The Estimated Effects of Service Learning on Students’ Intercultural Effectiveness

As the higher education landscape continues to diversify, intercultural effectiveness comes to the forefront among important outcomes for students. Service learning is one programmatic tool that institutions of higher education can use to foster the development of intercultural effectiveness. Service learning has been defined in a variety of ways, yet most definitions include the combination of knowledge gained within the classroom and service experiences within the community (see Astin, Oseguera, Sax, & Korn, 2002; Kenny & Gallagher, 2002; Kuh, 2008; Zlotkowski, 2002). Service learning has been documented as providing positive benefits to students primarily surrounding two areas: (a) citizenship (see Astin & Sax, 1998; Einfeld & Collins, 2008) and (b) multicultural competency and social responsibility (see Einfeld & Collins, 2008; Engberg & Fox, 2011; Giles & Eyler, 1998; Jones & Abes, 2004; Pascarella & Terenzini, 2005).

Not surprisingly, with such extensive evidence supporting its positive benefits to students, the Association of American Colleges and Universities (AAC&U) named service learning a “high-impact” educational practice (see Brownell & Swaner, 2010; Kuh, 2008). Emerging research suggests, however, that vetted good practices within AAC&U’s “high-impact” practices are more likely providing positive benefits to students (see Padgett, Keup, & Pascarella, 2013). These good practices have been found to positively benefit student development and include items such as quality interactions with faculty outside of class, integrative learning, academic challenge, and positive interactions with diverse peers (see Chickering & Gamson, 1987; Goodman, Baxter Magolda, Seifert, & King, 2011; Pascarella & Terenzini, 2005).

Utilizing data from the Wabash National Study of Liberal Arts Education (WNS), the current study is guided by the following questions: (a) What is the effect of service learning on undergraduate students’ intercultural effectiveness? (b) Are mediating variables influencing the effects of service learning on undergraduate students’ intercultural effectiveness?

METHOD

The current study draws from student-level data from the WNS, a longitudinal pretest/ posttest design study investigating the effects of college experiences on a host of cognitive and psychosocial outcomes. The current study’s sample (2006–2010 cohort) includes 17 four-year institutions—varying in institutional control, size, and selectivity—that participated in the WNS. Liberal arts colleges were purposefully oversampled within the WNS. More detailed information about the WNS and all outcome measures can be found at: http://www.liberalarts.wabash.edu. Listwise deletion was utilized, leaving the current study with a sample size of 1,934.

Dependent Measure

The dependent variable for the current study is intercultural effectiveness. Intercultural effectiveness was measured by the short form of the Miville-Guzman Universality-Diversity Scale (M-GUDS). The M-GUDS is a 15-item scale measuring universality-diversity orientation (UDO), which is “an awareness and potential acceptance of both similarities and differences in others that is characterized by interrelated cognitive, behavioral, and affective components” (Fuertes, Miville, Mohr, Sedlacek, & Gretchen, 2000, p. 158), with an internal reliability consistency of 0.85.

Tests of validity of the measure have been linked to several psychological constructs as well as diversity attitudes of college students (Fuertes, Miville, et al, 2000; Fuertes, Sedlacek, Roger, & Mohr, 2000). The M-GUDS was specifically chosen for this study because of its construct validity capturing students’ attitudes toward diversity. With much literature suggesting the link between service learning and diversity-related attitudes and competency, this construct was employed to examine the relationship.

Independent Measures

A dichotomous variable was included for sex (male versus female). Due to low sample sizes, a variable for race was dichotomized (students of color versus white). A proxy variable for socioeconomic status was included (received a federal grant versus did not). To control for precollege level of intercultural effectiveness, the precollege pretest measure score was included. Additional precollege covariates included high school racial composition (all-white high school composition versus not), academic ability (ACT composite score), high school volunteer involvement, interactions with high school teachers outside class, and academic motivation. To account for institutional type, two dichotomous variables were included (regional and research university versus liberal arts college).

The independent variable of interest was service learning, which was included as a dichotomous variable (did participate in service learning versus did not). A continuous variable was included for the number of diversity...