Variations in social capital among vocational rehabilitation applicants

Abstract

BACKGROUND: In general, people with disabilities have lower levels of social capital, a measure of the quality of social relations, than people without disabilities. People with disabilities who participate in the labor force, however, have been found to have higher levels of social capital than their peers who do not participate in the labor force.

OBJECTIVE: Using newly available data from the Survey of Disability and Employment (SDE), this study examined perceived social capital as it relates to supporting employment among applicants for state vocational rehabilitation (VR) services in three states: Mississippi, New Jersey, and Ohio.

METHODS: We used multivariate analysis to compare differences in levels of perceived (i.e. cognitive) social capital between applicants who were employed and applicants who were not employed, by disability severity, age at disability onset, health status, and individual characteristics.

RESULTS: VR applicants were more likely to benefit from social capital in their working lives if they reported currently working, less severe disability, and better perceived health.

CONCLUSIONS: VR counselors must recognize that persons applying for VR services vary greatly in their access to the social supports that are closely associated with employment.

Keywords

Vocational rehabilitation social capital employment

1 Introduction

State vocational rehabilitation (VR) agencies provide employment services to youth and adults who have disabilities. These services include assistance with job search or retention and financial support for obtaining assistive technology, education, and training. All of the services are provided with a goal of ensuring that applicants obtain or maintain competitive employment. In 2012, more than 560,000 persons applied for VR services at state agencies in the United States (Houtenville, Brucker, & Lauer, 2014). The characteristics of these applicants vary along a number of dimensions, as agencies serve a diverse group of people who have different types of disabilities and different levels of education and employment experience. Nationwide, slightly more than half (56%) of VR clients are employed when their cases are closed (Houtenville et al., 2014).

The research presented here explores the relationship between perceived social capital and employment among applicants to VR agencies. Social capital can be characterized as the quality of social relations. Prior work has suggested that deep reserves of social capital are required to maintain stable employment, both among the general population (Casey & Christ, 2005) and among vulnerable populations (Johnson, Honnold, & Threlfall, 2011). Levels of social capital may covary with employment for the VR population as well, as research on the general population of persons with disabilities suggests that higher levels of social capital are associated with positive employment outcomes (Araten-Bergman & Stein, 2014; Brucker, 2015). While existing research on VR has examined variations in employment outcomes associated with systemic differences among state agencies (Stapleton, Honeycutt, & Schechter, 2010) and among demographic factors, disability type, educational attainment, and occupation (Cimera, Avellone, & Feldman-Sparber, 2015; Cimera, Rumrill, Chan, Kaya, & Bezyak, J., 2015; O’Neill et al., 2015; Zhang et al., 2015), our study is the first to explore social capital as it relates to support of employment in the VR population.

We found that VR applicants who were employed and who had better health had higher levels of social capital. Applicants with severe limitations, conversely, had lower levels of social capital.

2 Literature review

Social capital can be conceptualized as a measure of the quality of social relations. High levels of social capital are characterized by high levels of trust and reciprocity, which lead to mutually beneficial outcomes (Araten-Bergman & Stein, 2014). Social capital can be measured at the individual or community level. Ehsan and DeSilva (2015) describe four measures of social capital, encompassing both structural (participatory) and cognitive (perceived) domains. Individual-level cognitive social capital captures levels of general trust in others, perceived social support, and sense of community. Ecological-level cognitive social capital measures trust in the environment and in politicians. Individual-level structural social capital reflects levels of actual participation in local or voluntary organizations, civic actions, actual support received from neighbors, and frequency of contact with family, friends, and neighbors (Ehsan & DeSilva, 2015). Ecological-level structural social capital draws from community-level measures of per capita group membership and engagement in public affairs (Ehsan & DeSilva, 2015). At the individual level then, social capital reflects the level of cohesion of the group to which an individual belongs. At a broader level, social capital can be viewed as a collective resource of a workplace (Macinko & Starfield, 2001) or community (Lochner, Kawachi, & Kennedy, 1999). The research we conducted examined individual-level cognitive social capital as it relates to support of employment.

People with disabilities generally have lower levels of social capital than other people, and social capital varies by disability type (Condeluci, Ledbetter, Ortman, Fromknecht, & DeFries, 2008; Meltzer et al., 2013; Stancliffe, Lakin, Taub, Chiri, & Byun, 2009). As Condeluci et al. note, certain types of disabilities affect the ability to form social connections. Persons with autism or schizophrenia, for example, may have lower levels of social capital than persons with other types of disabilities. Age at onset of disability may influence levels of social capital as well. Persons with intellectual or developmental disabilities (disabilities with onset in childhood) have uniquely low levels of social capital that are highly dependent on family relationships throughout the lifespan (Kramer, Hall, & Heller, 2013). A recent study investigating social capital disparities between people with and people without disabilities empirically confirmed these observations. Mithen, Aitken, Ziersch, and Kavanagh (2015) found that persons with intellectual and psychological impairments have greater levels of social capital disparity than those with sensory/speech and physical disabilities.

The association between disability severity and age at onset of disability has not been specifically examined for VR applicants, however. Understanding how levels of social capital vary by disability characteristics can help VR counselors develop plans for assisting applicants in obtaining necessary social supports. We thus first tested the hypothesis that social capital varies by disability characteristics including severity, age at onset, and health. We controlled for the possible effects of employment status and other individual characteristics on social capital.

Variations in labor force participation are also associated with differences in levels of social capital. Gilbride and Stensrud (2008) suggest that people with disabilities may have lower levels of social capital related to the labor market compared with people without disabilities because the former spend a disproportionate amount of time and energy establishing the necessary capital to navigate within the medical care, public assistance, and public education arenas. Prior research has confirmed an association between labor force participation and social capital for persons with disabilities, yet a causal relationship has not been established. Some research (Araten-Bergman & Stein, 2014) has suggested that persons with disabilities who are employed have higher levels of social capital than persons with disabilities who are not employed. Other research (Brucker, 2015) has found that persons with disabilities who are in the labor force (either working or seeking a job) have higher levels of social capital than persons with disabilities who are not in the labor force. Among persons with disabilities who are in the labor force, social capital has not been found to vary by employment status (Brucker, 2015). These findings suggest that further research is needed to understand how social capital varies by employment status for VR applicants. All VR applicants can loosely be considered as engaged in the labor force, as they are either employed or have taken active steps to obtain employment by connecting with VR. We therefore tested a second hypothesis, namely that levels of social capital vary by employment status for VR applicants, controlling for disability characteristics and individual characteristics.

3 Methods

3.1 Data

The Survey of Disability and Employment (SDE) was conducted by Mathematica Policy Research over a 6-month period in 2014 and 2015. The SDE gathered data from applicants to the New Jersey Division of Vocational Rehabilitation Services, the Mississippi Department of Rehabilitation Services, and Opportunities for Ohioans with Disabilities. The Kessler Foundation institutional review board reviewed and approved the survey research protocol. Eckstein, Sevak, and Wright in this issue provide more detail about the survey design and administration. A total of 2,804 surveys were completed. A small portion of the sample was missing information on the social capital measures of interest and was excluded (5.9%). The final analytic sample consisted of survey data from 2,639 cases.

3.2 Measures

3.2.1 Social capital

We focused on perceived social capital variables that were related to employment. Four questions gathered specific information about social support available in the community, reflecting cognitive social capital, as defined earlier. The questions asked respondents if there was anyone they could rely on for the following: help finding a job; borrowing money to pay an urgent bill such as electricity, gas, rent, or mortgage; transportation to get to work urgently; and help with a serious personal crisis that made it difficult for them to find or keep a job.

3.2.2 Disability

We examined several disability-related characteristics. First, we created a disability severity variable based on responses to two survey questions. We coded respondents as having a severe disability if they stated either that they had difficulty dressing or bathing or that, because of their condition, they had difficulty doing errands alone such as keeping a doctor’s appointment or shopping. Next, we created a disability age-at-onset variable. To calculate age categories at disability onset, we used a survey question that asked the approximate time when respondents became aware that they had a limiting condition. We categorized age at disability onset into groups coinciding with different periods of the working life: younger than 17 years (employment preparation), 17 to 24 years (transition into employment), 25 to 44 years (early career), and 45 to 65 years (later career). Lastly, we recoded the survey question on perceived health to develop a measure whereby higher values indicated higher levels of health, ranging from 1 (poor) to 5 (excellent).

3.2.3 Employment

The survey asked respondents whether they worked at one or more jobs for pay or income during the past week. We coded employment as a binary variable, assigning a value of 1 to applicants who reported that they had worked in the past week and a value of 0 to applicants who had not worked in the past week.

3.2.4 Demographic variables

Five individual background characteristics included gender (1 = male, 0 = female) and categorical measures of age group, race, educational attainment, and marital status. To define race, respondents were asked first to identify if they were “Spanish, Hispanic, or Latino” and then to identify their race, with the option to select all categories that applied. Response categories included White, Black or African American, American Indian or Native Alaskan, Native Hawaiian/Pacific Islander, Asian, or “other.” We used the information reported for both questions to create a single categorical race variable, assigning priority to reported Hispanic background if a person also reported a race. Owing to low frequencies, we assigned responses of Asian, American Indian, Native Hawaiian/Pacific Islander, and “other” to a single category. The final, mutually exclusive categories included non-Hispanic White, non-Hispanic Black, Hispanic, and “other.” We collapsed education level into three categories based on the highest educational degree the subject reported: no degree (less than high school); high school diploma or GED (for the General Educational Development test); or some college education or more. Finally, we collapsed marital status into three categories: married (included cohabiting); never married; and widowed, divorced, or separated.

3.3 Analytical plan

All analyses were conducted using Stata/SE version 14.0 (StataCorp LP). Data were weighted to address the complex sampling process used for the survey (unweighted N = 2,639, weighted N = 6,772). First we obtained descriptive statistics. Next, we created a bivariate table to show the relationships among our four social capital variables and our focal disability characteristics and employment variables. We used χ² and t tests to test for significant differences. Lastly, we ran logistic regressions, using each of the four social capital variables as separate dependent variables. These regressions allowed us to examine the relationships among disability characteristics, employment, and the four different types of social capital measured here, providing us with evidence to accept or reject our research hypotheses.

The limitations of this analysis include bias owing to omitted variables and limitations in generalizability as the data were collected from VR applicants in only three states. In addition, we conducted the analysis on a cross-sectional dataset, which does not allow for the establishment of causality between social capital and employment.

4 Results

Table 1 provides descriptive statistics on the four social capital measures, employment, and other measures used in the model. Approximately two-thirds of the sample reported having someone who could help them with finding a job, transportation, or a personal crisis. Less than 57% had access to someone who could help with financial concerns. One-quarter of the sample consisted of adults between the ages of 35 and 44 years and a third of the sample was adults between the ages of 45 and 54 years. Adults ages 25 to 34 constituted 22.6% of the sample and adults ages 55 to 65 accounted for 18.7% of the sample. The majority of the sample had at least a high school education; very few people (15.9%) reported that they had never obtained a high school diploma. The sample was evenly split by gender. Twenty-eight percent of the respondents were married and 45.7% were never married. The sample was racially diverse, with 50.5% identifying as non-Hispanic White, 37.5% as non-Hispanic Black, 6.4% as Hispanic, and 5.6% as non-Hispanic “other.” Slightly more than a third (36.2%) of the sample had been employed in the previous week. A third of the sample reported that the onset of their limitations had occurred between the ages of 25 and 44), with childhood and adolescence being the second largest age-at-onset group. The majority of the sample did not report severe limitations. Most applicants reported that their health was either good (33.6%) or fair (30.0%).

Table 2 shows the bivariate associations among the four social capital measures and each of the four focal variables (employment, age at disability onset, severe limitations, and perceived health). Each social capital measure varied significantly by employment status, disability severity, and perceived health. Age at onset was significantly associated with having someone who could provide assistance in finding a job, borrowing money, and helping with a personal crisis.

Table 3 shows the results of the multivariate logistic regressions. Our results confirm both of our hypotheses. Levels of social capital varied by disability characteristics and by employment status for VR applicants. Applicants with less severe disabilities and higher levels of self-reported health had greater social capital. Applicants who were employed experienced higher levels of social capital as well.

Applicants with severe limitations had significantly lower odds of having access to someone who could help them with job search, loan them money, provide transportation to work, or assist them in a personal crisis. Applicants with disability onset between the ages of 25 and 44 were significantly less likely than applicants with disability onset prior to age 17 to have access to personal financial supports.

Subjects with higher levels of self-reported health had significantly higher odds (odds ratio [OR]: 1.29, p < .01) of having someone to turn to for job search assistance. Those with higher levels of self-reported health had significantly higher odds of having access to transportation assistance. Applicants with higher levels of self-reported health had significantly higher odds of having someone available to loan them money.

Applicants who were currently working had significantly higher odds (OR: 1.66, p < .01) of having someone they could rely on to help them find a job. Applicants who were employed also had significantly higher odds (OR: 1.26, p < .05) of having someone available to loan them money. Applicants who were currently employed had significantly higher odds (OR: 1.73, p < .01) of having someone they could rely on for emergency transportation assistance to work. Persons who were employed had greater access to someone they could rely on for help with a personal crisis that might interfere with an ability to work (OR: 1.59, p < .01).

Compared with younger applicants, adults ages 45 or older were significantly less likely to have someone available to help them find a job. Applicants who were minorities or those who did not have postsecondary education had significantly lower odds compared with their reference groups of having someone from whom they could borrow money. Subjects who were widowed, divorced, or separated had less access to financial support compared with persons who were married or cohabitating.

Men had significantly higher odds than women of having access to transportation assistance. Applicants who were not White or who were age 35 or older had lower odds of having access to transportation help, compared with their reference groups.

5 Discussion

Variations in social capital existed among the applicants for VR services whom we studied. First, social capital varied significantly by employment status. Applicants who were currently employed already had a strong network of supports that were likely instrumental in helping them obtain and maintain employment. This result confirms other research that has suggested a strong tie between employment and social capital (Araten-Bergman & Stein, 2014). The new findings presented here, however, suggest that social capital varies between persons with disabilities who are employed and those who are not employed but have demonstrated an interest in finding a job by applying for VR services. Applicants to VR who were not employed had lower levels of social capital than those who were already employed.

Second, VR applicants who had severe limitations had lower levels of social capital. Persons with limitations in their self-care abilities or their ability to live independently or who may need high levels of support may be most compromised in this area. While the VR system can attempt to fill some of this gap in social capital for this population, the coordination of multiple support services within the community is likely necessary.

Third, in most cases, the age at onset of disability was not strongly related to levels of individual-level cognitive social capital. Persons who acquired disabilities at different ages had similar odds of having supportive persons available to assist them with looking for a job, with personal crises, or with transportation. Persons whose onset of disabilities occurred at age 25 or older, however, were less likely to have access to someone who could provide financial support. As the majority of VR applicants fell within this category, according to our analysis, it is important for VR counselors to be aware of this economic vulnerability. Ensuring that applicants are connected to financial supports that are available via the community may be important in addressing this concern. VR counselors have the capability to provide various types of financial support as a basic part of VR services and even after a VR case is closed, if a financial crisis threatens a person’s job stability. To supplement such support, many communities offer financial assistance on a crisis basis through nonprofit or religious organizations. Being cognizant of such opportunities can assist in boosting the social capital of persons with disabilities.

Fourth, applicants who reported better levels of health had significantly higher levels of social capital. Conversely, those with lower levels of health had lower levels of social support. While the research conducted here does not allow us to determine the causal relationship between these two variables, other research has found that higher levels of social capital lead to better health among the general population (Murayama, Fujiwara, & Kawachi, 2012). Future research can attempt to disentangle the relationship between social capital and health for persons with disabilities.

Lastly, the types of social capital measured here, including financial, job search, social support, and transportation assistance, provide a glimpse into the range of supportive services that persons with disabilities may need to obtain and maintain employment. VR counselors should strive to ensure that community-level supports are developed or enhanced. This may entail working beyond the usual employment support system to access nontraditional supports such as those available from other nonprofit or religious organizations. An example of accessing nontraditional community resources is a project recently funded by the Kessler Foundation that leverages social capital embedded in faith-based communities to support people with disabilities in becoming employed and staying employed. One element of the model involves enhancing the social capital of persons with disabilities by exploiting the built-in networks and influence of congregation members who are business owners, managers, human resource personnel, and employees (Nord, Timmons, Carter, & Gaventa, 2014).

6 Conclusion

Applicants to VR programs are more likely to benefit from social capital in their lives if they report currently working, are less severely disabled, and have better perceived health. Conversely, VR applicants who are not working are likely to have lower levels of social capital. In working with VR applicants, counselors should seek to enhance applicant connections with persons or organizations that can provide financial, job search, social support, and transportation assistance within the local community.

Conflict of interest

The authors have no conflict of interest to report.

Footnotes

Acknowledgments

This project was funded by the U.S. Department of Health and Human Services, National Institute on Disability, Independent Living, and Rehabilitation Research under cooperative agreement 90RT5017-01-01. The information developed by the Rehabilitation Research and Training Center on Individual-Level Characteristics Related to Employment among Individuals with Disabilities does not necessarily represent the policy of the Department of Health and Human Services, and endorsement by the federal government should not be assumed. The authors retain sole responsibility for any errors or omissions.

References

Araten-Bergman

, & Stein

M. A.

(2014). Employment, social capital and community participation among Israelis with disabilities. Work, 48, 381–390.

Brucker

D. L.

(2015). Social capital, employment and labor force participation among persons with disabilities. Journal of Vocational Rehabilitation, 43(1), 17–31.

Casey

, & Christ

(2005). Social capital and economic performance in the American states. Social Science Quarterly, 86(4), 826–845.

Cimera

R. E.

, Avellone

, & Feldman-Sparber

(2015). An investigation of the outcomes achieved by individuals with intellectual disabilities and mental illnesses. Journal of Vocational Rehabilitation, 43(2), 129–135.

Cimera

R. E.

, Rumrill

P. D.

, Chan

, Kaya

, & Bezyak

(2015). Vocational rehabilitation services and outcomes for transition-age youth with visual impairments and blindness. Journal of Vocational Rehabilitation, 43(2), 103–111.

Condeluci

, Ledbetter

M. G.

, Ortman

, Fromknecht

, & DeFries

(2008). Social capital: A view from the field. Journal of Vocational Rehabilitation, 29(3), 133–139.

Eckstein

, Sevak

, & Wright

(2017). Personal characteristics of vocational rehabilitation applicants: Findings from the Survey of Disability and Employment. Journal of Vocational Rehabilitation, 46(2), 177–185.

Ehsan

A. M.

, & De Silva

M. J.

(2015). Social capital and common mental disorder: A systematic review. Journal of Epidemiology and Community Health, 69(10), 1021–1028.

Gilbride

, & Stensrud

(2008). Why won’t they just do it?Rehabilitation Education, 22(2), 125–132.

10.

Houtenville

A. J.

, Brucker

D. L.

, & Lauer

E. A.

(2014). Annual compendium of disability statistics: 2014. Durham, NH: University of New Hampshire.

11.

Johnson

J. A.

, Honnold

J. A.

, & Threlfall

(2011). Impact of social capital on employment and marriage among low income single mothers. Journal of Sociology and Social Welfare, 38(4), 9–31.

12.

Kramer

, Hall

, & Heller

(2013). Reciprocity and social capital in sibling relationships of people with disabilities. Intellectual and Developmental Disabilities, 51(6), 482–495.

13.

Lochner

, Kawachi

, & Kennedy

B. P.

(1999). Social capital: A guide to its measurement. Place, 5(4), 259–270.

14.

Macinko

, & Starfield

(2001). The utility of social capital in research on health determinants. The Milbank Quarterly, 79(3), 387–427.

15.

Meltzer

, Bebbington

, Dennis

M. S.

, Jenkins

, McManus

, & Brugha

T. S.

(2013). Feelings of loneliness among adults with mental disorder. Social Psychiatry and Psychiatric Epidemiology, 48(1), 5–13.

16.

Mithen

, Aitken

, Ziersch

, & Kavanagh

(2015). Inequities in social capital and health between people with and without disabilities. Social Science & Medicine, 126, 26–35.

17.

Murayama

, Fujiwara

, & Kawachi

(2012). Social capital and health: A review of prospective multilevel studies. Journal of Epidemiology, 22(3), 179–187.

18.

Nord

, Timmons

, Carter

, & Gaventa

(2014). Building community capacity to advance employment: Putting faith to work. Journal of Vocational Rehabilitation, 40(2), 161–164.

19.

O’Neill

, Kang

H.-J.

, Sanchez

, Muller

, Aldrich

, Pfaller

, & Chan

(2015). Effect of college or university training on earnings of people with disabilities: A case control study. Journal of Vocational Rehabilitation, 43(2), 93–102.

20.

Stancliffe

R. J.

, Lakin

K. C.

, Taub

, Chiri

, & Byun

S. Y.

(2009). Satisfaction and sense of well being among Medicaid ICF/MR and HCBS recipients in six states. Intellectual and Developmental Disabilities, 47(2), 63–83.

21.

Stapleton

, Honeycutt

, & Schechter

(2010). Closures are the tip of the iceberg: Exploring the variation in state vocational rehabilitation program exits after service receipt. Journal of Vocational Rehabilitation, 32(1), 61–76.

22.

Zhang

, Kwan

, Trivdei

, St. Laurent

M.-L.

, Haines

, & Foley

(2015). Earnings and demographic variations among vocational rehabilitation occupational closures: Patterns and trends, 2008-2012. Paper presented at the 2015 meeting of the Population Association of America, San Diego, CA. http://paa2015.princeton.edu/uploads/151043