DRAFT

Geographic Disparities in Access to Service Learning[1]

Prepared for the 2008 Rural Sociological Society Annual Meetings

 

Randy Stoecker and Charity Schmidt

University of Wisconsin


 

Introduction

This project is part of a broader project to understand how service learning impacts community.  It began with a graduate seminar interviewing 64 Madison area community organization staff to understand their reactions of service learning (Stoecker and Tryon, forthcoming).   We used a broad definition of service learning that included any student performing a service for a small to medium size community organization for credit or as part of a program requirement (to include interns or practicum students).  A small to medium non profit was defined as having roughly less than a $1 million annual budget and/or twelve or fewer full-time staff.  Nonprofits then used the findings during subsequent focus groups to create a brochure entitled “Community Standards for Service Learning.” The process and findings are exhibited at http://comm-org.wisc.edu/sl.

The research project identified a number of concerns about service learning.  Community organization staff expressed a need to improve communication between organizations hosting service learners and higher education faculty.  Developing positive relationships was a second important theme for the agencies, who offered numerous suggestions for involved faculty.  A third focus was on providing aninstitutional infrastructure for service learning.  A fourth set of concerns addressed the problem of short-term service learning.  And a fifth topic focused on managing service learners, identifying the responsibilities of students, faculty and non-profit staff. The final theme highlighted challenges presented by varying economic, racial/ethnic and cultural backgrounds between students and service populations.  Many of the challenges described by nonprofit staff are rooted in the unequal standing between higher education and community organizations and  the prioritization of the student experience over community impact (Stoecker and Tryon, forthcoming).

This pilot project led to deeper concerns over the service learning programs across the state.  If service learning has the potential to burden small to medium nonprofits and produce negative externalities in urban communities, how might it impact our rural communities?  This shift in focus from city to state resulted in the current two-year project to understand what service learning programs look like in Wisconsin and who they are serving, as well as to assess their impact on community nonprofits and the people they intend to benefit. Our first step in that process was identifying whether rural community organizations were even getting access to service learning.

What do we know about Urban-Rural differences in Service Learning?

As much as we would like to think that service learning serves communities, we have very little evidence that is actually the case (Cruz and Giles, 2000).  In fact, the overwhelming majority of our research on service learning has focused on the institutional side of the relationship, especially on how service learning impacts students.  There are some community satisfaction studies (Vernon and Ward, 1999; Ferrari and Worrall, 2000; Birdsall, 2005).   But the articles that evaluate community reactions in more depth (Vernon and Ward, 1999; Jones 2003; Birdsall, 2005; Muzak and Woollard, 2008 Sandy and Holland 2006; Worrall 2007), while increasing in number, remain very few.

Rural service learning has also been neglected in the literature (see Mihalynuk et al., 2007).  Authors consider  the general issues facing service learning in rural contexts (Holton, 2007), guides for doing service learning in rural areas (Boethel, 1999), and in one case study community reactions to service learning (Vernon and Ward, 1999). They have not, however, looked at how much service learning has a presence in rural areas. We thus know of no research delineating the geographic spread of service learning. This is not surprising, given the lack of research on the community side of service learning in general.  The best information available is from the annual Campus Compact (2007) survey of service learning that found agriculture and rural environment listed in the “other category” of issues that students addressed, indirectly suggesting that only a tiny percentage of service learning was being conducted in rural areas.

Of course, it seems logical to suspect that there would be less service learning in rural areas, given that there is less population and fewer nonprofit and governmental organizations.  But we know little detail about the geographic distribution of service learning.  Are rural organizations as likely to get access to service learning as urban organizations?  Are rural colleges and universities as likely to send out service learners?  This paper will take a look at those questions, and the answers will hold some surprises.

Methodology

The definition of service learning that developed from the Madison project interviews and on which the state-wide project was based (as described above) is a simplified and bureaucratic one.  Most higher education advocates identify the element of “reflection” as fundamental to distinguishing between service learning and service.  However, as our past research shows, nonprofit staff are unaware of whether or not there exists a reflection component in the classroom.  However, because we were dependent on higher education service learning and other staff for data, we were often dependent on whatever definition they used in creating or maintaining their own databases.

We chose to focus on small to medium nonprofits. Those organizations, because of their lack of resources, have both the greatest potential to benefit from service learning or be harmed by it. For these reasons, we also excluded service learning hosted directly by public schools and governmental departments.

Determining which nonprofits had access to service learners proved to be more challenging than expected.  Not only is service learning often carried out without reference to community impact, we found that higher education institution  representatives rarely know who their community partners are.  There is seldom a service learning office or coordinator that documents and tracks such programs.  When this project began, only three of the fifty-two targeted higher education institutions in Wisconsin had existing lists of community partners engaged in service learning. As of Fall, 2007, a year after the Americorp*VISTA initiatives began, only ten of the institutions had a service learning or related office, or a staff member designated to coordinate projects and information. 

A partial solution to this dilemma was to utilize Americorp*VISTA members as on-site resources in order to obtain lists of community partners for every higher education institution in Wisconsin .  The current project began in 2006, the same year that the Americorp*VISTAS began their initiatives in mobilizing more student-community engagement (Smith, 2008; Corporation for National and Community Service, 2007).  The initiative placed them in service learning coordinator positions.  This dynamic sometimes benefited the research by providing new staff resources to create the lists.  However, it also complicated the data collection process, making it difficult to determine the state of service learning institutionalization prior to Americorp*VISTA placements and generating a potential for unequal representation among educational institutions. Additionally,  some campuses have numerous Americorp*VISTA members and others have none, and placements change on an annual basis.  In addition, institutional reliance upon Americorp*VISTA members for the development of entire service learning programs not only introduces inconsistency into the data collection but often leads to programs that only exist for the duration of the VISTA placement.

Such data is difficult to create after the fact when there is no system in place to do so. When Americorp*VISTA members were not present, other resources for the data included service learning offices, higher education staff, faculty and department chairs and institutional websites.  Such variation rendered the data problematic.  While many campuses had existing lists of potential community partners, they did not have lists of actual community partners, and we had to contact the potential partners in an attempt to determine which of them had actually hosted service learners.  It was also difficult to capture data on service learning in a common timeframe, as some campuses  provided data for one  targeted academic year as requested, while others could only proud de data for other yeans.

We ultimately collected data from 27 of 52 targeted higher education institutions in Wisconsin.  We were most successful with the public universities, and then the private institutions.  We were less successful with the UW Colleges, and Wisconsin technical colleges (see Table 1).  In many cases, officials at the colleges and technical schools explained that they had no service learning program, though they believed there was service learning occurring.  We agree that there likely was some service learning still occurring, but probably in small numbers. 

Table 1:  Higher Education Institutions Contributing Data

Total

Number Contributing Data

UW universities

13

12

UW colleges

13

4

Technical colleges

6

2

Private institutions

20

9

 

Our goal of developing a complete database of all the small and medium-size nonprofit organizations receiving service learning in 2005-2006 could thus not be achieved.  The incompleteness of the data for that year meant that fwe could only get data for 2006-7 from some institutions and, in a few cases, for 2007-8.  Our suspicion is that our database may be too small for some institutions and too large for others, but that the differences in collection periods is unlikely to affect findings about the geographic spread.

The lack of systems for tracking the community impact of service learning conducted by higher education institutions shows the bias of service learning to serve the institution rather than the community.  If higher education is not documenting service learning projects, what does this say about higher education’s perspective on student engagement with communities?  And if higher education doesn’t document those projects, how can we assess the impact that service learning is having on our communities?  Faculty carefully track the students in service learning through mandatory grading and course evaluation practices.  Service learning courses are designated as such in timetables and such requirements are documented for each student.  Thus, it should only be a matter of the labor needed to get the existing records of every service learning student.  But the organizations that host those service learners are invisible; they don’t exist in those records.  It is not important to know the organizations when the focus is on the students.  And the issue is not, as suggested by one service learning educator in Wisconsin, that the state has such a strong culture of service that everyone considers it natural rather than an activity that requires tracking.  The lack of higher education administrative support, and consequently the lack of service learning, shows that service learning, at least, is not part of the state culture.  If the higher education faculty did care about truly making a difference, they would do the necessary evaluation to find out the extent to which their service learning programs were having the desired impact.  We are encouraged by the number of institutions who began tracking the community organizations receiving service learning during the course of this research, and partly in response to our requests for such information.  But we still have a far way to go. 

At this point, however, and because of the lack of consistent tracking protocols, our data are necessarily messy and incomplete. They cover a longer historical period than we would like and were gathered by such different methods, with such different degrees of completeness across higher education institutions, that we are certain we are missing a number of nonprofit organization hosts.  On the other hand, we do have a quite large database—large enough that we can still get a useful picture of the geographic spread of service learning in Wisconsin. 

What is the Geographic Spread of Service Learning?

To understand the geographic spread of service learning, we need to define  rural and urban.  There are many definitions, perhaps the most common relying on standard metropolitan area designations.  Such designations do not work well for our purposes, however.  First, such designations include incorporated and unincorporated areas that have a strong agricultural base.  Only the city of Milwaukee has anything approaching multiple concentric rings of urbanization, and only Milwaukee, Madison, and Green Bay surpass populations of 100,000. Wisconsin’s other urban centers can, at best, be considered only small cities. There are consequently many communities within a technically urban county that are not functionally urban.

As we will also see, the data itself informed our definition of urban vs. rural. As we will show, the reach of service learning is so tightly concentrated around particular higher education institutions that, in nearly all cases except for Milwaukee, urban means inside the city limits.  Consequently, we limited the definition of urban to cities with 30,000 residents or more, and considered all cities in Milwaukee County to be urban/suburban. We counted as suburban any incorporated area sharing a boundary with those cities (see Table 2).  All other areas were counted as rural. 

Table 2:  Cities of Wisconsin with population above 30,000 (excluding Milwaukee County suburbs)

City

2000 Census Population

Milwaukee           

596974

Madison           

208054

Green Bay           

102313

Kenosha            

90352

Racine           

81855

Appleton

70087

Waukesha           

64825

Oshkosh           

62916

Eau Claire           

61704

West Allis           

61254

Janesville/Beloit

59498/35775

La Crosse           

51818

Sheboygan            

50792

Wauwatosa           

47271

Fond du Lac           

42203

Brookfield           

38649

Wausau           

38426

New Berlin           

38220

Manitowoc

34053

Menominee Falls

32647

Superior*           

27368

*Superior is included because it is part of the broader metropolitan cluster with Duluth.

 

We ended up with a database of 925 nonprofit organizations that hosted service learners in any of the academic years of 2005-6, 2006-7, or 2007-8.  Remember that each nonprofit organization is counted only once regardless of how many times it may have appeared in the data. 

We established the geographic spread of service learning through the use of zip code boundaries. Zip codes are quite useful in urban areas, as they correspond with city boundaries in most cases.  In rural areas, however, zip codes can cover quite a distance.  We will see, however, that does not create a problem in our data.

Our initial data collection unit is the nonprofit organization.  We do not know how many students from how many classes each organization hosted.  We then use that data to abstract to the zip code level to count the number of nonprofit organizations hosting service learning in each zip code.

The first thing we notice from the data is the extremely tight concentration of service learning around higher education institutions themselves. Of the 882 zip codes in Wisconsin, only 156 had nonprofit service learning hosts. Figure 1 shows the concentration of service learning for each zip code. 

Be aware that the situation is even more extreme than the map shows, as we shaded a zip code if it had only a single nonprofit organization hosting service learners.  Of the 156 total zip codes with nonprofit service learning hosts, there were 54 zip codes with only one organization host.  Of the total, only 124 organizations were rural, while 43 were suburban, and a whopping 758 were urban.

As you might suspect, and as we suspected, the relative lack of nonprofit organizations in rural areas is likely to impact the absolute numbers of organizations hosting service learners.  So we need to look at how many organizations there actually are in each area and then develop a ratio of service learning host organizations to the total.

We developed two databases of the total number of nonprofit organizations.  The first list comes from the State of Wisconsin Department of Revenue database, which, after eliminating large nonprofits, produces a database of 32,079 organizations. This database has been the standard for studies of nonprofit organizations in the state.  However, it excludes civic organizations, chambers of commerce, business and union organizations, fraternal and recreational groups, professional societies, social clubs, and  veteran organizations and their auxiliaries. To develop as inclusive a database as possible, then, we combined this database with both the IRS registration database and the annual tax return database.  The registration database includes all nonprofits, while the tax return database normally only includes organizations with more than $25,000 in annual income.  Eliminating duplicates and large organizations such as hospitals and private schools left us with 64,987 organizations--double the number of organizations in the Wisconsin database. 

We find the ratio data to be quite surprising.  While we expected some disparity between urban and rural organizations in their access to service learning, we also expected some of that disparity to be moderated by the lack of organizations in those areas.  But the data, shown in Table 3, in fact show that rural organizations are far less likely to have access to service learning.  Using the Wisconsin exempt organizations database, we find that 5.3 percent of urban nonprofits hosted service learners, while only 0.6 percent of rural nonprofits did so.  Urban nonprofits are nearly nine times more likely to host service learners.  The combined IRS-Wisconsin database, with twice the number of organizations shows the same situation.  With this data, which appears to include a larger proportion of rural nonprofit organizations, 2.5 percent of the total urban organizations hosted service learners compared to only 0.4 percent of rural organizations, more than a six-fold difference.

Table 3:  Percent of nonprofits hosting service learners by region

Wisconsin Exempt Organizations: urban/surburban

Wisconsin Exempt Organizations: rural

Total number of nonprofits

15494

16524

Number of nonprofit service learning hosts

801

124

Percent of total nonprofits hosting service learners

5.3%

0.6%

Combined Wisconsin and IRS database:  urban/surburban

Combined Wisconsin and IRS database:  rural

Total number of nonprofits

32179

32808

Number of nonprofit service learning hosts

801

124

Percent of total nonprofits hosting service learners

2.5%

0.4%

 

It is important to keep in mind that, while the findings provide an unsettling snapshot of the geographic distribution of service learning across the state, there are many challenges inherent in the data itself.  Because of the difficulties in collecting the community partner lists and finding zip codes, it is hard to determine the true quantity and quality of the data.  Because of the variation in sources, and timeframes, both the accuracy and commonality of the data are in some ways uncertain.  Additionally, there were some nonprofits for which it was tough to ascertain their location (i.e. zip code), especially with less formal organizations or clubs.  There is also some uncertainty about the actual population served by service learning in reference to the nonprofit location (For example, it may be the case that rural children have access to an urban based Big Brothers, Big Sisters Program).  In short, the lack of institutionalization of service learning on higher education campuses results in inconsistent and incomplete data.  And, most significantly, this data allow us to develop an  accurate assessment of the impact of service learning on communities and their nonprofit organizations.      

Definitional, issues are also important to keep in mind here.  Our definition of rural includes a number of population centers greater than 10,000 though smaller than 30,000.  Some may not classify such places as rural, and such communities are also the most likely to include the largest concentrations of nonprofit organizations.  We classified them in rural because of their relative isolation from the larger cities and their surrounding rural context.  Superior is included as an urban area because it borders the Minnesota city of Duluth.  The classification of Stevens Point as rural, with a population of 24,551,  is arguable, but it is situated in a rural context and falls below the 30,000 threshold.  Designating Stevens Point as urban does not change the final ratios by more than a tenth of a percent. It is also important to remember that our definition of service learning is limited to partnership with small to medium size nonprofit organizations.  Much service learning in rural areas does not follow the typical model of partnership with a community organization.  Instead service learning may occur with local governments, informal resident groupings, and even in one-on-one relationships.  Those occurrences are not counted here.

Even with more stable data and broader definitions, however, we doubt that we would see that much difference in the ultimate conclusions.  Rural nonprofits have dramatically less access to service learning.

Implications and Conclusions

To the extent that service learning can add to the problem-solving capacity of excluded and oppressed communities, rural communities are once again left at the back of the line in accessing needed services.  But why might this be?

The single obvious answer is distance (Holton, 2007).  Given the tight clustering of service learning close to campus, it is not just rural areas that are excluded, but entire areas of cities that extend further from campus.  This is the case even when the institutions such as the University of Wisconsin-Madison offers students free transportation.  But there may be more to the equation than the simple fact that rural service learning sites are further from campus.  We do not know whether this bias toward propinquity in service learning is the result of students feeling uncomfortable with venturing far from campus, finding more appealing service sites nearby, or not wanting to spend time traveling to and from a service learning site.  All of these explanations are likely partly true.  In urban areas, students from outside the city may be fearful of venturing into poor neighborhoods further from campus. There is some suggestion that even the difference of a few blocks can affect the popularity of a service learning site (Stoecker, Stern, and Hathaway, 2008).  The most popular service learning sites also have enough of a track record with students that they have systems in place for managing themsand such systems may make the service learning experience more convenient for the average student. 

Institutional efforts to expand service learning beyond zip codes adjacent to campus also depend on institutional faculty and staff going the extra mile.  There are faculty who develop service learning relationships with organizations not just in other zip codes but at the other end of the state and even in other states.  To the extent that good service learning is based on strong relationships, the further the distance the greater the effort needed.  But there is little institutional infrastructure to support such efforts to reach beyond campus.  As usual, those who have are also more likely to receive.  Organizations close to campus are more likely to know how to access students and build relationships, while those further away are going to find it more difficult. 

What are some ways of attacking this disparity?  The University of Wisconsin-Eau Claire may have one of the strategies, which shows a greater spread of service learning to more distant zip codes.  UW Eau Claire provides programs to support summer service learning, and nearly all of their geographic spread is likely the result of students conducting service learning in their home towns, sometimes in other states, over the summer.    A number of Appalachian community organizations such as Big Creek People in Action in West Virginia have developed systems for housing students in the community for short-term service learning assignments. The U-Links center, in rural central Ontario, is able to recruit students for projects from Trent University, 90 minutes away.  Part of their strategy is to design projects that only require students to make occasional trips to the community, and sometimes the community partner actually comes to campus. Support research projects that requires students to dig up information in libraries, or analyze survey data, allow the student to serve the community from afar.

Another source of possible support for expanding the geographic scope of service learning is the Carnegie Classification of Institutions of Higher Education, which added an elective classification labeled as Community Engagement in 2006.  This is defined as “the collaboration between institutions of higher education and their larger communities (local, regional/state, national, global) for the mutually beneficial exchange of knowledge and resources in a context of partnership and reciprocity,” and includes the practice of service learning. (The Carnegie Foundation for Advancement of Teaching, 2007).  The change has provided incentive for higher education campuses to expand their service learning programs.

At first glance, such an incentive holds great potential for improving the current state of service learning across the country.  The application includes questions regarding the existence of civic engagement offices, system-wide documentation and tracking mechanisms, as well as mechanisms for exploring community perception and community impact. However, the classification analysis is dependent upon a voluntary data collection process, rather than secondary analysis of existing data sources (in contrast to its other classifications).  It does not provide campuses with a standard systematic framework with which to institutionalize such elements of civic engagement.  This situation could lead to the promotion of more service learning, but not necessarily better service learning, which could cause more distress for already resource-limited community organizations.   Additionally, there is no mention of geographic diversity or promotion of civic engagement outside of the campus’ direct community.  Its quite clear from this research that many higher education campuses aren’t collecting the data that could help them answer the application questions, much less meeting the goals themselves.  Also, some Americorp*VISTA placements are dedicated to the task of applying for the classification rather than the more fundamental tasks of institutionalizing service learning.  The Carnegie Classification on Civic Engagement, therefore, should not make us more optimistic in regards to increasing rural communities’ access to service learning.

It is important to understand, then, that just because rural communities are not getting access to service learners does not mean they are necessarily losing out on anything. There are growing questions over the extent to which service learning actually serves communities or simply enhances the education of students (Stoecker and Tryon, forthcoming).  It is possible that simply trying to shift existing service learning practice to rural areas could actually further burden, rather than benefit, those communities.  If we are going to seriously consider equalizing the flow of service learning students into rural areas, we are going to need to consider how to do so in such a way that brings real benefits.  That means carefully structuring service learning projects through the participation of those community members who will be most impacted by it, training the students so that they have the necessary skills and attitudes prior to engaging in service, and providing the support that community organizations need to get the most out of service learning.

References

Birdsall, J. T. (2005). Community Voice: Community partners reflect on service learning. Retrieved on 11-26-2006  from: http://www.mc.maricopa.edu/other/engagement/Journal/Issue5/Birdsall.pdf .

Boethel M (1999). Service Learning: A Strategy for Rural School Improvement and Community Revitalization: The Exponential Results of Linking School Improvement and Community Development, Issue Number 2. Southwest Educational Development Lab, Austin, TX. http://www.sedl.org/prep/benefits2/issue2/

Campus Compact. (2007). 2006 Service statistics: Highlights and trends of Campus Compact's annual membership survey. Providence, RI: Campus Compact. http://www.compact.org/about/statistics/2006/

 

Carnegie Foundation for the Advancement of Teaching. (2007).  2008 Community Engagement Classification. http://www.carnegiefoundation.org/classifications/sub.asp?key=1213&subkey=2215 

Corporation for National and Community Service.  (2007).  Strategic Plan. http://www.nationalservice.gov/about/focus_areas/index.asp

Cruz, N. & Giles, D. (2000). Where’s the Community in Service-Learning Research?.  Michigan Journal of Community Service Learning, Special Issue, 28-34.

Ferrari, J. R.,  & Worrall, L. (2000). Assessments by Community Agencies: How “the Other Side” Sees Service-Learning. Michigan Journal of Community Service Learning Vol 7 2000 pp???.

Holton, N.  (2007).  Rural Service Learning: Turning Special Challenges into Great Opportunities .  Presented at the Community College National Center for Community Engagement meetings. http://www.mc.maricopa.edu/other/engagement/2007Conf/Papers/Holton.pdf

Jones, S. (2003). Principles and profiles of exemplary partnerships with community agencies. In B. Jacoby & Associates (Eds.), Building Partnerships for Service Learning (pp.151-173). San Francisco, CA: Jossey-Bass.

Mihalynuk,  Tanis V., Seifer, S. D.,  and Community Campus Partnerships for Health.  (2007).  Higher Education Service-Learning in Rural Communities.   Learn and Serve America’s National Service Learning Clearninghouse.  http://www.servicelearning.org/instant_info/fact_sheets/he_facts/rural_communities/

Muzak, J. and Woollard, L. (2008). Finding the Fit: Community Partners as Co-educators in Community Service-Learning.  In Community-University Partnerships: Connecting for Change.  Proceedings of the Third International Community-University Exposition (CUexpo 2008),  D. E. Clover and C. McGregor (eds.)

Sandy, M. & Holland, B. (2006). Different worlds and common ground: community partner perspectives on campus-community partnerships. Michigan Journal of Community Service Learning 13(1), 30-43.

Smith. F. (2008). What Do AmeriCorps* VISTA members DO? Wisconsin Campus Compact.  http://www.uwp.edu/departments/community.partnerships/wicampuscompact/vista/WorkPlan.cfm

Stoecker, R.  Stern, E. and Hathaway, P.  (2007). Managing Service-learners through a Student Intermediary and a Project-Based Service-learning Model.  Unpublished manuscript.

Stoecker R. and Tryon, E.  (forthcoming). Unheard Voices.  Community Organizations and Service Learning.  Philadelphia:  Temple University Press. 

Vernon, A., & Ward, K. (1999). Campus and community partnerships: assessing impacts & strengthening connections.  Michigan Journal of Community Service Learning 6, 30-37.

Worrall, L. (2007). Asking the Community: A Case Study of Community Partner Perspectives. Michigan Journal of Community Service Learning 14,5-17.

 

Note:

[1] Thanks to Matt Kures and Andy Lewis, from the University of Wisconsin Cooperative Extension Center for Community and Economic Development, for assistance with the GIS an analysis and the Wisconsin exempt organization data.  Thanks also to the National Center for Charitable Statistics for the Internal Revenue Service Core data files for Wisconsin.  This project was supported by federal Department of Agriculture Hatch funding grant WIS01041.  Please direct all correspondence to rstoecker@wisc.edu