Service-Learning as a Catalyst for Community Development

Abstract

Service-learning has the potential to create mutually beneficial relationships between schools and communities, but little research explores service-learning from the community’s perspective. The purpose of this study was to (a) understand how community-based organizations (CBOs) benefited from partnering with students and (b) examine whether organizational capacity (e.g., organization size) or employee vision (e.g., belief in the abilities of children) had a greater impact on the extent to which organizations benefited from their partnerships. The sample consisted of 129 CBOs that received a grant from K-12 students engaged in a service-learning program. Organizational capacity was more predictive of CBO behaviors, such as involving students in service projects and interacting with students, whereas employee vision was more predictive of positive CBO beliefs, such as the future potential of their partnership. This study helps service-learning practitioners and researchers understand how to better support CBOs that wish to form meaningful partnerships with schools.

Keywords

service-learning school/community partnerships urban education community development

Mounting evidence has linked service-learning to a host of desirable outcomes for students and schools. For students, service-learning can increase school engagement (Billig, Root, & Jesse, 2005; Furco, 2001; Scales, Blyth, Berkas, & Kielsmeier, 2000), reduce risk of high school drop-out (Bridgeland, Dilulio, & Burke Morrison, 2006; Family Strengthening Policy Center, 2007), and improve civic and social competencies (Billig et al., 2005; Melchior, 1999). Service-learning, moreover, is associated with important school-level outcomes, including improved student–teacher and peer relationships (Melchior, 1999) and fewer behavioral problems (Follman & Muldoon, 1997). As the body of empirical literature on the individual and school-level impacts of service-learning grows, however, we continue to know very little about the impact of service-learning on community partners (Basinger & Bartholomew, 2006; Imperial, Perry, & Katula, 2007; Worrall, 2007) and what factors predict the extent to which community partners are able to benefit from partnerships with schools.

Further exploration of this topic is critical for many reasons. The National Service-Learning Clearinghouse (n.d.) defines service-learning as “a teaching and learning strategy that integrates meaningful community service with instruction and reflection to enrich the learning experience, teach civic responsibility, and strengthen communities.” Although this definition touches on strengthening communities as an outcome, it mostly positions service-learning as a teaching and learning strategy. Thus, it is unsurprising that the pedagogical benefits of service-learning have assumed a more prominent position in the research literature than the community benefits. Yet, the “service” and the “learning” components of service-learning are in fact mutually reinforcing. Students learn more when they perceive that they have delivered a meaningful service (Eyler & Giles, 1999), and organizations are potentially more likely to invest in student learning when they perceive a real organizational benefit (Batenburg, 1995).

More broadly, understanding how and why organizations benefit from service-learning is particularly important in high-poverty urban contexts, where school/community partnerships help meet the physical and socio-emotional needs of students (Sanders, 2003) but tend to be one-sided: Community-based organizations (CBOs) provide services to schools, but rarely do schools reciprocate (Sanders, 2003; Stone, Doherty, Jones, & Ross, 1999). Service-learning is one mechanism through which schools in urban areas can contribute to community development, and in doing so, potentially begin to dispel deficit orientations that adults often hold toward low income and minority urban youth (Christens & Dolan, 2011; Males, 2006).

In the present study, we investigate how the Penny Harvest (PH), a service-learning program that operates in hundreds of New York City elementary and middle schools, fosters community development through enabling community partners to leverage their partnerships with schools into new opportunities to serve their constituents, educate a new generation about their work, and access new friends and resources. Furthermore, we explore which organizational-level factors influence the degree to which organizations are successful at leveraging their partnerships into these opportunities. We conducted this research, with consultation from a faculty member and a graduate student at a nearby university, when we were employed at Common Cents, the organization that operated the PH. Although we were each intellectually curious about the community benefits of service-learning, the research was motivated by our practical desire to help CBOs learn how to maximize the educational and organizational impact of their partnerships with schools. We regularly communicated with CBOs through workshops, e-mails, and mailings, and we hoped for this research to inform those communications.

Literature Review

Partnerships between CBOs and urban schools tend to take the form of school-linked services, whereby organizations provide school-based health services to students or operate socio-emotional or youth development programs (Tyack, 1992). Such partnerships originated during the Progressive Era as a response to immigration, industrialization, and urbanization (Tyack, 1992), and although highly critical in under-resourced neighborhoods, school-linked services tend to perpetuate existing notions of low-income children as “needy” or “deficient” (Tyack, 1992). Thus, when students offer a meaningful service to their communities, adults may begin to re-orient their deficit perceptions of low-income urban youth, resulting in positive youth and community development (Christens & Dolan, 2011). Although meaningful service should be a critical component of service-learning in any population, its value is arguably more salient in low-income communities due to its potential to re-frame the historical one-sidedness of school/community partnerships.

Despite the promise of service-learning partnerships in urban settings, very little research has been conducted regarding the motivations for CBOs to engage in service-learning, particularly at the K-12 level. From interviews with several CBO representatives who had engaged in service-learning partnerships with schools, Batenburg (1995) concluded that organizations have three purposes for engaging in service-learning partnerships. First, CBOs engage in partnerships to better serve their clients. Not only do students offer extra hands but also several organizations report that students offer unique skills and perspectives that benefit both their organizations and their constituents (Sandy & Holland, 2006; Worrall, 2007). For example, Cairn (2003) cites activities such as refugee children translating health and housing information for elders, children sparking happiness in senior citizens, and teenagers speaking against drug use to peers and young children. Each of these examples demonstrates the unique capabilities of youth volunteers.

Second, CBOs that are committed to the public good engage in partnerships to educate young people about their work, including exposing youth to social justice and developing civically engaged citizens (Batenburg, 1995). Furthermore, these CBOs consider education as a strategic way to educate a future generation of volunteers, donors, and board members. Batenburg also notes that many CBOs attract idealistic staff who are committed to a better world. Mired down in the day-to-day struggles of trying to solve massive social problems, the opportunity to work with children enables staff to re-connect with their hope for a better future.

Finally, CBOs engage in partnerships to generate new support and resources. For example, CBOs report that they appeal to funders by advertising the diversity of their volunteer pool and the efficiency with which they use youth volunteers (Worrall, 2007). In addition, through collaborating with higher education institutions, CBOs garner access to additional resources and contacts (Sandy & Holland, 2006). At a national summit of CBO leaders convened by the Consortium for the Advancement of Private Higher Education (Leiderman, Furco, Zapf, & Goss, 2002), CBOs perceived that benefits of partnering with higher education institutions included further advancement of their mission, new perspectives and insights sparked by interaction with students and faculty members, and access to knowledge and research on campus, grant opportunities, and an expanded resource base (Leiderman et al., 2002).

Despite these perceived benefits, there are many challenges to school/community service-learning partnerships. For example, organizations tend to measure success as a tangible product that directly benefits the organization—number of hours students spent with constituents, number of grocery bags stuffed, and so on—whereas schools measure success based on student learning outcomes, such as attendance rates or standardized test scores (Abravanel, 2003; Batenburg, 1995). In addition, structural and cultural differences between schools and agencies compound conflicts brought upon by already differing goals; CBOs might become frustrated with a partnership because it is too difficult to reach a teacher, forgetting that teachers have little down time during their days to return phone calls and e-mails, particularly in urban schools where the high-stakes testing environment focuses on maximum instructional time.

Given all of the challenges associated with school/community partnerships, what factors may predict the degree to which an organization is able to benefit from a service-learning partnership? One theory is that organizational capacity, including organizational size, staff experience, number of staff involved in the partnership, and number of adult volunteers involved in the partnership, influences a CBO’s ability to engage students. Sandy and Holland (2006) found that although many CBOs reported that service-learning partnerships with college students enabled them to fulfill their mission of promoting social justice, smaller organizations were less capable of pursuing this goal. In addition, Littlepage, Gazley, and Bennett (2012) found that prior experience working with students was one of the strongest predictors of an organization’s desire to engage college students as volunteers. The number of volunteers involved in a partnership may also influence the degree of success, as Edwards, Mooney, and Heald (2001) found that organizational volunteers tend to supervise college student volunteers. Finally, Abravanel (2003) suggested that service-learning will be more productive for an organization when more staffs are involved. To the best of our knowledge, the number of involved staff and volunteers has not been empirically tested as predictors of organizational benefit.

However, organizational capacity may have little to do with predicting organizational benefit from a service-learning partnership. Littlepage et al. (2012) found that controlling for more qualitative factors, such as an organization’s perceived benefit of partnership, organization size no longer significantly predicted the desire to engage college students in service-learning. This study suggests that an organization’s enthusiasm and vision for the benefits of service-learning may be a more salient predictor of how much it benefits from the partnership. A small organization with a less rigid structure may even grant employees more freedom to engage in activities related to service-learning. Batenburg (1995) also found that the respondents he interviewed who were enthused about service-learning believed in its potential to educate a future generation about their work and improve their own job quality.

The Present Study

There are several gaps in the current research literature. Although there have been some studies regarding the motivations for CBOs to participate in service-learning partnerships, we know little about how CBOs actualize these perceived benefits. Furthermore, if CBOs are able to benefit from service-learning partnerships in the ways in which they intend, we know little about why certain organizations are more successful in this endeavor than others. The studies that have pursued this question have examined predictors of engaging students in volunteering (Littlepage et al., 2012; Sandy & Holland, 2006) but have not explored which factors predict broader organizational-level benefits, such as cultivating future supporters and garnering friends and resources. Perhaps small organizations will still choose to partner with schools—in fact, they may need the help more—but they may not have the capacity to truly maximize their partnership through educating about their work and garnering friends and resources. Also, despite the potential for service-learning to build needed partnerships between schools and local organizations (Sanders, 2003), the very small existing literature regarding the community-level impacts of service-learning focuses on higher education rather than K-12 education. To situate our study in previous theory, we used Batenburg’s (1995) framework specifying three objectives of community partners for entering into service-learning partnerships: (a) direct service provision to their clients, (b) education and outreach in their field of expertise, and (c) generation of friends and resources. The purpose of this study was (a) to understand to what extent CBOs that had received PH grants had accomplished these three goals and (b) to examine whether organizational capacity or employee vision had a greater impact on the extent to which CBOs accomplished these three goals.

Method

This research was conducted with Common Cents, a New York City based non-profit organization that runs a service-learning and youth philanthropy program called the PH. The PH operates in more than 800 New York City elementary and middle schools, most of which are comprised by a majority of socioeconomically disadvantaged and racial minority students. Students in each participating school form leadership committees and decide how to allocate roughly $1,000, collected by the whole school, to local CBOs. While some CBOs just receive grants, others form lasting partnerships with the students who funded them through engaging students in service projects, educating students about their work and issue area, and connecting with members of the school community, such as parents and teachers. Whether or not CBOs take these extra steps is mostly up to their own discretion.

Data for the present study were collected through a survey of all CBOs that received a PH grant in Spring 2009. All organizations that received a grant that year and were located in New York City were invited to complete an electronic survey. We crafted the survey based on focus groups and interviews we conducted with CBOs over the previous several years. We e-mailed the survey to CBOs at the end of the school year, when they were most capable of reporting accurately on their involvement with the PH. Respondents were allowed 1 month to complete the survey, and two reminder e-mails were sent to all organizations. The survey took about a half an hour to complete, and respondents were entered in a drawing to win a selection of retail items. There were 165 initial respondents. Due to 36 organizations failing to complete the entire survey, the final sample size was 129. Respondents represented a diverse array of organizations, from small voluntary organizations to well-known national agencies; they also addressed a range of issues, including health, homelessness, hunger, environment, and animal welfare.

Measures

The survey consisted of a mixture of close-ended and open-ended questions. Open-ended questions asked respondents to describe the impact of partnering with a school(s) and publicizing their partnership, how children contributed to their organization in a unique way, and the aspect of the PH that most impacted their organization. Close-ended items included the following.

Independent Variables

Organizational capacity

There were four variables designed to measure organizational capacity. Respondents indicated the number of employees who worked at their office or workplace, how many years they had been involved in the PH, the number of staff members who were involved with the PH, and whether volunteers or constituents were involved with the PH, a “yes” or “no” question.

Employee vision

There were three variables designed to measure employee vision. The variable, belief in children, was measured by the item, “Young children are valuable philanthropists because they tend to recognize different community needs than adults do.” Responses were on a scale from 1 to 5 (1 = strongly disagree, 5 = strongly agree). Respondents also indicated, on a scale from 1 to 5, the extent to which they believed PH students with whom they interacted were professional, prepared, capable of critical thinking, curious, engaged, comfortable with adults, immature, and apathetic. Responses were averaged together to form the variable, perception of children, and negative descriptors were reverse coded. In addition, respondents indicated their favorite aspect of the PH, choosing between the grants, educating a new generation about their work, and connecting with a school community, including students, school staff, and parents. We considered visionary employees as those who chose one of the latter two options.

Dependent Variables

Serving constituents

Regarding the first objective of CBOs to enter into partnerships, serving constituents, respondents indicated whether or not they had engaged students in a service project and, on a scale from 1 to 5, the extent to which students had helped their organization in a way only children could, or in other words, the extent to which children gave a unique grant.

Education

Four distinct variables represented the education objective. First, respondents noted the number of times they had interacted with PH students, including talking on the phone, visiting the school for a presentation, giving a tour of the organization, or supervising a service project. The variable, students learned, was comprised by averaging scores on the extent to which respondents believed students had learned about (a) their organization and (b) the larger issue it addresses (α = .86). The variable, likelihood of continued support, was comprised by averaging scores regarding the extent to which respondents believed students would continue to (a) support their organization, (b) support the larger issue the organization addresses, and (c) be active in their communities (α = .81). The variable, job improvement, was comprised by averaging the scores on two items, the extent to which respondents believed the PH had made their day-to-day job more interesting and had made them believe more in the importance of their work (α = .85). Each of these latter three variables was measured on a 5-point Likert-type scale from 1 to 5 (1 = strongly disagree, 5 = strongly agree).

Friends and resources

Regarding the third objective, expanding access to potential supporters through partnerships with students and schools, organizations indicated whether they had advertised their partnership with students in their newsletter, on their website, in materials they sent to donors, or in a press release. Responses were summed to represent the variable, advertising. For example, if an organization had both advertised their grant in their newsletter and on their website, they would have received a score of two. Respondents also indicated whether they had communicated with parents, teachers, and/or school principals by dispersing information about their organization, making a presentation in front of them, and/or one-on-one conversations. Responses were summed to represent the variable, outreach to school community. For example, if an organization had reported making a presentation in front of teachers, having one-on-one conversations with principals, and dispersing information to parents, they would have received a score of 3.

Analytic Strategy

First, 36 incomplete cases were deleted through listwise deletion. Data were then checked for multicollinearity, homoscedasticity, and non-linearity. Study hypotheses were tested with a series of three-step hierarchical regression analyses for each of the eight dependent variables. The first step regressed the organizational capacity variables, organization size, number of staff involved, number of volunteers/constituents involved, and years of experience on one of the dependent variables. The second step regressed the employee vision variables, belief in children, perception of children, and favorite part of the PH, on one of the dependent variables. The third step regressed all of the variables on one of the dependent variables.

Different types of regression were used depending on the dependent variable type. We used multiple linear regression for the dependent variables, unique grant, belief that students learned, belief in continued student support, and job improvement. Although the dependent variables were technically measured on an ordinal scale, we chose to use linear regression. There is some debate regarding whether linear regression is acceptable for Likert-type items, but literature suggests that this method is acceptable when there are at least five response points, when assumptions of normality and equal variance of residuals were met, and when it is reasonable to assume that intervals between response points (strongly disagree, disagree, etc.) are equal (Glass, Peckham, & Sanders, 1972; Lubke & Muthen, 2004). The advantage of linear regression is its ease of interpretation.

We used Poisson regression for the variables, interaction with PH students, advertising of partnership, and school community outreach, as these variables represented counts of how many times organizations had engaged in certain activities (Cameron & Trivedi, 1998). Poisson regression is preferable to linear regression in this situation, because these variables were not normally distributed; in other words, the majority of the responses fell toward the low end of the distribution. Our data did not violate the required assumption that the mean and variance are equal; thus, we felt comfortable using Poisson regression rather than the negative binomial model.

Finally, we used binomial logistic regression (Menard, 2002) for the dependent variable, engaged students in service, as respondents responded yes or no to this question. Binomial logistic regression is the most common method for handling dichotomous outcome variables, as linear regression assumes a normal distribution. For the models fitted by Poisson and logistic regression, we exponentiated the resulting beta coefficients and reported incidence rate ratios (IRRs) and odds ratios, respectively. IRRs are interpreted as the rate by which the predictor variable increases or decreases the frequency of a certain activity.

We determined the best fitting model differently depending on the type of regression. For linear regression analyses, we determined the omnibus significance of each model and used an F test to compare significant models that were nested. For Poisson and logistic regression analyses, we calculated the Hosmer–Lemeshow pseudo R², which represents the difference between the null and residual deviance divided by the null deviance (Menard, 2002). In addition, we calculated a chi-square value, the difference between the null and residual deviance, for each model and compared these values through a chi-square test of difference. In cases where the chi-square test of difference was not significant, we retained the more parsimonious model.

Results

The descriptive statistics in Table 1 revealed that, overall, CBOs benefited from their partnerships. More than one third of CBOs engaged students in a service project, and on average, responses regarding whether students gave a unique grant fell in between neutral and agree (M = 3.43, SD = 1.03). Responses regarding the extent to which students learned (M = 4.11, SD = 0.65) and likelihood of continued student support (M = 4.12, SD = 0.58) fell between agree and strongly agree, on average, whereas reported job improvement due to the PH fell between disagree and neutral (M = 4.12, SD = 0.58). On average, CBOs interacted with students 7 times (SD = 8.33) over the course of the school year. CBOs advertised their grant in at least one way (SD = 1.11) and communicated about their organization in an average of one way (SD = 2.43) with parents, teachers, or principals.

Table 1.

Descriptive Statistics.

	M	SD	Range	%
Organization size	17.00	17.51	0-51+
Other staff involved in PH	1.13	1.01	0-4
Volunteers and/or constituents involved in PH (Y/N)				54.40
Years of experience	1.96	1.06	0-4
Belief that children are important community resources	3.87	0.88	1-5
Perception of children	4.17	0.56	2-5
Favorite aspect of PH
The grants				39.71
Educating a new generation about my work				30.88
Connecting with a school community				29.41
Service project (Y/N)				34.60
Unique grant	3.43	1.03	1-5
Job improvement due to PH	2.33	1.05	1-5
Likelihood of continued student support	4.12	0.58	2-5
Likelihood that students learned	4.11	0.65	1.5-5
Interaction with PH students	7.00	8.33	0-44
Advertising of PH grant	1.39	1.11	1-5
Outreach to school community	0.99	2.43	0-13

Note. N = 129. PH = Penny Harvest.

Qualitative Results

In their open-ended responses, many CBOs provided qualitative evidence that they were accomplishing the three objectives of partnering with schools (Batenburg, 1995). Many CBOs reported that children helped their organization in ways that “only children could.” For example, several discussed the positive impact of interacting with youth on their constituents, particularly on the elderly and the disabled. Children were described as bringing laughter and cheer to these populations. When asked what was unique about their PH grant, one CBO responded, “Because it came from children! In fact, it was not so much the grant itself but the participation of the children in the service project that breathed new life into our programs.” Another CBO admitted that the contribution from children was unique because, “They add a sense of creativity and fun to all that they participate in.”

CBOs also discussed grants they believed were unique to children. For example, a representative from a senior center reported that PH students had recognized a lack of fun activities for seniors and thus purchased a karaoke machine for the senior center. Students and seniors then sang karaoke together. Many CBOs reported that their PH grants helped them pay for the “little things” that adults often overlook, such as paying for children afflicted with cancer to attend sporting events with their families, purchasing real silverware at a homeless shelter to help preserve the dignity of residents, and throwing birthday parties for hospitalized children. In such desperate financial times, adult donors typically concentrate funding on absolute necessities, yet children tended to recognize and fund the small projects that enable organizations to offer their constituents something special.

When asked to describe if and how PH students had remained involved with their organization, many organizations did not have mechanisms for tracking sustained involvement. However, several mentioned that PH students had continued volunteering and supporting their organization into high school and college. One large national organization reported receiving inquiries regarding volunteer opportunities from PH alumni all the time. Another CBO mentioned that students kept in touch with the organization through Facebook. In addition, there was substantial evidence that CBOs devoted time and energy to the PH because they hoped for continued support from a new generation. One CBO described the most impactful aspect of the PH: “This is a great way to encourage children to become social entrepreneurs. The next George Soros could be sitting in the classrooms we reach. They will remember us when they grow up.” Many CBOs also discussed the importance of students learning about their organization and their cause. For example, one CBO described the most impactful part of the PH as children learning about asthma and thus enrolling in the CBO’s school-based asthma program.

Several CBOs also described the personal impact of the PH when asked to describe the greatest impact of the program. Describing the day when a check from PH students arrived in the mail, a CBO representative said, “We were both overwhelmed with just the thought that these kids appreciate what it is we do and why we do it.” Other organizations described their experiences working with PH students as “heartwarming,” “touching,” “moving,” “refreshing,” and “spirit-filled.” PH students also seemed to inspire CBO representatives, whose work often addresses vast social problems, to continue to hope for a better future. For example, one CBO said, “Showing [children] that there is more to life than the ‘self’ alone instills sensitivity in the future generation, and therefore hope for tomorrow. The children have brought back sensitivity and awareness to a new level.” Another respondent asked, “If children can produce this kind of outcome, how much more can adults do?”

CBOs described how PH grants had helped them appeal to their donors and expand their support networks. Several CBOs mentioned that their donors were interested to hear that the organization was diversifying its funding sources and were impressed by youth involvement. CBOs reported engaging their trustees and board members in the PH through asking them to deliver presentations to students and attend end-of-year celebrations. One CBO said that after PH students had awarded a grant, they also sang at a Board of Trustees meeting, energizing the trustees as well as exposing students to an important aspect of the non-profit world. In addition, many CBOs described the opportunity to reach out to parents, teachers, and principals associated with the school community as one of the primary benefits of the PH. Not only did direct outreach from CBOs to these populations garner additional support, but organizations also described how PH students inspired adults associated with the school community to also support their organizations through volunteering and fundraising. In addition, CBOs reported that more families became aware of their services after they had received a PH grant, and the grant was often a launching pad for many organizations to expand their programming and services into more schools or even create new programs in schools.

Quantitative Results

The purpose of the quantitative data was to understand what factors contributed to the ability of CBOs to benefit from the outcomes detailed above. We were specifically interested in whether organizational capacity or employee vision had a greater impact on the ability of CBOs to provide better services for their constituents, educate a new generation, and access new friends and resources.

Objective 1: Better serving constituents

First, we tested which factors related to the ability of CBOs to provide a better service to their constituents. The results are displayed in Table 2. Regarding whether CBOs engaged students in service, Model 1, which included the organizational capacity variables, significantly explained 9.1% (χ² = 15.79, 4, ρ < .01) of the variance, whereas Model 2, which included the employee vision variables, only explained a non-significant 2.89% of the variance. The full model significantly explained 11% (χ² = 19.05, 8, ρ < .05) of the variance. According to a chi-square test of difference, the difference between Models 1 and 3 was not significant; thus, Model 1 was retained as the best fit. In Model 1, significant predictors included whether or not volunteers and/or CBO constituents were involved in the PH and the respondent’s number of years of experience with the PH. Controlling for organizational size, organizations that engaged volunteers and/or constituents were 3.1 times (ρ < .05) more likely to engage PH students in service. In addition, the odds that a CBO engaged students in service increased by about 50% (ρ < .05) for every additional year of respondent experience. In regard to Model 2, surprisingly, a higher perception of children was significantly associated with about a 50% (ρ < .05) reduced odds of engaging students in service. None of the variables we tested significantly predicted whether organizations believed the grants they received were unique to children; thus, we do not display the results. In sum, the organizational capacity variables were a stronger predictor of the ability to better serve constituents.

Table 2.

Hierarchical Logistic Regression Results From Predicting Whether Organization Engaged Students in Service Projects.

	Service project
	Model 1		Model 2		Model 3
	OR	SE	B	SE	B	SE
Intercept	0.13**	0.57	6.24	1.51	1.04	1.74
Organization size	1.00	0.01	—	—	1.00	0
Other staff involved in PH	0.99	0.21	—	—	0.99	0.2
Volunteers/constituents	3.10*	0.47	—	—	3.07*	0.5
Years of experience	1.47*	0.18	—	—	1.45*	0.2
Perception of children	—	—	0.48*	0.35	0.53^†	0.38
Belief in children	—	—	1.10	0.22	1.14	0.24
Favorite aspect of PH
Connecting with school community	—	—	1.29	0.45	1.21	0.48
Educating about work	—	—	1.30	0.46	0.97	0.50
Pseudo R²	.091		.029		.110
Model χ²	15.8**		n.s.		19.1*

Note. N = 129. OR = odds ratio; PH = Penny Harvest.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Objective 2: Educating about work

Next, we tested whether organizational capacity or employee vision better predicted to what extent CBOs were able to educate them about their work, garner the support of a new generation, become more energetic about their jobs, and interact with students. The results of linear regression analyses for the first three of these variables are displayed in Table 3, and the results for Poisson regression analysis for number of interactions with students are displayed in Table 4.

Table 3.

Hierarchical Linear Regression Results From Predicting Belief That Students Learned, Belief in Continued Student Support, and Job Improvement.

	Belief that students learned						Belief in continued student support						Job improvement due to PH
	Model 1		Model 2		Model 3		Model 1		Model 2		Model 3		Model 1		Model 2		Model 3
	B	SE	B	SE	B	SE	B	SE	B	SE	B	SE	B	SE	B	SE	B	SE
Intercept	3.96***	0.15	2.35***	0.43	2.12***	0.47	4.14***	0.14	2.30***	0.38	2.31***	0.42	2.37	0.25	4.32***	0.72	4.52***	0.79
Organization size	0.00	0.00	—	—	0.00	0.00	0.00	0.00	—	—	0.00	0.00	0.00	0.00	—	—	0.00	0.00
Other staff involved in PH	−0.05	0.06	—	—	−0.04	0.06	−0.05	0.06	—	—	−0.03	0.05	−0.02	0.22	—	—	−0.06	0.10
Volunteers/constituents	−0.03	0.14	—	—	−0.01	0.13	0.03	0.12	—	—	0.02	0.12	−0.06	0.22	—	—	−0.08	0.22
Years of experience	0.11	0.05	^† —	—	0.12*	0.05	0.00	0.05	—	—	0.01	0.05	0.01	0.09	—	—	−0.02	0.09
Perception of children	—	—	0.36***	0.10	0.38***	0.10	—	—	0.05	0.05	0.37***	0.06	—	—	−0.48**	0.16	−0.50**	0.17
Belief in children	—	—	0.03	0.06	0.03	0.06	—	—	0.37***	0.09	0.05	0.09	—	—	−0.05	0.10	−0.06	0.11
Favorite aspect of PH
Connecting with school community	—	—	0.23^†	0.13	0.23^†	0.13	—	—	0.14	0.11	0.15	0.11	—	—	0.50*	0.21	0.52*	0.22
Educating about work	—	—	0.16	0.13	0.15	0.13	—	—	0.18	0.11	0.18	0.12	—	—	0.23	0.22	0.27	0.23
R ²	.003		.147		.183		.001		.179		.182		.000		.096		.103
Model F	1.08		5.54***		3.5***		0.20		7.04***		3.47**		0.06		3.41**		1.8

Note. N = 129. PH = Penny Harvest.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table 4.

Hierarchical Poisson Regression Results From Predicting Staff Interaction With PH Students.

	Interaction with PH students
	Model 1		Model 2		Model 3
	IRR	SE	IRR	SE	IRR	SE
Intercept	1.47***	0.08	9.96***	0.26	9.58***	0.26
Organization size	1.01***	0.00	—	—	1.01***	0.00
Other staff involved in PH	1.14***	0.03	—	—	1.12**	0.03
Volunteers/constituents	1.74***	0.08	—	—	1.77***	0.09
Years of experience	1.49***	0.03	—	—	1.49***	0.03
Perception of children	—	—	0.98	0.06	0.89^†	0.06
Belief in children favorite aspect of PH	—	—	0.89**	0.04	0.91*	0.04
Favorite aspect of PH
Connecting with school community	—	—	1.11	0.08	1.00	0.08
Educating about work	—	—	1.31***	0.08	1.51	0.08
Pseudo R²	.284		.018		.291
Model χ²	327.3***		n.s.		335.1***

Note. N = 129. PH = Penny Harvest; IRR = incidence rate ratio.

†

p < .10. *p < .05. **p < .01. ***p < .001.

As displayed in Table 3, employee vision significantly explained the variance in belief that children learned, R² = .15, F = (4, 129) = 5.54, ρ < .001; belief in continued student support, R² = .18, F = (4, 129) = 7.04, ρ < .001; and job improvement due to the PH, R² = .10, F = (4, 129) = 3.41, ρ < .01. Organizational capacity did not significantly predict any of the education variables. According to the chi-square test of difference, Model 2 was not significantly different from Model 3 for any of the dependent variables; thus, the model with employee vision alone was a better fit than the full model.

There were significant positive relationships between respondents’ perceptions of children and belief that students learned (B = 0.36, p < .001) and belief that children are important community resources and likelihood of continued support (B = 0.37, p < .001). In addition, compared with CBOs who identified receiving grants as their favorite part of the PH, on average, individuals who noted connecting with a school community as their favorite part of the PH scored .50 (p < .05) points higher regarding their belief that the PH had improved their job. Surprisingly, a stronger perception of children had a significant negative association with respondents’ beliefs that their job had improved due to interacting with PH students (B = −0.48, p < .01). The only organizational capacity variable that significantly predicted one of these three outcomes was years of experience, which significantly predicted belief that students learned in the full model (B = 0.12, p < .05).

In contrast, organizational capacity was more strongly related to the frequency with which CBOs interacted with PH students, as displayed in Table 4. The organizational capacity variables explained 28.4% (χ² = 327.29, 4, ρ < .001) of the variance in frequency of interaction; the employee vision variables only explained a non-significant 1.8% of the variance. Model 1 was retained as the best fitting model. Every unit increase in organization size, staff involved in the PH, and years of experience, respectively, increased the expected number of times CBOs interacted with PH students by 1% (p < .001), 14% (p < .001), and 49% (p < .001), respectively. Organizations that involved volunteers and/or constituents in the PH were expected to interact with students 75% more than those that did not engage volunteers and/or constituents (p < .001). Regarding employee vision, compared with individuals whose favorite part of the PH were the grants, respondents who most enjoyed educating a new generation about their work were expected to interact with students on over 30% more occasions (p < .001). Surprisingly, the more respondents believed that children were important community resources, the less they interacted with students. Although this result was statistically significant, the effect size was small (λ = .89, p < .01). In sum, employee vision was a stronger predictor of beliefs regarding the benefits of service-learning as a tool for educating youth about one’s work, whereas organizational capacity was a stronger predictor of actual interactions with young people.

Objective 3: Access to new friends and resources

Finally, we tested which factors were related to an organization’s ability to leverage its partnership into new publicity opportunities, as displayed in Table 5. None of the predictors were related to the number of ways in which organizations advertised their grant or partnership, and both models explained a negligible proportion of variance; thus, we do not display the results for these models. In regard to the number of ways in which CBOs reached out to parents, teachers, and principals, organizational capacity explained 12.9% (χ² = 50.1, 4, ρ < .001) of the variance, and the full model explained 14.3% (χ² = 54.5, 8, ρ < .001) of the variance. The chi-square for Model 2 was not significant, and Model 1 was retained as the best fitting model. For every additional staff member involved in the PH, the frequency of outreach increased by a factor of nearly 30% (p < .001), and CBOs that involved volunteers and/or constituents in the PH experienced a higher rate of outreach frequency by a factor of 70% (p < .01). Belief in children surprisingly had a marginally significant negative association with outreach, decreasing the rate of outreach by a factor of 13% (p < .01). In sum, organizational capacity was a stronger predictor of community outreach than employee vision.

Table 5.

Hierarchical Poisson Regression Results From Predicting Number of Ways in Which CBO Reaches Out to School Community.

	Outreach to school community
	Model 1		Model 2		Model 3
	IRR	SE	IRR	SE	IRR	SE
Intercept	0.83	0.2	1.69	0.52	0.49	0.59
Organization size	1	0	—	—	1	0
Other staff involved in PH	1.3***	0.06	—	—	1.27***	0.1
Volunteers/constituents	1.69**	0.16	—	—	1.79***	0.2
Years of experience	1.03	0.06			1.03	0.1
Perception of children	—	—	1.15^†	0.1	1.24	0.1
Belief in children	—	—	0.87^†	0.1	0.90^†	0.1
Favorite aspect of PH	—	—
Connecting with school community	—	—	1.08	0.2	0.99	0.2
Educating about work	—	—	1.13	0.2	0.95	0.2
Pseudo R²	.129		.015		.143
Model χ²	50.1***		n.s.		54.5***

Note. N = 129. CBO = community-based organization; PH = Penny Harvest; IRR = incidence rate ratio.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Discussion

The purpose of this article was twofold. First, we sought to understand to what extent organizations that had received grants from a youth philanthropy program maximized the impact of working with children, engaged a new generation in their work, and leveraged the grant into further opportunities for publicity. Second, we aimed to explore whether organizational capacity or employee vision regarding partnering with children and schools better predicted these outcomes. This article contributes to the literature on service-learning and to broader literature on school/community partnerships in several ways. First, it is one of only a few studies to examine community-level impacts of service-learning (Imperial et al., 2007), particularly at the K-12 level. Second, this study extends prior research, which has only investigated organizational-level predictors of engaging students in volunteering; we also examined broader CBO outcomes of service-learning, such as educating children about their work and garnering additional friends and resources. A more complex understanding of how CBOs benefit from service-learning is critical given service-learning’s focus on meaningful community impact and its potential to forge mutually beneficial relationships between schools and community organizations (Sanders, 2003).

Our descriptive and qualitative data indicated that many organizations maximized the impact of their PH grants in multiple ways. Due to partnerships with PH schools, CBOs were able to continue funding special projects that adults had neglected; infused energy into their constituents through engaging youth in service projects; were personally inspired by the generosity of children; motivated students to continue supporting their organizations into adulthood; impressed donors by garnering support from young people; and expanded their support networks by including parents, teachers, and principals. These data offer a testament to the capabilities of children and the reciprocal benefits of school/community partnerships. Although these outcomes would be helpful for CBOs in any context, they may be particularly valuable in urban settings, where school/community partnerships have historically been enacted from a deficit perspective (Tyack, 1992). When youth contribute to organizations in a variety of tangible and meaningful ways, school/community partnerships emphasize the strengths—rather than the deficits—of low-income urban youth. As our data show, the visibility of these strengths inspired many respondents. Although we cannot claim that service-learning partnerships actually influenced job performance or productivity, organizational literature suggests that “affective commitment”—emotional identification with and attachment to the organization—predicts job performance (Meyer, Paunonen, Gellatly, Goffin, & Jackson, 1989). Future research may explore the influence of these “softer” outcomes on actual job performance.

However, many CBOs received grants without forming a meaningful partnership with students or leveraging their grant into additional organizational benefits. We hoped to understand why certain CBOs were more successful at this than others by exploring whether organizational capacity or vision had a greater impact. Organizational capacity was more predictive of whether CBOs engaged students in service, interacted with students, and reached out to the school community. However, employee vision was more predictive of the extent to which CBOs believed students learned about the organization, would support the organization in the future, and the extent to which the respondent’s job had improved as a result of the PH. These results indicate that both organizational capacity and employee vision are important, yet accomplish different goals. If CBOs do not have the capacity to frequently interact with students or reach out to the community, a visionary employee may still effectively energize youth about the work of the organization. Nevertheless, this research corroborates previous research regarding the importance of organizational capacity for successful engagement in service-learning partnerships (Sandy & Holland, 2006). Delivering presentations to students, offering organizational tours, attending school-based events, and engaging students in service projects all take time and human resources. Contrary to the findings of Littlepage et al. (2012), organizational capacity still mattered when controlling for employee vision. However, experience of the respondent and the number of engaged volunteers/constituents were consistently stronger predictors than organizational size, suggesting that smaller organizations still can have the capacity to benefit from service-learning.

Thus, the main conclusion that we draw from these findings is that a combination of organizational capacity and employee vision is likely optimal. Although this finding may seem intuitive, this research potentially has practical implications as well. The most consistent significant predictor of organizational outcomes was engaging volunteers or constituents in the service-learning partnership. This was significant even when controlling for organizational size. Thus, a visionary employee who believes in the potential of service-learning coupled with a stable volunteer base that can engage students and reach out to the community may be a particularly effective combination. Service-learning may be a strategic way for CBOs to engage baby boomers who are predicted to volunteer more than previous generations (Einolf, 2009), but desire more “skills-based volunteering” opportunities in lieu of more traditional activities such as stuffing envelopes (Holland & Robinson, 2008). Future research should further investigate the relationship between engaging volunteers/constituents and the organizational benefits of service-learning, as our data are cross-sectional and the relationship between engaging volunteers and organizational benefits may be spurious. For example, there may be an underlying variable that renders CBOs more able to effectively engage volunteers/constituents and benefit from service-learning.

Several surprising findings merit further discussion. Stronger perceptions of PH students were associated with lower beliefs among respondents that their jobs had improved as a result of the PH. During focus groups and interviews we conducted as a precursor to survey development, many respondents revealed how inspired they became due to being surprised that young people could make such a powerful difference. Thus, there may have been less room for job satisfaction improvement among respondents who already had positive perceptions of children. In addition, stronger perceptions of children was associated with lower odds that a CBO would engage students in service, and stronger beliefs that students were important community resources were associated with significantly fewer occasions on which CBOs had interacted with PH students. First, it is important to note that both of these independent variables were positively skewed, so very few CBOs actually had negative perceptions of children. Most of the variation was in regard to whether respondents reported to be neutral, to agree, or to strongly agree. Despite this statistical caveat, there is a possibility that more engagement with students predicted lower perceptions of them. Although CBOs unanimously reported positive experiences working with PH students in their open-ended responses, perhaps the more CBOs interacted with children, the more likely they were to encounter a more negative experience, even if the majority of their experiences were positive. These findings potentially speak to the challenges embedded in navigating school/community partnerships discussed earlier (Batenburg, 1995), including different schedules, goals, and expectations. Although this study illustrates the diverse rewards many CBOs reap from engaging in these partnerships, our data also suggest the positive potential of engaging volunteers and/or constituents who may be able to devote more time and energy than full-time staff to building relationships with schools.

Limitations and Implications

This study had several limitations. First, as with all cross-sectional data, we were unable to make causal conclusions. Thus, all statistically significant relationships mentioned in this article are merely correlational. For example, it is impossible to conclude from our data that employee vision temporally precedes the ability for CBOs to effectively educate about their work. A positive experience educating about one’s work may contribute to being visionary about the potential of service-learning, or alternatively, there may be an underlying variable that contributes both to employee vision and education.

Second, the response rate was only around 30%, so we cannot assume our sample was representative of all CBOs that received PH grants. Thus, the percentage of organizations that were able to maximize the impact of their grants was likely overestimated. We considered random sampling, but including every organization that had received a grant that year enabled Common Cents to learn about best practices through the open-ended responses, and we decided that the benefits to the organization exceeded the costs to the research design. The organizations in our sample still represent a large variety of sizes, types, and issue areas. Furthermore, the primary goal of this study was to understand how organizations maximized the impact of their partnerships, rather than the extent to which they did so. A final limitation is that the models predicting advertising the grant and the extent to which students made a unique grant explained a negligible percentage of the variance. Thus, future research should explore what other factors may be relevant to these variables. Qualitative research may initially serve this purpose best.

Despite these limitations, this study is an important step toward better understanding how community partners can optimally benefit from service-learning and youth philanthropy. Although there are a few useful guides to help CBOs who plan to partner with schools in service-learning projects (see Abravanel, 2003; Cairn, 2003; Roehlkepartain & Bailis, 2007), they rarely offer advice regarding how to maximize the partnership for the benefit of the organization. More research should probe qualities of service-learning partnerships that mutually benefit youth and CBOs, and this research should be incorporated into such guides. This study gives us hope to believe that all organizations, regardless of size, can successfully leverage the benefits of being a community partner but may just need some extra assistance.

Against the backdrop of budget cuts for schools and anti-poverty programs, the ability for schools and communities to share and maximize their resources is as important as ever. This study suggests that when students make a grant or conduct a service project for a community organization, there is a resulting “boomerang effect.” That is, adults—non-profit staff, community members, parents, teachers, principals, and so on—become inspired to give back to youth, investing in their educational, social, and civic development. They form what Benson (2003) calls an asset-building community, in which parents, educators, and community members work together to invest in the well-being of young people. These dynamic and reciprocal relationships between youth and schools and the surrounding community have the potential to bridge youth development and community development in urban settings. We hope that this article spurs further discussion and research regarding the potential for this process to flourish.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Joanna D. Geller is a PhD candidate in community research and action at Vanderbilt University and is also a senior research associate at the Annenberg Institute for School Reform at Brown University. Her action research focuses on providing youth, families, community stakeholders, and teachers with a voice in their schools and communities.

Natalie Zuckerman is the officer of community partner relations at New York Cares in New York City. She has over 5 years of experience in non-profit and education program development, evaluation, and implementation, including as a community development fellow at Common Cents. She holds a Master’s in Education in Prevention Science and Practice from the Harvard Graduate School of Education, and a bachelor of science in applied psychology from New York University.

Adam Seidel is the director of operations at Roxbury Prep Lucy Stone Campus, a growing middle school located in Dorchester, MA. Prior to this role, he served on the founding staff as the director of special projects at another Roxbury Prep middle school. Before working directly in schools, he worked at Common Cents for 9 years, serving multiple roles, including managing director, national program director, and community development fellow. He holds a Bachelor of Arts and Sciences in Politics from Oberlin College and a Master’s in Public Administration from Baruch College.

References

Abravanel

S. A.

(2003). Building community through service-learning: The role of the community partner (Education Commission of the States Issue Paper). Retrieved from http://www.ecs.org/clearinghouse/44/03/4403.pdf

Basinger

Bartholomew

(2006). Service-learning in nonprofit organizations: Motivations, expectations, and outcomes. Michigan Journal of Community Service Learning, 12, 15-26.

Batenburg

(1995). Community agency and school collaboration: Going in with your eyes open (ERIC Document Reproduction Service No. ED 383901). Annual Meeting of the Educational Research Association. Available from http://www.eric.ed.gov/

Benson

P. L.

(2003). Developmental assets and asset-building community: Conceptual and empirical foundations. In Lerner

R. M.

Benson

P. L.

(Eds.), Developmental assets and asset-building communities: Implications for research, policy, and practice (pp. 19-45). New York, NY: Plenum.

Billig

S. H.

Root

Jesse

(2005). The relationship between quality indicators of service-learning and student outcomes: Testing professional wisdom. In Root

Callahan

Billig

S. H.

(Eds.), Improving service-learning practice: Research on models to enhance impacts (Advances in service-learning research) (Vol. 5, pp. 97-115). Greenwich, CT: Information Age.

Bridgeland

J. M.

Dilulio

J. J.

Burke Morrison

(2006). The silent epidemic: Perspectives of high school dropouts. Washington, DC: Civic Enterprises. Retrieved from http://files.eric.ed.gov/fulltext/ED513444.pdf

Cairn

(2003). Partner power and service-learning: Manual for community-based organizations to work with schools. Minneapolis: ServeMinnesota. Retrieved from http://www.servicelearning.org/filemanager/download/118/MNCBOManual2003.pdf

Cameron

A. C.

Trivedi

P. K.

(1998). Regression analysis of count data. Cambridge, UK: Cambridge University Press.

Christens

B. D.

Dolan

(2011). Interweaving youth development, community development, and social change through youth organizing. Youth & Society, 43, 528-548

10.

Edwards

Mooney

Heald

(2001). Who is being served? The impact of student volunteering on local community organizations. Nonprofit and Voluntary Sector Quarterly, 30, 444-461.

11.

Einolf

(2009). Will the boomers volunteer during retirement? Comparing the baby boom, silent, and long civic cohorts. Nonprofit and Voluntary Sector Quarterly, 38, 181-199.

12.

Eyler

Giles

(1999). Where’s the learning in service-learning? New York, NY: Jossey-Bass.

13.

Family Strengthening Policy Center. (2007). Youth service-learning: A family-strengthening strategy. Washington, DC: National Human Services Assembly.

14.

Follman

Muldoon

(1997). Florida Learn & Serve 1995-96: What were the outcomes? Bulletin of the National Association of Secondary School Principals, 81(591), 29-36.

15.

Furco

(2001). Is service-learning really better than community service? A study of high school service program outcomes. In Furco

Billig

(Eds.), Service learning: Essence of pedagogy (pp. 23-50). Greenwich, CT: Information Age Publishers.

16.

Glass

G. V.

Peckham

P. D.

Sanders

J. R.

(1972). Consequences of failure to meet assumptions underlying the analyses of variance and covariance. Review of Educational Research, 42, 237-288.

17.

Holland

Robinson

(2008). Community based learning with adults: Bridging efforts in multiple sectors. New Directions for Adult and Continuing Education, 118, 17-30.

18.

Imperial

M. T.

Perry

J. L.

Katula

M. C.

(2007). Incorporating service learning into public affairs programs: Lessons from the literature. Journal of Public Affairs Education, 13(2), 243-264.

19.

Leiderman

Furco

Zapf

Goss

(2002). Building partnerships with college campuses: Community perspectives (Consortium for the Advancement of Private Higher Education). Retrieved from depts.washington.edu/ccph/pdf_files/engaging_monograph.pdf

20.

Littlepage

Gazley

Bennett

T. A.

(2012). Service-learning from the supply side: Community capacity to engage students. Nonprofit Management & Leadership, 22, 305-320.

21.

Lubke

G. H.

Muthen

B. O.

(2004). Applying multigroup confirmatory factor models for continuous outcomes to Likert scale data complicates meaningful group comparisons. Structural Equation Modeling, 11, 514-534.

22.

Males

(2006). Youth policy and institutional change. In Ginwright

Noguera

Cammarota

(Eds.), Beyond resistance! Youth activism and community change (pp. 301-318). New York, NY: Rutledge.

23.

Melchior

(1999). National evaluation of Learn and Serve America school and community-based programs. Washington, DC: The Corporation for National Service.

24.

Menard

(2002). Applied logistic regression analysis. Thousand Oaks, CA: Sage.

25.

Meyer

J. P.

Paunonen

S. V.

Gellatly

I. R.

Goffin

R. D.

Jackson

D. N.

(1989). Organizational commitment and job performance: It’s the nature of the commitment that counts. Journal of Applied Psychology, 74, 152-156.

26.

National Service-Learning Clearinghouse. (n.d.). What is service-learning? Retrieved from http://www.servicelearning.org/what-service-learning

27.

Roehlkepartain

E. C.

Bailis

L. N.

(2007). Developing partnerships for service-learning: Starting points for community-based organizations. Retrieved from http://www.servicelearning.org/instant_info/fact_sheets/cb_facts/developing_partnerships

28.

Sanders

M. G.

(2003). Community involvement in schools: From concept to practice. Education and Urban Society, 35, 161-180.

29.

Sandy

Holland

(2006). Different worlds and common ground: Community partner perspectives on campus-community partnerships. Michigan Journal of Community Service Learning, 13, 30-43.

30.

Scales

Blyth

Berkas

Kielsmeier

(2000). The effects of service learning on middle school students’ social responsibility and academic success. Journal of Early Adolescence, 20, 331-358.

31.

Stone

Doherty

Jones

Ross

(1999). Schools and disadvantaged neighborhoods: The community development challenge. In Ferguson

R. F.

Jones

W. T.

(Eds.), Urban problems and community development (pp. 339-380). Washington, DC: Brookings.

32.

Tyack

(1992). Health and social services in public schools: Historical perspectives. Future of Children, 2, 19-30.

33.

Worrall

(2007). Asking the community: A case study of community partner perspectives. Michigan Journal of Community Service Learning, 14(1), 5-17.