Service learning has been used as a teaching tool for every type of learning from academic links between theory and practice to social problem-solving and interpersonal skill development . Now service learning is being explored as a method to meet higher education’s historic mission to educate adults to become civically engaged members of the community . As M. Stimmann Branson points out, “each new generation is a new people that must acquire the knowledge, learn the skills, and develop the dispositions... of private and public character that undergird a constitutional democracy” (p. 2) . As increasing numbers of each new generation enter higher education, methods are being sought by faculty and administrators to help students learn such democratic knowledge and skills in addition to the academic content of their disciplines. Can participation in service learning be a viable force for ensuring long-term civic engagement as vigorously as it serves as a viable teaching tool? Authors such as D. Owen believe that the two share an innate connection,  and in searching for methods to address what many have documented as the growing individualization, and political and community alienation of younger adults, service learning has become the method to which many  are turning in order to increase students’ engagement with their communities.
Before even addressing whether service learning influences civic engagement, it is important to define what is meant by civic engagement. Often, civic engagement is defined as political involvement . While it appears that political involvement in the United States has been decreasing since the 1970s, this decline seems to have become a public policy magnet in the last 10 years. Certainly, when definitions of civic engagement are limited to voting behavior and political participation, adults in the 18- to 24-year-old range appear to be rather uninvolved. For example, in 1996, only 32% of this cohort voted;  by 1998, this already low number had dropped to 20% . These can be very disheartening numbers if political behavior is the definition for civic engagement.
If civic engagement is defined more broadly, however, the picture can look downright optimistic. Van Benschoten also reported that 64% of this same age cohort had joined nonpolitical organizations, and 53% of these young adults reported volunteering in their communities (Ibid.). R. D. Putnam reports that civic engagement is down, even when more broadly defined,  but as Schudson points out,  while Putnam (“The Strange Disappearance”) is correct in that participation in conventional forms of civic activity, such as membership in the Elks club or the League of Women Voters, has decreased; “grassroots” activism, such as the pro-life movement and the pro-choice movement, and membership in national organizations, such as Sierra Club and the National Rifle Association, have increased. Perhaps, then, younger cohorts are not disengaged, they are differently engaged. And when civic responsibility is defined as meaning “active participation in the public life of a community in an informed, committed, and constructive manner, with a focus on the common good,” then certainly service learning, with its focus on participation in a community in order to meet local needs and learn more deeply the academic objectives of a classroom, seems to be a practical vehicle for fostering civic skills and knowledge at the same time that it forges connections between academic content and experiential learning .
Research, however, is mixed about whether there is a connection between service learning participation and civic engagement. Perry and Katula concluded that, “while some studies of service learning participants show enhanced compassion and interest in social problems generally, those attitudinal changes do not consistently translate into behavioral changes.”  T. Walker (“The Service/Politics Split”) surmised that when the low rates of political engagement and higher rates of volunteer activity are taken into consideration, “educators cannot simply assume that service contributes to political engagement,” for otherwise, political engagement levels would have already increased to match the higher levels of volunteer activity. J. J. Cowan stated that, “we must see service for what it is: an invaluable part of building a healthy national community—but not necessarily a bridge to greater political engagement.” In contrast, Allen affirmed that “at its best, service learning goes beyond volunteerism because it increases students’ personal involvement in academic and civic life.” M. X. Delli Carpini and S. Keeter believed indirect evidence can be found to suggest that, “including service as part of students’ educational experience can increase their motivation and opportunity to learn about politics, which in turn could increase the likelihood of their continued engagement in public life.... The key to success is likely to be found in the nature of the service experience and how well the experience is integrated into the classroom.”
The conflicting conclusions about service learning’s role in civic engagement development may be explained by S. Hunter and R. A. Brisbin, Jr., who stated that while, “many authors claim participating in service learning can lead students to better understand and practice democratic and civic values, few detailed evaluations of the effect of service learning on students’ political behavior and attitudes have been published.” What has been published is a mixture of results and conclusions from service learning research on K-12 school programs and higher education programs. Adding to the mixture is the differing definitions of civic engagement used in these studies. In some studies, the focus is on assessing service learning and political knowledge or involvement (Walker, “The Service/Politics Split”), while in other studies, the focus is on community involvement more widely defined [17–19].
In an attempt to lessen the confusion about whether service learning is related to civic engagement development, it might be appropriate to follow the model created by Gottlieb and Robinson’s, and Putnam’s (“The Strange Disappearance”) broader definitions of civic engagement that include both political and social involvement in community life in assessing the civic engagement of students. Additionally, as Delli Carpini and Keeter seem to suggest, if part of the explanation for service learning influences on student civic engagement development lies in the approach taken by the service learning program in creating service experiences, it would also be appropriate to assess student civic engagement levels in service learning programs that were developed or influenced by specific mentoring, such as exist in the training provided through Campus Compact, the Community College National Center for Community Engagement (CCNCCE), or the American Association of Community Colleges (AACC), to name a few . Such a focus would also remove the confusion around what level of education such a relationship might be found. Since many of the studies on civic engagement seem to be focused on K-12 service learning programs, [21, 22] this study was conducted to add to the literature on service learning’s ties to civic engagement development at the post-secondary level, and specifically at the community college level.
In order to assess civic engagement in students, two civic engagement surveys were created. The two surveys contained the same 27 questions, but an additional seven questions specifically addressing service learning in the class in which the student had participated were added to the end of one of the surveys. This survey became the post-course survey and, in addition to the comparison between service learners and non-service learners at the end of an academic term, I used the post-course civic engagement survey to query the service learners about their experiences with service learning as an educational tool. The pre-course survey was to be given at the beginning of a semester as students were being introduced to service learning. Its purpose was twofold. First, I used it to establish a baseline of civic engagement actions and community knowledge so that changes in actions or knowledge during the semester could be identified. Second, I used it to ascertain the similarities between the students who chose to participate in service learning and students who chose not to participate. Historically, one problem with only evaluating end-of-semester surveys about service learning involvement and attitudinal or behavioral changes is that one does not know whether these service learners were different from non-service learners before the service learning experience even began. Surveys took approximately 10 to 15 min to complete.
To construct both surveys, other civic-engagement surveys were reviewed, factors mentioned in the literature were identified, and other service-learning researchers were consulted. Topics that appeared in multiple surveys and those repeatedly mentioned in the literature on civic involvement were included. Questions about political involvement such as voting behavior, attendance at local town, city, or tribal council meetings, running for political office were included; as were questions about broader community involvement such as working to address a community problem, signing a petition, writing an opinion letter to a newspaper, knowing about local community service agencies, and commitment to volunteer in the future.
To verify that the surveys would be reliable and valid instruments, a pilot study was conducted during the spring 2004 term. To narrow the definitions and approaches to service learning that faculty who participated in this study might use, I went to the service learning directors from the 36 community colleges across the country that had previously been involved in Corporation for National and Community Service-funded (CNCS) service learning development grants from AACC (1994 through 2003). The definition used by colleges who participated in AACC service learning grant activities was that, “service learning combines community service and classroom instruction, with a focus on critical, reflective thinking, community-identified needs, and personal and civic responsibility.” The 36 service learning directors were asked to participate in this pilot study, and 14 agreed to participate. Each was asked to identify up to 30 students who were participating in service learning in the spring 2004 term, plus up to an additional 30 students who were not participating in servicelearning that term. The service learning program directors identified faculty who used service learning and faculty members who did not. These faculty members agreed to distribute the pre-course civic engagement survey to service-learning students and non-servicelearning students in their classes at the beginning of the spring 2004 term. The latter were either in the same class as the service learning students but chose not to participate in service learning, or were in a comparable class, i.e., another section of the same course, but one that did not offer a service learning component. In many cases, the same instructor taught both sets of students. At the end of the spring 2004 semester, the same faculty members distributed the post-course survey to the same service learning students and non-service learning students that had taken the pre-course survey. During summer 2004, the data were examined and the survey questions modified according to the feedback received through the pilot study.
The full study began with the fall 2004 semester. All 36 previously funded colleges were invited to continue their participation in the study (i.e., with new groups of students each semester). Additionally, eight community colleges had been selected by AACC to participate in the latest CNCS-funded Community Colleges Broadening Horizons through Service Learning (Horizons) grant project from 2004 through 2006. The program directors for these eight colleges were required to administer the pre-course and post-course civic engagement surveys each semester of the grant period. Each was instructed to choose at least two classes at his or her college for survey administration. The instructions for class selection were the same as the instructions for the pilot study. The two classes could be made up of either one class that required students to participate in service learning and one that did not, or two classes that offered service learning as an option so that some students in these classes would choose to participate while the rest of the students would not. The pre- and post-course civic-engagement surveys and the student informed-consent form were posted to the project’s listserv so that each service learning project director could print the surveys and consent form from the e-mail attachment and have access to it in the future if needed. Directions for administration were included in the e-mail, and emphasized that the same students who were given the pre-course survey were to be given the post-course survey and that only matched pairs of surveys with signed student informed consent forms would be analyzed.
By the end of the first year of survey use (Summer 2005), 166 matched pre- and post-course civic engagement surveys had been returned. In the fall 2004 term, 105 matched pre- and post-course civic engagement surveys were received from six colleges: two colleges from the previous grant periods, and four colleges from the current grant. Of these, 73 were from servicelearners and 32 were from non-service learners. In the spring 2005 term, 61 matched pre- and post-course civic engagement surveys were received from five colleges: the same two colleges who had responded in the fall term from the previous grant periods; and three colleges from the current grant: two that had responded in the fall term, and one that had responded for the first time. Of these, 34 were from servicelearners and 27 were from non-service learners. All of the current Horizons colleges had sent pre- and post-course surveys for both terms, however, not all colleges had matched pre- and post-course surveys, and thus these surveys were not included in the analyses.
The demographic profile of students consisted of age, enrollment status (full-time or part-time), employment status (full-time, part-time, or none), caretaker responsibility for family members (yes or no), and volunteer behavior during the past year (regularly, once in awhile, or none). Age was included because so much of the literature on civic engagement has cited the low levels of political involvement among young adults (typically defined as less than 25 years old) as evidence for the decline in civic engagement. I included the last four demographics in an attempt to gauge the students’ commitment levels as a possible alternate explanation for low civic participation.
Breaking the demographics down by academic term, in the fall 2004 term, service learners were more likely to be 25 and older (32%) than non-service learners (19%). The majority of both types of learners were full-time students (79% of service learners, 75% of non-service learners), while service learners were slightly more likely to work part-time (51%) than non-service learners (44%). Both groups of learners were primarily composed of non-caretakers (73% of servicelearners, 75% of non-service learners). Finally, service learners were more likely to have volunteered regularly during the previous 12 months (18%) than non-service learners (9%) and less likely to have done no volunteering during the past 12 months (25%) as compared to the non-service learners (41%).
In the spring 2005 term, the demographic profile was different. Service learners were slightly less likely to be 25 and older (21%) than non-service learners (26%). Service learners were also more likely to be full-time students (82%) and full-time workers (70%) than non-service learners (70 and 30%). Service learners and non-service learners were similar in that the majority of each group were non-caretakers (71% of service learners and 73% of non-service learners). Finally, more service learners volunteered regularly during the past 12 months (23%) than non-service learners (12%), and fewer did no volunteering during the past 12 months (23%) than non-service learners (44%). The spring profile of participants is interesting in that the service learners appeared to have more time commitments through work and school, yet were still more likely to have volunteered during the previous 12 months than the less-time-committed non-service learners.
Pre-course Civic Engagement Survey Comparison: Fall 2004
Before analyzing the relationships between the service learners’ and the non-service learners’ pre- and post-course surveys, t test analyses were conducted comparing the fall 2004 service learners’ responses on the pre-course surveys and the fall 2004 non-service learners’ responses on the pre-course surveys. I ran this test to determine whether the two groups had similar responses to the survey questions before the service learners had any service learning experiences. One of the few methods to determine if service learning does indeed insert a detectible influence on the development of civic engagement, knowledge, and commitment is to measure both groups of students on civic engagement knowledge and commitment before students experience service learning. For this comparison, service learners who, on survey question 25 (At this college, how many courses have you taken that included service learning?), answered “a” (none) were compared to non-service learners who answered “a” (none) on question 25. Interestingly, 19% of non-service learners indicated by choosing answers “b” through “f” on question 25 that they had previously taken one or more service learning classes and thus they were excluded from the t test analysis because they had experienced service learning in the past and could no longer be counted as non-service learners. When the t test comparison was done on the two fall 2004 groups of students, the hypothesis being tested was that no statistically significant differences would exist between the pre-course survey responses of the two groups; the analysis supported this hypothesis. There was no statistically significant difference between the service learners (n = 42, M = 5.71, SD = 2.95) and the non-service learners (n = 26, M = 6.31, SD = 3.03), t = −.8, p = 0.43 (two-tailed), α = .05, in their responses to the pre-course civic engagement survey.
Pre-course Civic Engagement Survey Comparison: Spring 2005
It was also hypothesized that no statistically significant difference would exist between the pre-course civic engagement scores of the spring 2005 service learners who had no previous service learning experience and the pre-course civic engagement scores of the spring 2005 nonservice learners with no previous service learning experience; however, when the spring 2005 pre-course survey responses were compared, a statistically significant difference was detected. This difference indicated that service learners with no previous service learning experience (n = 23, M = 7.13, SD = 2.36) and non-service learners with no previous service learning experience (n = 16, M = 5.44, SD = 2.25) were different from each other before any of the students had experienced service learning, t = 2.24, p = .03 (two-tailed), α = .05. To understand what differences existed at the beginning of the semester for the spring 2005 participants, I compared answers to individual questions for the two groups of students. There were six questions on which service learners scored 10 percentage points or higher than nonservice learners. Service learners were more likely to have voted in national elections (12% higher), signed a written petition related to a political or social issue that was important to them (13% higher), known the name of their community’s chief elected official (20% higher), worked with an individual or organization to address a problem in their community (13% higher), known which community service agencies could help a student who was homeless (34% higher), and believed that they would volunteer in their communities in the next 12 months (57% higher).
ANCOVA Analyses: Fall 2004
To test the hypothesis that students who participate in service learning would demonstrate more civic engagement than students who do not participate in service learning, the data from the fall 2004 semester were analyzed with a one-way ANCOVA. Service learners who indicated on question 25 that they had taken two or more classes at the college that included service learning (n = 15, M = 10.08) were compared to non-service learners who indicated on question 25 that they had taken no classes that included service learning (n = 22, M = 7.31). Results supported the hypothesis in that service learning students scored statistically significantly higher on the post-course civic engagement survey than non-service learners (F = 4.47, p = .02 one-tailed, α = .05).
ANCOVA Analyses: Spring 2005
The analysis of survey data from spring 2005 also revealed a statistically significant difference (F = 9.28, p = .01 one-tailed, α = .05) between service learners who had taken two or more classes that included service learning (n = 11, M = 7.08), and non-service learners who had taken no previous classes that included service learning (n = 14, M = 4.44).
Discussion and Conclusions
Overall, the results supported the premise that, at least in community college students, participation in service learning may have an influence on increasing students’ civic engagement levels when civic engagement is defined as more than just political action. Both in fall 2004 and spring 2005, servicelearning students scored statistically significantly higher on the post-course civic engagement survey than non-service learning students. In spring 2005, however, service learners were scored higher and thus were already statistically significantly different than non-service learners in their answers on the pre-course civic engagement survey. Why these students were statistically different from each other on their answers to the pre-course survey when the fall 2004 students were not is unclear. It is possible that the sample sizes in the spring 2005 groups were not large enough to be representative of the population, but until additional data are gathered, this possibility remains untested.
The fall 2004 data are more persuasive as the service learners and non-service learners were not significantly different in their answers on the pre-course civic engagement survey and yet were statistically significantly different in their answers on the post-course survey. For the fall 2004 participants, at least, it appears that broadening the definition of civic engagement to include both political and social involvement in community life may have allowed for different types of engagement to be detected. In addition, choosing to study in service learning programs that had been influenced by the same development model may have created more consistent experiences for service learners that allowed them to be statistically detectible. Again, replications of these results are needed before firm conclusions can be made, but civic engagement does seem to be fostered in community college students by participation in service learning when the definition of engagement is broadened and variations in the development models for the programs being assessed is minimized.
The benefit of broadening the definition of civic engagement beyond a one-definition-fits-all model might be seen in the detection of different levels of civic engagement in this study and is supported by other researchers, who identified the surfacing of three different types of civic engagement in their study of high school programs designed to foster education’s democratic purposes . They found that at the end of program involvement, some students could be described as “personally responsible citizens” who hold “a job, pay taxes, give blood, and obey the law”; while others were better described as “participatory citizens” who are “active in community affairs” through planning “community events and participating on local boards;” and still others could be characterized as “social reformers” who seek “to understand the causes of societal problems” and address their root causes (p. 237; in later research, “social reformer” was changed to “justice-oriented citizen”). While results from the current study are not specific enough to support or refute these three delineations of citizenship emerging in community college students after service learning participation, it could be hypothesized that, in general, service learning that is focused on the impact of individual service acts in community life would, at a minimum, foster development of personally responsible citizens. An explanation for how service learning does this may have been provided by Campbell  when he used social capital development to explain how service learning could lead to increased civic engagement. The connection between the two is the norm of reciprocity. Social capital, defined as “features of social organizations such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit” (p. 67), increases the development of the norm of reciprocity . Those who have developed a belief in mutual reciprocity consider that benefits enjoyed now should be repaid in the future, and such a belief increases civic activity . Even if all service learning programs do is enhance “compassion and interest in social problems,” this should at a minimum foster the development of personally responsible citizenship, which is not the same as not fostering civic development at all .
Whether the connection between service and civic engagement is due to the belief in reciprocity or some other construct, how the connections between the service experience and civic engagement are made tie back into the aim of the service learning activity that the student is engaging in. Studying service learning programs that have been influenced by the same development model allows the researcher to be more specific in what the hypothesized outcomes should be. For example, some have suggested that “the vast majority of large service-learning initiatives emphasize volunteerism and charity.” It should not be surprising, then, when the description of civic engagement sounds like a description of the responsible citizen because the “research has concentrated on a conception of citizenship that privileges individual acts of compassion and kindness over social action.” (p. 45). Yet if different conceptions of citizenship were not considered in these studies, what questions were left out that, had they been present might have revealed some development of the participatory citizen or social reformer, or of additional types of citizens not yet described? Kahne, Westheimer, and Rogers recognized that research has been used far less to investigate the “degree to which service-learning programs contribute to participants’ understandings of organized movements... and the ways government and corporate sectors constrain and enable solutions to social problems” (p. 45). As they suggest, knowing the approach that service learning programs take seems to be important in understanding what form of civic engagement will be fostered in students, and thus what is still lacking is an identification of the servicelearning components that change students into social reformers as compared to those components that foster participatory citizens or responsible citizens. If these factors can be identified, then faculty members can be more conscious of how they integrate service learning into their course activities in order to achieve their desired learning goals, and evaluators can be more specific in designing measures to assess goal attainment. If developing justice oriented citizens is not a goal of a particular service learning experience or program, then that program should not be criticized for the lack of justice-oriented citizens. Service learning need not transform every student into a social reformer for it to be deemed successful or valuable. Success should instead be tied back to the goals that the faculty member had for students who participated in service learning—both academic and civic. Studying the types of civic engagement development in service learning programs that are randomly chosen may be applying too narrow a question to too broad a population to detect any consistent findings. Research is needed that focuses on identifying various approaches of service learning programs and then measuring the types of civic engagement development that are fostered by these specific approaches. Then it will be clearer that service learning is up to the challenge of fostering civically engaged and academically competent students.