actual academic performance (Arnold & Kuh, 1992; Astin, 1985; Bryan & Schwartz, 1983). Moreover, upon graduation Marmaros and Sacerdote (2002) found that fraternity and sorority members are able to use their social networks to obtain higher-paying jobs compared to non-members.

Race and Educational Attainment

In the United States, racial minorities have lower educational attainment compared to European Americans (U.S. Census Bureau, 2000). With specific regard to the groups evaluated in this paper, African Americans have lower educational attainment than European Americans (Cross & Slater, 2001). Estimates show that the gap in college graduation rates between European American and African American students is nearly two to one (Kalmijn & Kraaykamp, 1996). Further, African American females graduate at higher rates than their male counterparts, but at rates still below those of European Americans (U.S. Census Bureau, 2000; Cross & Slater, 2001). 16.4 percent of African American males and 16.8 percent of African American females have earned a Bachelor's degree, compared to 30.8 percent of European American males and 25.5 percent of European American females (U.S. Census Bureau, 2000).

Classic and contemporary studies have relied heavily on parental characteristics to explain the "black-white" gap in educational outcomes (Blau & Duncan, 1967; Corcoran, 1995; Duncan, Brooks-Gunn, & Klebanov, 1994; Duncan, Featherman, & Duncan 1972; Jencks et al., 1972; Haveman, Wolfe, & Spaulding, 1991; Hill & Duncan, 1987; Sewell & Hauser, 1975; Hauser & Featherman, 1976). Research indicates that one reason why African Americans have attained less education is that they are more likely to be reared in homes with lower levels of human capital (i.e., parents' education) and financial capital (i.e., household income) (Mizell, 1999). In addition to affording the economic resources to pay for education, the home life provided by well-educated and high income parents may support and encourage the child more than those home environments that are economically deprived (Parcel & Menaghan, 1994; Mau, 1995). In this respect, parents may be able to transmit social class, which includes educational attainment, to their progeny (Kohn, 1969).

Similar to parental socioeconomic status, family structure and region of origin have been important background factors in educational research on African Americans. A disproportionate number of African American children are reared in single parent households, which are related to lower educational attainment (Duncan, Brooks-Gunn, & Klebanov, 1994). Children from two-parent homes have greater access to social capital (Parcel & Dufur, 2001). Compared to single parents, two-parents are able to develop and maintain parent-child social bonds by spending time with their children, monitoring their children's activities, and paying attention to the general welfare of their children. These efforts help children internalize appropriate social norms related to educational achievement. Similar to single-parent homes, growing up in the Southern region of the U.S. has been found to generate lower educational attainment (Hogan & Pazul, 1982; Jensen, Findeis, & Wang, 2000). The culture, economic structure and greater number of impoverished African Americans traditionally concentrated in the South are all factors that prior research indicates contribute to lower educational achievement, contrasted to those living in other regions (Glasmeier & Leichenko, 1999; Hogan & Pazul, 1982; Lieberson, 1980; Mizell, 1999).

In this study, we hypothesize that membership will be positively related to the probability of college graduation. An important component of this research is to investigate whether the process shaping educational attainment for African Americans differs compared to European Americans. Given the documented educational differences between African Americans and European Americans, we test whether race moderates the relationship between Greek-letter organization membership and the probability of college graduation. Although we anticipate that all college students will benefit from Greek-letter membership, we hypothesize that given their economic disadvantage that African Americans will benefit more from Greek-letter membership, compared to their European American counterparts. In other words, because in general European Americans have greater resources at their disposal to guarantee their educational progress, the importance of the social capital generated through membership in Greek social organizations will be more important to the achievement of African Americans. We further explore whether race conditions the impact of sociodemographic variables (e.g., gender and region of origin), human capital, and financial capital on educational outcomes.

Data and Measures

To explore the relationship between Greek-letter organization membership and the probability of college graduation, we utilized data from the General Social Survey (GSS). The GSS is conducted by the National Opinion Research Center and is a national probability sample of non-institutionalized, English-speaking U.S. citizens over 18 years of age. Data for the GSS have been collected from 1972 to 2002. The survey was administered annually beginning in 1972 and biennially after 1994 (Davis & Smith, 1998). In any given year, the response rate for the study ranges from 70 to 82 percent (Davis & Smith, 1998).

Narrowing the Sample. The GSS question assessing Greek-letter organization membership, our main explanatory variable of interest, was administered in selected years only: 1974, 1975, 1977, 1978, 1980, 1983, 1984, 1986, 1987, 1988, 1989, 1990, 1991, 1993, and 1994. Therefore, we were limited to these years for the current study. Inclusion of these years in this study reduced the overall sample from about 38,000 to roughly 24,000. Next, only those respondents who reported at least a high school degree and some college experience (greater than twelve years of education) were retained in our subsample. This restriction assured that respondents had an opportunity to belong to a Greek-letter organization. This decreased the sample size to around 9,000 respondents.

Then, because educational opportunities, especially for racial minorities and women, have increased over time, the age range of respondents was narrowed from 19 to 89 years of age, to 24 to 40 years of age. This restriction further reduced the sample size to around 4,000. Finally, despite the relatively high response rate of the GSS, missing data further restricts the number of complete cases to 3,712 respondents. In supplementary analyses not shown, we estimated probit selection models to explore the missing data. These analyses revealed that those respondents with missing values did not differ significantly from those with no missing values. Therefore, only the complete cases analyses are presented below. Table 1 displays the descriptive statistics and bivariate associations among the variables in this study.

Dependent Variables

In this study, our dependent variable is college graduation (1=yes). Roughly half of our sample, (53.1%) has a college degree. This proportion is higher than in the general population. However, recall that to be in the current sample that respondents had to have some college experience. This restriction alone increases the probability that college will be completed.

Independent and Control Variables

The independent variable, our measure of social capital, used in the analysis is membership in a Greek organization (9.3%, 1=yes). Greek-letter membership is positively correlated with the probability of college graduation (r=.191, p<.001).

We selected several control variables. These variables can be divided into sociodemographic information, human capital, and financial capital. The sociodemographic variables included race, gender, age, family structure and region of origin. We distinguished African Americans (1=yes) from European Americans. African Americans make up approximately nine percent of the sample, and being African American is negatively correlated with the probability of graduation. With regard to gender, we compared males, coded as 1, to females. Men comprise almost half (48.1%) of the sample. Male gender is positively correlated with the probability of college graduation and Greek-letter organization membership. Further, we controlled for age, measured in years, to the extent that older respondents may have had more time to complete education and to take advantage of social capital resources derived from membership in fraternities or sororities. The average age of the sample was 31.85 years. Also, growing up in a two parent family (1=yes) was compared to all other forms of family structures. This structure, about 90% of the sample, is positively correlated with college graduation. Recall that our sample is restricted to those respondents who at least began a college career. In the U.S., individuals growing up in two parent homes are not only more likely to start college, but also to graduate (Corcoran, 1995; Kao & Thompson, 2003). Finally, growing up in the South (1=yes) is compared to all other regions of the country. Our measure of the southern region includes: Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia and the District of Columbia (Davis & Smith, 1998). A little over a quarter (25.2%) of the sample hails from the South. For the current sample, growing up in the South was positively correlated with Greek-letter membership.

The human capital measures in this study were operationalized as the parents' years of schooling (Coleman, 1988, 1990). The mean levels of education for fathers and mothers were 12.61 and 12.32, respectively. Both measures of human capital were positively related to the respondent's probability of college graduation and Greek-letter membership.

Financial capital was measured by a GSS item that asked each respondent to identify his/her family income at the age of sixteen. The forced-choice response item ranged from 1 (far below average) to 5 (far above average), with a mean of 3.14. Family income was positively correlated with the probability of graduation and Greek-letter membership.

Analytic Strategy

We completed the analyses for this paper in two parts: (1) a logistic regression analysis; and (2) interaction effects used to test differences by race. In part one, logistic regression models were used to estimate the probability of college graduation and included two steps. The first step consisted of a model with all the control variables regressed on the probability of college graduation. In the second step, Greek-letter membership was added to the model. These steps allowed us to establish the overall contribution of membership to the model net of all other relevant variables. Further, we were able to assess whether and how Greek-letter membership changed the relationship between the probability of college graduation and the control variables. Finally, in the second part of the analyses, we estimated interactions to test for differences by race. The primary goal of the interaction analysis was to determine whether race moderates the effects of Greek-letter membership on educational attainment. Additionally, given known differences in educational achievement by race (see e.g., Cross & Slater, 2001), we calculated whether race qualifies the effects of any of the sociodemographic, human capital, or financial capital variables in the model.

Results

Table 2 displays the logistic regression analysis. Equation 1 of Table 2 accounted for only the sociodemographic, human capital and financial capital variables. Being African American is related to decreased graduation rates (b=-.732, p<.001). Compared to whites, the odds of graduation for African Americans are reduced by about 50% [([e.sup.-.732]-1) X 100]. Men are 24% [([e.sup..214] - 1) X 100] more likely than women to complete their college education, while one year of age represents a small (2%) but significantly higher probability of graduation.

In addition to age it was possible that differences in the sample might have emerged as a result of cohort effects. Therefore, we created dummy variables representing when the respondent most likely had been a college student (e.g., 1970s, 1980s and so forth). None of these variables were significant in the analyses and were pruned from the model.

Being reared by two parents boosts the probability of graduation by 86% [([e.sup..621] -1) x 100]. With regard to human capital, operationalized as parents' schooling, fathers' education (b=.109, p<.001) and mothers' education (b=.077, p<.001) increase the likelihood of college graduation. Each year of father and mother's education results in about a 12% and 8% higher probability of graduation, respectively.

In Table 2, Model 2, Greek-letter membership, our primary independent variable, is added to the model. This variable positively influenced the likelihood of college graduation (b=1.555, p<.001). Thus, membership in a Greek-letter organization increases the odds of college graduation by a little over 370%, compared to non-members. Interestingly, membership had the largest effect on graduation compared to all other variables in the model. Though adding Greek-letter membership to the model does increase the r-square by a little over four percent, little else in the logistic regression model changes. However, there is some variation in the size of the effects. The size of the effect of being African American decreases (b=-.732 versus b=-.804), as does the effect of being male (b=.214 versus b=.190). The influence of age on the probability of graduation increases slightly (b=.023 versus b=.025). The influence of father's education decreases slightly (b=.109 versus b=.108), as does the impact of mothers' education (b=.077 versus b=.072).

Interaction Analysis

The main effects models (Table 2) showed that there is a positive relationship between Greek membership and the probability of college graduation. However, with the exception of showing that African Americans are less likely to graduate, this analysis did little to elucidate racial differences in the process shaping college graduation. We accomplished the task of further exploring the racial variation in the sample using interactions. As described above, we estimated a series of interactions between race and all other variables in the model. Given that this research focused on how Greek-letter membership influences the probability of college graduation, the main focus for this part of the analysis was the interaction between race and membership [i.e., African American (1=yes) X Membership (1=yes)]. Additionally, other interactions with race and the sociodemographic, human capital, and financial capital variables were used to test further for racial variation in the process shaping educational attainment.

Table 3, Equation 1 shows the two interactions that were significant in the estimation of the probability of college graduation. First, race qualified the effects of Greek-letter membership. African Americans who are members of Greek-letter organizations benefit from this membership significantly more than their European American counterparts. Solving for the interaction, the impact of membership for African American (b=2.567) is about 55% larger than for European Americans (b=1.405): [1.405 (Membership) + 1.162 (African American X Membership) = 2.567]. Nevertheless, the overall effect of being African American still remains negative (b=-.922). Figure 1 depicts the interaction between race and Greek-letter membership. In our sample of respondents, who all have some college experience, the proportion of African American Greek-letter members who graduate from college is roughly 90% (see Figure 1). This proportion is about 80% for European Americans. Conversely, European Americans who are not Greek-letter members, with about 50% of them graduating, fair more favorably than their non-member African American counterparts, with about 25% graduating.

[FIGURE 1 OMITTED]

Second, the positive effect of father's education on the probability of college graduation is moderated by race. The size of the effect of father's education on college graduation for European Americans (b=.116) is about five times larger than for African Americans (b=.024): [b =.116 (Father's education) -.092 (African American X Father's education = .024)]. Figure 2 displays this interaction. At low levels of father's education, the proportion of respondents completing college is around 40%, more than 10 percentage points below the sample mean of 53.1% (see Figure 2). However, as father's education approaches the mean, African Americans appear to be at a distinct disadvantage. At the sample mean (12.610 years) for father's education, the proportion of African Americans graduating is still below 50% while the proportion of European Americans who graduate is closer to 60%. Moreover, as father's education increases, so does the more favorable trajectory for European Americans. At one standard deviation above the mean (a little over 16 years--a college degree) of father's education, the proportion of African Americans, who are graduating is now about 60%. At the same level of father's education, the proportion of European Americans who are graduation is approximately 75%.

[FIGURE 2 OMITTED]

Discussion and Conclusions

In this research project, we explored the impact of belonging to Greek-letter social organizations on the probability of college graduation. We further investigated whether there was racial variation in the process that shapes this outcome. Similar to others (e.g., Whipple, Baier, & Grady, 1991), we found that Greek-letter membership is positively related to college graduation. In fact, membership was the most robust predictor of whether an individual graduated from college This finding supports a social capital framework in which membership in voluntary organizations provides interpersonal resources that aid members in achievement (Coleman, 1990; Eastis, 1998).

Compared to their European American counterparts, Greek-letter membership is more advantageous for African Americans in terms of college graduation. At first blush, this finding may seem counter intuitive. After all, other research clearly indicates that African Americans are at a disadvantage in terms of educational attainment (Kao & Thompson, 2003). However, part of the disadvantage faced by African Americans in terms of socioeconomic attainment is being isolated from the mainstream and lacking rich social connections (Wilson, 1996). Therefore, for those African Americans who are able to attend college, Greek-letter organizations may provide social connections that help in this process. For whites who tend to have more resources at their disposal, they may simply be relying more on other types of resources.

Race also moderates the effects of father's education on college graduation. European Americans receive a greater benefit to their education from father's education than do African Americans. This finding confirms other research that indicates that whites are able to parlay their attainment into greater benefits for their children and themselves compared to racial minorities (Jencks et al., 1972; Hill & Duncan, 1987; Feagin & Sikes, 1994). For example, because of historical factors that include discrimination, highly educated blacks may not be able to transfer privilege to their children as easily as whites (Corcoran, 1995). Interestingly, we did not find differential effects by race for mother's education. This pattern may be an artifact of the traditional model of socioeconomic inheritance in which children's statuses are more likely to be affected by their father's attainment (Duncan, Featherman, & Duncan, 1972). More than half our sample was thirty years-old or older, which means that they grew up in the 1960s and 1970s, when fathers were more likely than mothers to be working outside the home and to have higher levels of education.

This research is limited in some respects. First, despite using a more nationally representative sample than the majority of studies on Greek-letter membership, the GSS data do not contain information about important features of the respondent's Greek organization that might be important to educational attainment. For example, some fraternities and sororities are predominantly African American. Having this information would have allowed us to differentiate the extent to which race and membership make a difference as a result of (or lack of) cultural solidarity. Second, no measure was available on the importance of Greek membership to an individual's identity or self-image. Other research links role identity to effects on individual motivations, satisfaction, and continued association with a role or organization (Simon, 1995). We cannot assess the extent to which role salience, satisfaction or identity impacts Greek membership or educational attainment. Third, the data for this study does not include information on the many economic costs of Greek letter membership. We partially address this issue by including family income, which is positively related to membership, as a covariate in this study. Nevertheless, other research shows that moderate to low income members of Greek organizations often struggle with joining fees, local and national dues, and the financial cost of participating in sponsored (e.g., fund raisers) and social (e.g., formal dances) events associated with Greek life (Byer, 1997). Therefore, future research should explore the extent to which economically disadvantaged youth, regardless of race, have difficulty both joining and maintaining their membership in college fraternities and sororities. Finally, while we use data collected from 1972 to 2002, the analyses are essentially cross-sectional and provide only a snapshot in time. Longitudinal data may have provided an opportunity to assess how the role of Greek membership changes over time. For example, does Greek membership become more important as graduation draws near or is it as important in the second year of college as it is in the fourth year? Only longitudinal data would allow an assessment of such issues.

In conclusion, this research found that Greek-letter membership is positively related to the probability of college graduation. Further, race moderates the effect of membership and father's education on college graduation. Despite these differences by race, it should be noted that this research also shows that the process shaping education is remarkably similar for African Americans and European Americans. Background resources, human capital, and organizations that produce social capital are important regardless of race. Moreover, this research included findings that merit further investigation. For example, this study was limited to two racial/ethnic groups. Other research should incorporate other groups (e.g., Asian Americans or Hispanic groups) to allow a better understanding of whether and how Greek-letter membership may shape educational outcomes. Finally, other research should explore whether belonging to Greek organizations or other types of voluntary organizations influences other educational out-comes, such as choice of major or decisions to attend graduate school.

References

Arnold, J. C., & Kuh, G.D. (1992). Brotherhood and the bottle: A cultural analysis of the role of alcohol in fraternities. Bloomington, IN: Center for the Study of the College Fraternity.

Astin, A.W. (1975). Preventing students from dropping out. San Francisco, CA: Jossey-Bass.

Astin, A.W. (1985). Achieving educational excellence. San Francisco, CA: Jossey-Bass.

Blau, P.M., & Duncan, O.D. (1967). The American occupational structure. New York: Free Press.

Bourdieu, P. (1986). The forms of capital. In J.G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241-258). New York: Greenwood Press.

Bradshaw, H.L., & Kahoe, R. (1967). Differential effects of fraternity and sorority membership upon academically promising students. Journal of Educational Research, 61, 62-64.

Brewer, G.A. (2003). Building social capital: Civic attitudes and behavior of public servants. Journal of Public Administration Research and Theory, 13, 5-26.

Bryan, W.A., & Schwartz, R.A. (1983). The eighties: Challenges for fraternities and sororities. Carbondale, IL: ACPA Media.

Buckner, D.R. (1961). A comparison of the scholarship of freshman fraternity pledges and residence hall men. Journal of College Student Personnel, 3, 20-24.

Byer, J.L. (1997). Comparisons of socioeconomic background characteristics of Greeks and independents. Research Note for the Educational Resources Information Center. Washington, D.C.: U.S. Department of Education.

Coleman, J.S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, $95-$120.

Coleman, J.S. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press.

Corcoran, M.E. (1995). Rags to rags: Poverty and mobility in the United States. Annual Review of Sociology, 21,237-267.

Crookston, B.B. (1960). Academic performance of fraternity pledges. Journal of College Student Personnel, 1, 19-22.

Cross, T., & Slater, R.B. (2001). The troublesome decline in African-American college student graduation rates. Journal of Blacks in Higher Education, 33, 102-109.

Davis, J.A., & Smith, T.W. (1998). General social surveys, 1972-1998: Cumulative codebook. Chicago, IL: National Opinion Research Center.

Duncan, G.J., Brooks-Gunn, J., & Klebanov, P.K. (1994). Economic deprivation and early childhood development. Child Development, 65, 296-318.

Duncan, O.D., Featherman, D.L., & Duncan, B. (1972). Socioeconomic background and achievement. New York: Seminar.

Eastis, C.M. (1998). Organizational diversity and the production of social capital: One of these groups is not like the other. American Behavioral Scientist, 40, 547-678.

Ehrhart, J.K., & Sandler, B.R. (1992). Campus gang rape: Party games? Washington, DC: Center for Women Policy Studies.

Engs, R.C., & Hanson, D.J. (1988). University students' drinking patterns and problems: Examining the effects of raising the purchase age. Public Health Reports, 103, 667-673.

Feagin, J.R., & Sikes, M.P. (1994). Living with racism: The black-middle class experience. Boston, MA: Beacon.

Foley, M.W., & Edwards, B. (1999). Is it time to disinvest in social capital? Journal of Public Policy, 19, 141-173.

Glasmeier, A.K., & Leichenko, R.M. (1999). What does the future hold? What globalization might mean for the rural south. Southern Rural Sociology, 15, 59-83.

Gonzalez, G.M., & Broughton, E.A. (1986). Status of alcohol policies on campus: A national survey. NASPA Journal, 24, 49-59.

Granovetter, M.S. (1985). Economic action, social structure and embeddedness. American Journal of Sociology, 91,481-510.

Hauser, R.M., & Featherman, D.L. (1976). Equality of schooling: Trends and prospects. Sociology of Education, 49, 99-120.

Haveman, R.H., Wolfe, B.L., & Spaulding, J. (1991). Childhood events and circumstances influencing high school completion. Demography, 28, 133-157.

Hill, M.S., & Duncan, G.J. (1987). Parental family income and the socioeconomic attainment of children. Social Science Research, 16, 39-73.

Hoscschild, J.L. (2003). Social class in public schools. Journal of Social Issues, 59, 821-840.

Hogan, D.P., & Pazul, M. (1982). The occupational and earnings returns to education among black men in the North. American Journal of Sociology, 87, 905-920.

Jencks, C., Smith, M., Acland, H., Bane, M.J., Cohen, D., Gintis, H., Heyns, B., & Michelson, S. (1972). Inequality: A reassessment of the effect of family and schooling in America. New York: Basic.

Jensen, L., Findeis, J.L., & Wang, Q. (2000). Labor supply and underemployment in the Southern United States. Southern Rural Sociology, 16, 96-124.

Kalmijn, M., & Kraaykamp, G. (1996). Race, cultural capital, and schooling: An analysis of trends in the United States. Sociology of Education, 69, 22-34.

Kao, G., & Thompson, J.S. (2003). Racial and ethnic stratification in educational achievement and attainment. Annual Review of Sociology, 29, 417-442.

Kohn, M.L. (1969). Class and conformity: A study in values. Homewood, IL: Dorsey.

Kuh, G.D. (1991). Environmental influences on alcohol use by college students. Office of Educational Research and Improvement. Washington, D.C.: U.S. Department of Education (ED 331-6).

Lieberson, S. (1980). A piece of the pie: Black and white immigrants since 1880. Berkeley, CA: The University of California Press.

Loundsbury, J.W., & DeNeul, D. (1995). Psychological sense of community on campus. College Student Journal, 29, 270-277.

Magoon, T.M., & Maxwell, M,J. (1965). Demographic characteristics between high and low achieving university students. Journal of College Student Personnel, 6, 367-373.

Marmaros, D., & Sacerdote, B. (2002). Peer and social networks in job search. European Economic Review, 46, 870-879.

Mau, W. (1995). Educational planning and academic achievement of middle school students: A racial and cultural composition. Journal of Counseling and Development, 75, 518-526.

Mizell, C.A. (1999). Earnings' potential: The consequences of family background, adolescent self-concept and adult labor force factors. Journal of African American Men, 4, 89-119.

Parcel, T.L., & Dufur, M.J. (2001). Capital at home and at school: Effects on student achievement. Social Forces, 79, 881-912.

Parcel, T.L., & Menaghan, E.G. (1994). Parents' jobs and children's lives. New York: Aldine De Gruyter.

Pike, G. (2003). Membership in a fraternity or sorority, student engagement, and educational outcomes at AAU public research universities. Journal of College Student Development, 44, 369-382.

Pugh, R.C., & Chamberlain, P.C. (1977). Undergraduate residence: An assessment of academic achievement in a predominately university community. Journal of College Student Personnel, 17, 138-141.

Putnam, R.D. (1995). Turning in, tuning out: The strange disappearance of social capital in America. PS: Politics and Society, 28, 664-683.

Putnam, R.D. (2000). Bowling alone: The collapse and revival of American community. New York: Simon and Schuster.

Requena, F. (2003). Social capital, satisfaction and quality of life in the workplace. Social Indicators Research, 61,331-360.

Samter, W. (1992). Communicative characteristics of the lonely person's friendship circle. Communication Research, 19, 212-239.

Sewell, W.H., & Hauser, R.M. (1975). Education, occupations and earnings: Achievement in the early career. New York: Academic.

Sherron, R.H. (1969). A study of academic and nonacademic predictors and criteria of success among the Morehead scholars at the University of North Carolina. Unpublished doctoral dissertation, University of North Carolina, Chapel Hill, NC.

Simon, R.W. (1995). Gender, multiple roles, role meaning and mental health. Journal of Health and Social Behavior, 36, 182-194.

Smith, A.W. (1989). Educational attainment as a determinant of social class among Black Americans. The Journal of Negro Education, 58, 416-429.

Taylor, B. (2000). Trade tmions and social capital in traditional communist states: The case of China. Policy Sciences, 33, 341-354.

Trice, A.D., & Knapp, L. (1992). Relationship of children's career aspirations to parents' occupations. Journal of Genetic Psychology, 153, 355-357.

United States Census Bureau. (2000). Educational attainment in the United States. Population Division, Education & Social Stratification Branch.

Warman, R.E. (1962). Pledges view of fraternity effect on scholarship. The Fraternity Month, 30, 30-32.

Whipple, E.G., Baier, J.L., & Grady, D.L. (1991). A comparison of black and white Greeks at a predominantly white university. NASPA Journal, 28, 140-148.

Wilson, W.J. (1996). When work disappears: The world of the new urban poor. New York: Vintage.

RONALD E. SEVERTIS JR.

The Ohio State University

Department of Sociology

C. ANDRE CHRISTIE-MIZELL

Kent State University

Department of Sociology

Table 1. Intercorrelations Among Variables for a General Social
Survey Sample (N=3712).

Measure 1 2 3 4

1. College graduation(1
 =yes)

2. Education (years) .829 ***

3. African -.136 *** -.122 ***
 American(1=yes)

4. Male (1=yes) .059 *** .096 *** -.072 ***

5. Age (years) 0.014 .065 *** -0.008 .031 *

6. Grew up in a two- 083 *** .093 *** -.058 *** 0.017
 parent home
 (1=yes)

7. Grew up in the -0.011 -0.019 .156 *** -.037 *
 South (1=yes)

8. Household 108 *** .092 *** -.125 *** 0.022
 Income (1)

9. Father Education .250 *** .267 *** -.160 *** 0.004
 (years)

10. Mother Education .210 *** .220 *** -.115 *** 0.008
 (years)

11. Greek-letter .191 *** .193 *** 0.015 .036 *
 Membership
 (1=yes)

Means 0.531 15.443 0.086 0.481

Standard Deviations 0.499 1.817 0.280 0.500

 5 6 7 8
1. College graduation(1
 =yes)

2. Education (years)

3. African
 American(1=yes)

4. Male (1=yes)

5. Age (years)

6. Grew up in a two- 0.016
 parent home
 (1=yes)

7. Grew up in the -0.02 -0.01
 South (1=yes)

8. Household -.102 *** 0.020 -.042 *
 Income (1)

9. Father Education -.144 *** .041 * -.078 *** .357 ***
 (years)

10. Mother Education -.121 *** .059 *** -.039 * .283 ***
 (years)

11. Greek-letter -0.032 0.009 .058 *** .057 ***
 Membership
 (1=yes)

Means 31.853 0.922 0.252 3.135

Standard Deviations 4.828 0.268 0.434 0.777

 9 10 11

1. College graduation(1
 =yes)

2. Education (years)

3. African
 American(1=yes)

4. Male (1=yes)

5. Age (years)

6. Grew up in a two-parent
 home
 (1=yes)

7. Grew up in the
 South (1=yes)

8. Household
 Income (1)

9. Father Education
 (years)

10. Mother Education .566 ***
 (years)

11. Greek-letter .069 *** .080 ***
 Membership
 (1=yes)

Means 12.610 12.318 0.093

Standard Deviations 3.748 2.862 0.290

* p<.05; ** p<.01; *** p<.001

(1) Family of origin household income when the respondent was 16 years
of age--ranges from 1 (far below average) to 5 (far average).

Table 2. Logistic Regression Analyses for the Probability of
College Graduation (I =yes) for a GSS Sample (N=3712).

 (1) (2)
 Logit Odds Logit Odds
 (SE) (SE)
Sociodemographic
Variables

African American -.732 *** .481 -.804 *** .448
 (1=yes) (.133) (.137)
Male (1=yes) .214 ** 1.238 .190 ** 1.209
 (.069) (.070)
Age (years) .023*** 1.023 025 *** 1.025
 (.007) (.070)
Grew up in two parent .621 ** 1.860 .634 *** 1.885
 home (I =yes) (.133) (.136)
Grew up in the South .121 1.129 .066 1.068
 (1 =yes) (.081) (.083)
Human Capital
Father's education .109 *** 1.115 .108 *** 1.114
 (years) (.012) (.012)
Mother's education .077 *** 1.080 .072 *** 1.074
 (years) (.015 (.016)
Financial Capital
Household income .021 .007
 ranges 1 (far below (.049) 1.021 (.049) 1.007
 avg.) to 5 (far
 above avg.)
Social Capital
Member of Greek 1.555 *** 4.735
 organization (1=yes) (.154)
Max-resealed R-square .120 .162

*** p<.001 ** p<.01 * p<.05; The coefficients in parentheses are
standard errors.

Table 3. Race Interaction analyses for the probability of college
graduation (1=yes) for a GSS sample (N=3712).

 (1) (2)
 Logit Odds Logit Odds
 (SE) (SE)

Sociodemographic
Variables
African American -.922 *** .398 .239 1.270
 (1=yes) (.148) (.407
Male (1-yes) .190 ** 1209 199** 1.220
 (.070) (.071)
Age (years) .025 *** 1.025 025*** 1.025
 (.007) (.007)
Grew up in two parent .638 *** 1.893 643 *** 1.902
 home (1 =yes) (.136) (.136)
Grew up in the South .068 .053
 (1 =yes) (:083) 1.070 (.083) 1.054

Human Capital
Father's education .110 *** 1.116 116 *** 1.123
 (years) (.012) (.013)
Mother's education 071 *** 1.074 072 *** 1.075
 (years) -0.016 (.016)

Financial Capital
Household income .005 .006
ranges 1 (far below (.050) 1.005 (.050) 1.006
 avg.) to 5 (far aboveavg.)

Social Capital
Member of Greek organi- 1.405 *** 4.075 1.542 *** 4.676
 zation (1 =yes) (.160) (.153)

Interaction Terms
African American X 1.162 *
Member of Greek (.488) 3.196
 organization
African American X -.092 **
Father's education (.034) 0.912
Max-resealed R-square .163 164

***p<.001 **p<.01 *p<.05.
Note: The coefficients in parentheses are standard errors.