Linking welfare clients to jobs: discretionary use of worker social capital.

The overarching theme of the 1996 welfare reform law was to move clients from dependency to self-sufficiency by facilitating their entry into the labor market. While numerous mechanisms were used to do this, this study explores discretionary actions taken by workers to help clients find jobs, namely, tapping into their own social capital. Respondents in one urban and one rural county in a southern state reported using their own social capital to get information regarding job openings and to exert influence to get clients hired. Notably, respondents at all levels of the bureaucracy expected this behavior to occur. Both the positive and negative aspects of social capital emerged as points of discussion in the rural county. Potential benefits and risks of worker social capital use are discussed as are future research directions and implications.

**********

A plethora of research undertaken in recent years addresses the impact of the TANF program on clients. This research documents a variety of factors that affect the success or failure of clients in securing and maintaining employment, including the economic conditions of an area, skill levels of recipients, child care, transportation, and client attitudes (Brayfield and Hofferth, 1995; Ong, 1996; Hofferth, 1999; Danziger et al., 1999; Kalil, Schweingruber, and Seefeldt, 2001). The Work First strategy driving the Personal Responsibility and Work Opportunity and Reconciliation Act of 1996 (P.L. 104-193) focuses on pushing welfare recipients into the labor force as rapidly as possible (Midgley and Rainford, 2000). Much of the discourse surrounding welfare reform focuses on the client's responsibility to work while the specific responsibilities and tasks to be undertaken by the state in the welfare-to-work era eludes most inquiry (Brodkin, 1997). Thus, a closer look at how state agencies and individual workers implement their side of the welfare-to-work contract is warranted (Brodkin, 1997). Otherwise, as Brodkin (1997) points out, bureaucratic processes are reduced to the proverbial black box. One area not sufficiently discussed in the literature is the formal and informal mechanisms used by welfare workers to find employment for their clients. This study explores one informal job search mechanism: worker social capital. It seeks to determine whether or not workers use their own social capital to help clients find jobs, ascertain the factors that influence this use and examine the attitudes and opinions of workers regarding its use.

Theoretical Framework

Social Capital and the Job Search

Social capital exists in the social relations of individuals (Lin, 1999) and like Bourdieu's (1985) conception of the term, social capital, in this paper, refers to elements of social relationships that result in economic benefits to individuals. This includes the social-structural resources available to individuals that facilitate actions that further their interests (Coleman, 1990). Especially useful in the job search are the social resources present in the networks of others, known as alters, in an individual's network. Individuals whose alters have higher levels of wealth, status, and power have greater access to information and influence that can improve stratification outcomes (Lin, Ensel, & Vaughan, 1981). Research has demonstrated that social networks are essential in obtaining both professional and entry-level blue-collar jobs (Granovetter, 1981; Kaye and Nightingale, 2000). Newman (1999) affirmed these findings in reference to the low-skilled work force during her research in Harlem, "employers can be very choosy, and they use social networks, among other things, as a mechanism for streamlining the choice-making process" (p. 84).

Worker Social Capital And Discretion

Job readiness and search classes, on-the-job training, community work experience placements (CWEP) and other subsidized employment opportunities provide welfare recipients avenues to enter the work force. Some recipients also use their own personal networks to find jobs. This study explores whether or not caseworkers use their own social capital to help clients find jobs. Arguably, this would benefit clients as long as workers' networks include individuals with higher status and better education than those of their clients since such people would have access to better jobs.

Using personal networks to help clients is not in the job description of case managers. However, this does not rule out the practice since research indicates broad bureaucratic discretion in welfare-to-work programs, both at the state and caseworker level (Brodkin, 1997). We argue that the prevalence of discretionary action by street-level bureaucrats (Lispky, 1980) makes use of their own social capital to help clients find jobs a distinct possibility. Furthermore, workers have an incentive to do this since their effectiveness is measured by participation rates, or the percentage of recipients participating in a countable activity.

Research Design

Qualitative methods are employed because this such methods are appropriate when studying a topic about which little is known (Padgett, 1998, p. 7). While much research on welfare reform exists, it focuses on the outcomes of reform and the impact of the policy change on clients (Long, 2001; Freedman, 2000; Brady-Smith et al., 2001 and others). Other studies assess how agencies implement existing policy and manager/client perception of programs (Danziger and Seefeldt, 2000; Sandfort, 1999). The phenomenon of interest here, worker social capital use, is not included in any agency policy manual and it is not in the job description of employees in welfare agencies. Little is known about this topic because researchers simply have not asked questions about it.

Guided Conversations

Guided conversations with key informants were conducted to gather information. A guided conversation is a discussion between a respondent and a researcher that is guided by the researcher. The goal of this type of data collection is to increase understanding of an issue and to uncover other issues. This "discovery process often leads the researcher to explore issues and ideas not part of the original study" (Lofland and Lofland, 1995). The advantage of this is that researchers gets closer to reality than they would if they only asked questions defined by previous research. A guided conversation can be differentiated from an interview because it is less structured. Interview schedules used with this methodology are "lists of things to be sure to ask about when talking to the person," rather than a list of questions that must be asked (Lofland and Lofland, 1995: 85).

Data Collection Procedure

In this type of research, much of the success of the data collection relies on the researchers' ability to ask follow-up questions to informant leads. To maximize the quality of data collection both researchers were present for conversations whenever possible. One researcher led the conversation and the other monitored and asked follow-up questions. This type of observer triangulation enhances the rigor of qualitative research (Padgett, 1998). Due to time constraints, case managers were interviewed by only one of the researchers.

Each of the guided conversations was taped and the researchers transcribed the tapes. In all cases, a room with a door was provided for the conversation to enable respondents to speak candidly. Signed consent was obtained from all participants. While direct quotes are used in this paper, care was taken to insure the anonymity of respondents: no names are included and all references to individuals or places that could be used to identify individuals were removed.

Sample Selection. Both purposive and convenience sampling techniques were used to select respondents. Purposive sampling was used in order to get the perspective of individuals in different positions within the welfare bureaucracy. Thus, respondents included state-level administrators, managers in charge of welfare offices, supervisors of case managers, and case managers. All top-level administrators in the state office were included as were the managers of selected sites. A convenience sample was used for case managers and supervisors to minimize agency disruption. Managers selected these participants based on their availability during scheduled interview times.

Site Selection. Research sites included state offices, one rural county and one district office in an urban county. Site selection used both purposive and convenience sampling. Since differences in rural and urban social structures are widely documented in the literature, both types of counties were included in the study to elicit experiences in each type of locale. The largest urban county in the state was selected because of its size and the rural county because it is one of the poorest in the state. Both counties were poorer than the national average with higher levels of unemployment and declining populations (U.S. Bureau of the Census, 1997; U.S. Department of Labor, 2000; U.S. Census Bureau, 2000). Since the urban county had more than one welfare office, one district office was chosen as the study site. To inform the selection, state administrators were asked to identify a typical welfare office. Typical was defined as average, not the best and not the worst. Another criterion was that the office not be participating in other aspects of the welfare evaluation project in order to decrease agency disruption.

Data Analysis and Coding

Each interview tape was transcribed and both researchers coded the data independently from the written transcripts. The first step involved marking all references to client job search or worker use of social capital. Second, both coders developed initial codes describing statements extracted from the transcripts. Care was taken to create numerous and varied codes (Lofland and Lofland, 1995, p. 192). The next step, focused coding, involved analyzing the codes and determining which were useful and which could be eliminated. The two coders compared notes and modified codes when they overlapped or needed to be subdivided.

This paper focuses on responses related to the personal network theme identified through the coding. The subcategories of this theme included (1) general examples of self using personal networks; (2) specific examples of self using personal networks; (3) general statements that others use personal networks; (4) specific incidences of others using personal networks; (5) statements that personal networks are seldom used; and (6) statements that personal networks are never used.

Findings

Findings demonstrate that some workers use their social networks to gather information about available jobs and share this information with clients. Some workers also exert influence through their networks to help clients obtain employment. Finally, negative and positive dimensions of social capital as related to the job search of clients emerge in conversations with rural respondents.

Worker Social Networks and Job Information

Social network contacts are seen as important to the job candidate because they may possess information about available jobs not already known by the job candidate (Granovetter, 1973; Lin, Ensel, & Vaughan, 1981). Some workers noted that they shared information obtained through their own social networks about jobs with their clients. One supervisor in the urban county, for example, noted that she obtained information about two job opportunities from classmates in graduate school. A case manager in the urban area discussed a similar experience that occurred recently. She shared job availability information with a coworker that she obtained by chance from an acquaintance:

   "An incident just happened this morning where another worker
   asked me did I know anyone that was hiring. This lady that I
   know ... works for [a hotel in the city] ... She asked me what I
   did and when I explained to her what I did she gave me her card
   and said that if ! had anybody to send down there to her
   [to do so]."

Both of these comments support Lin and his colleagues' (1981) argument that high status contacts contribute to positive job search outcomes. In these cases, high status contacts with individuals possessing graduate degrees provided information about jobs to workers and they passed along this information to recipients.

Another urban case manager expressed general knowledge that workers obtained information about jobs available from their personal networks and relayed this information to clients. Additionally, a rural case manager noted that she relayed information received through her social network regarding job openings to her clients. However, since the area has a high unemployment rate, this did not occur often.

Numerous respondents stressed that worker social capital was not the most common mechanism used by clients to find jobs. One case manager noted, for instance, that she did not think recipients commonly found jobs through such informal means, even though she did have a recent example. Similarly, another case manager in the urban county stated that she had on occasion heard about jobs through her personal network and informed clients. However, this did not happen regularly. The administrator of the rural parish, in fact, noted that most clients found jobs through word of mouth. While some of this happened in the family support office, it mostly occurred in their own communities.

Worker Social Networks and Influence Exertion

Social network contacts are also important because they enable some individuals to exert influence on the hiring process (Granovetter, 1973; Lin, Ensel, & Vaughan, 1981). One urban case manager noted that she had several alters in her network, including her sister and friends, who obtained information about job openings. She did, however, note discretion in using these ties to actually exert influence.

   "I wouldn't call on a personal friend to get a job for somebody
   unless I have a feeling about that client, unless I know that
   client is going to work out because I don't want to have to hear
   about this from a personal friend. Yes, for clients that I trust
   and that want to work-then I will call on someone I know and say,
   `Look I have someone finishing in bookkeeping, don't y'all have
   something? Anything? If we could just get them in, they can prove
   themselves.' And they say, `Okay, there you go again.' For select
   clients I do pull some stuff. Other than [that], I go through job
   fairs and such ..."

The case manager reported that she had sent someone to a particular business and the employer was not satisfied. The business owner is now skeptical of her clients and refuses to give them a chance when she calls.

A tendency to reserve one's social capital to help the most needy clients surfaced in the rural parish. An administrator noted,

   "... If I hear a worker is having trouble placing a client, I know
   this guy that manages this place. I'll call him and see if he will
   try this. Sure, I was never above calling in for favors ..."

Similarly, a rural worker noted that when she had a large number of clients to be placed, she called on a relative of one of the other individuals in the office to develop a community work experience placement.

Such behavior appears to be condoned by administrators as well as case managers. When asked whether case managers use their own connections to find jobs for clients, a rural parish administrator answered affirmatively:

   "Sure they do. Their neighbor may own something down the street,
   they will say, `Let me call them and see what I can do.' They use
   personal contacts. It's kind of an overlapping thing. I've done it
   before, all of us probably have."

Two state-level administrators support this view also. In fact, even though not part of any official job description, they expected this behavior to occur. One stated,

   "(County) administrators (especially in smaller communities) are
   well known and can exert influence on employers. They approach
   employers and ask them to hire or provide [CWEP] placements ..."

Another state administrator noted that even at the higher levels of the bureaucracy, individuals tap into their social networks to exert influence to have recipients hired. She gave a hypothetical example,

   "If I happen to know that I have a friend who ... is in the daycare
   business ... I say, `You need to come to this meeting ... I think
   you need to do something to help us.'"

Even though the use of worker social capital for exerting influence is not an explicit duty of workers, it does fit within the realm of acceptable actions of workers and is sanctioned, at least informally, by administrators. As demonstrated in previous research, case workers in public welfare agencies use discretion in executing their job duties (Brodkin, 1997). Specifically, there appear to be three instances when workers use their social capital to exert influence: (1) when the client is especially competent, (2) when the client is especially vulnerable, and (3) when there are a large number of clients needing jobs.

Negative and Positive Ramification of Social Capital

Alejando Portes (1998) in a critical review of the idealistic rhetoric surrounding social capital, emphasizes that social capital has a negative side. He argues that current research and discourse regarding social capital focuses only on the positive attributes, failing to acknowledge the characteristics of social capital that have negative effects on individuals. For example, the closed social systems praised by social capital proponents also restrict access to opportunity and individual freedom.

Research on the social structure of rural communities reveals the potential negative effects of social capital on the job prospects of the poor. Studies of rural communities demonstrate large social divisions between the poor and the nonpoor and job opportunities linked to personal contacts, influenced by individual and family reputations and controlled by elites (Duncan, 1992; Duncan & Lamborghini, 1994; Tickamyer and Duncan, 1990; Vindich & Bensman, 1958).

Conversations with administrators and workers in the rural county reflect both the positive and negative effects of social capital. The administrator of the rural county noted that it was difficult to place clients in private corporations. She thought this was because of the community's small size:

   "It's worse because they know everybody. They know the client.
   They know their momma ... momma's history. It's harder in a small
   place."

The regional specialist agreed with this stating:

   "... if they come from a family without a very good reputation ...
   we've had employers say on more than one occasion, `I'm not hiring
   them, that's old so and so's child.'"

A similar incident was mentioned by a rural supervisor in the context of trying to place clients in community work experience placements.

   "Yes, a lot of times [businesses ask who the client is before
   agreeing to the placement]. And a lot of times if they are known
   for drugs or something like that, they say they wouldn't prefer
   to have that person. So they pick and choose."

When asked whether the closed social setting could also have a positive effect on increasing the ability of workers to exert influence to get clients hired, the answer was swift,

   "No, not here. It's the old stigma. They are just on welfare.
   That's a lot of what you have to overcome".

This respondent elaborated, revealing the positive ramifications of social capital. She noted that in order to overcome this obstacle, case workers sweet talk employers:

   "We'll tell them, We've got this girl. She's got two babies. Give
   her a shot. We'll get her here and get her day care. Just take her
   and see if you can train her. If it doesn't work out, I'll come get
   her. We'll do something else."

Later, the rural administrator continued discussing this topic,

   "... Now that's the advantage to being in a small [county]: where
   the businesses know all the clients, the case managers know all the
   businesses. It works both ways. They know what type of person each
   business will take. [Itl all boils down to the case managers sweet
   talking them ... Picking up the client ... Helping out."

In rural areas, the closed social system seems to restrict opportunities for some while making it possible for workers to overcome these restrictions by exerting pressure on employers.

Study Limitations

Before discussing the implications of this study, it is important to note that it has several limitations. The nature of qualitative research with a small and non random sample of respondents means that this work cannot be considered representative of workers in public welfare agencies nationally or even in the South. Additionally, due to the methodology, the study elicits information and raises new questions; it does not prove or disprove hypotheses. In spite of these limitations, this work brings the voices of workers into the discourse regarding worker roles in the client job search process.

Discussion

Social welfare programs are complex constructs whose daily operations are determined by a multitude of factors ranging from federal laws to the actions of individuals. Within the system, individual workers put policies into practice using their own discretion. This study found that welfare workers in both a rural and an urban county in a southern state used their own social capital both to obtain information about available jobs and to exert influence to get clients hired.

Most interesting was the extent to which this behavior was expected and informally supported by state-level administrators, although not a part of the job officially. Also, a categorization of occasions that warranted the exertion of influence by workers emerged. Such action occurs with especially competent clients, especially vulnerable clients or in situations where many clients are in need. In the rural county, both the positive and negative effects of social capital were noted in conversations. There the job search of clients was impeded when employers had a negative impression of the client or the client's family. However, the closed social system that provided employers negative information about clients and their families also gave workers the opportunity to pressure business owners to hire clients. As one respondent succinctly stated, "It works both ways."

The finding that some workers used their own social capital to assist clients and that such action was implicitly expected by administrators indicates that the topic warrants further exploration and study. The final section addresses potential benefits and risks of such action. It also identifies avenues for future research and policy implications.

Potential Benefits

The most intriguing potential benefit of worker use of their own social capital in the client job search relates to one of the fundamental missions of the social work profession: poverty alleviation. The poor in the United States are typically isolated from the social and economic mainstream. In addition to being poor, welfare recipients are often disadvantaged due to closed social systems. Worker social capital offers a potential link to the socioeconomic mainstream and may be used to overcome the negative effects of social capital.

Such action by welfare workers also challenges one critical view of the welfare system by identifying elements within the system which promote progressive reform even when this is not dictated by law. The welfare system in the United States has historically been criticized for its remedial nature. The latest welfare reform has been criticized for being punitive and for focusing on decreasing caseloads instead of decreasing poverty. This research presents evidence that some workers in the welfare system have, on their own accord, surpassed the limited nature of current and historic policies. In effect, they have bridged the socioeconomic divide that isolates the poor from the nonpoor by using their own social networks to help clients.

These findings also identify instances where worker discretion is not categorically bad. Research on worker discretion generally portrays such action as detrimental to clients. Discretionary action often includes service rationing, rationalizing program objectives, delivering services only to the more cooperative clients (Goodsell, 1981; Lipsky, 1980), and failing to elicit information from clients or respond to their needs in order to move them through the system more quickly (Brodkin, 1997).

In addition to helping clients, worker use of their own social capital may benefit the agency and employers. When worker social capital use results in job placement, participation rates also increase, helping to secure funding for the agency. In addition, employers often hire individuals known by a personal contact because obtaining sufficient and accurate information about job candidates is difficult. The potential for a good match is increased in this case because the worker knows both parties. An added amount of social control exerted on the recipient by the worker may also make the placement more sustainable.

Potential Risks

While this form of discretionary action occurs and may benefit the client, agency and employer, potential risks of such action also exist. It may open the door to the development of dual relationships between social workers and their clients which is prohibited in the NASW Code of Ethics (1996) (Section 1.06.c). Using alters in one's personal network to obtain information about jobs or to exert influence to help clients gain employment makes it possible for clients to become members of the social networks of social workers.

In addition to the dangers faced by clients, workers may also be harmed by such practices. Expecting workers to use their personal resources to meet participation benchmarks could be viewed as worker exploitation. This can be compared to asking them to use their monetary capital to help clients. DeFilippis (2001) makes a similar argument about the use of social capital in community development initiatives. Expecting workers to use their own social capital in this way would be especially problematic if workers are not connected to individuals of a high social status or if they work in areas that are economically depressed.

Pressuring or requiring workers to use their own social capital may also damage their social capital because their relationship with a recipient is contractual rather than social. Social relations contribute to the formation of social norms, rules that regulate behavior in certain settings. Such norms serve as social capital when they facilitate individual action that would not be possible or would be more costly if they did not exist (Coleman, 1988, 1990). When individuals exert influence to get a friend or relative a job they expect the person to behave in a respectful manner in the job out of a sense of obligation and because of potential informal sanctioning by network alters. Since the client is not actually a part of the workers' network, this form of social control is not present, thus increasing the risk involved in such action. Such risk was discussed by one on the respondents.

   "... I talked them into hiring someone who didn't work out and
   so every time they say, `Oh, no. We're not hiring any more of your
   people.' I tell them, `look, it wasn't my fault.' That's why you have
   to be careful."

Furthermore, a potential administrative concern is that such action undermines the intent of the program. An underlying tenant of T.A.N.F. is encouraging the self-sufficiency of recipients. Expecting workers to find jobs for clients may shift the responsibility of the job search from clients to workers. Another administrative concern is that it may be impossible to promote the discretionary use of worker social capital while discouraging other types of discretionary action.

Future Research and Policy Implications

One of the most noteworthy changes that has occurred in the post-welfare reform era is the shift of priorities from eligibility and case management to job finding and maintaining placements. Respondents emphasized the shift in these work tasks throughout the guided conversations. Admittedly, such tasks are not new to the field of social work: job finding and coaching can be traced to supported employment programs for the mentally and physically challenged (Shaefer, Hill, Seyfarth, & Wehman, 1987). However, for welfare offices and workers, this is a new priority.

Elaborate formal models of job development exist; some states have merged Departments of Labor with welfare departments, while other areas hire specific job finders. In this study, nearly all workers were implicitly or explicitly responsible for job finding and coaching, even in the office with a designated job finder. Because these findings cannot be extrapolated to broader populations, further research should examine whether workers commonly evoke their social capital in offices with elaborate job development models and job finders. Thus, while it is important to acknowledge the role of the formal job finder and the parallels to supported employment, the implications from our research are substantively different.

Connections matter in finding jobs in the post-welfare reform era, albeit formal or informal connections. Currently, the ramifications of workers utilizing their informal connections to fulfill work requirements are unclear. This research documents some evidence of this practice but it does not address the prevalence or results of such action. This is a clear next step for future research. Such research is important because this phenomenon may be affecting welfare reform evaluation results without being detected.

While it is true that due to their position in the social status hierarchy, workers could use their social capital as a bridge to better jobs for clients, we are not prepared to advocate or disavow such action. We instead ask whether or not this is an action that should be expected or allowed by welfare administrators and the broader social work community. Given the potential benefits and risks noted earlier, should this practice be encouraged, discouraged, or formalized?

References

Beggs, J, Haines, V., & Hurlbert, J. (1996). Revisiting the rural-urban contrast: personal networks in nonmetropolitan and metropolitan settings. Rural Sociology, 61:306-25.

Bourdieu, Pierre (1985). Forms of capital. In John G. Richardson (Ed.) Handbook of Theory and Research for the Sociology of Education (Pp. 241-258). New York: Greenwood Press.

Brady-Smith,C., Jeanne Brooks-Gunn, Jane Waldfogel, and Rebecca Fauth. 2001. Work or Welfare? Assessing the Impacts of Recent Employment and policy changes on very young children. Evaluation and Program Planning, 24, 4:409-425.

Brayfield, A., and Hofferth, S.L. (1995). Balancing the Family Budget: Differences in child care expenditures by race/ethnicity, economic status, and family structure. Social Science Quarterly, 76, 158-177.

Brodkin, Evelyn Z. (1997). Inside the welfare contract: discretion and accountability in state welfare administration. Social Service Review. 71 (1): 1-33.

Coleman, J. (1988). Social capital in the creation of human capital, American Journal of Sociology, 94, 95-120.

Coleman, J. (1990). Foundations of Social Theory. Cambridge, MA: The Belknap Press of Harvard University Press.

Danziger, S., Corcoran, M., Danziger S., Heflin, C., Kalil, A., Levin, J., Rosen, D., Seefeldt, K., Siefert, K., & Tolman, R. (1999). Barriers to the employment of welfare recipients. Ann Arbor, MI: University of Michigan Poverty Research and Training Center.

Danziger, S., and Kristin S. Seefeldt. (2000). Ending Welfare Through Work First: Manager and Client Views. Families and Society. 81: 593-601.

DeFilippis, James (2001). The myth of social capital in community development. Housing Policy Debate, 12(4): 781-806.

Duncan, Cynthia (1992). Persistent poverty in Appalachia: Scarce work and rigid stratification. In C. Duncan (Ed., pp 111-134) Persistent poverty in America. Westport, CT: Auburn.

Duncan, Cynthia and Nita Lambourghini. (1994). Poverty and context in remote rural communities. Rural Sociology, 3 (59):437-461.

Freedman, S. (2000). National Evaluation of welfare to work strategies: four-year impacts often programs on employment stability and earnings growth, New York, NY: Manpower Demonstration Research Corporation.

Hoffereth, Sandra. (1999). Child Care, Maternal Employment, and Public Policy. Annuals of the American Academy of Political and Social Science, 563: 20-38.

Granovetter, Mark. (1973). The strength of weak ties. American Journal of Sociology, 78 (6): 1360-1380.

Granovetter, Mark. (1981). Toward a sociological theory of income differences. In Ivar Berg (ed.) Sociological Perspectives on Labor Markets. New York: Academic Press.

Goodsell, Charles T. (1980). Looking once again at human service bureaucracy, Journal of Politics, 42:761-778.

Kaye, Kelleen and Demetra Smith Nightingale (2000). The low wage labor market: Challenges and opportunities for economic self-sufficiency. U.S. Department of Health and Human Services: Washington, D.C. 2000.

Kalil, A., Heidi Schweingruber, and Kristin S. Seefeldt (2001). Correlates of Employment Among Welfare Recipients: Do Psychological Characteristics and Attitudes Matter? American Journal of Community Psychology, 29 (5): 701-723.

Lin, Nan. (1982). Social resources and instrumental action. In Peter V. Marsden and Nan Lin (Eds.) Social Structure and Network Analysis. Beverly Hills: Sage.

Lin, Nan. (1999). Building a network theory of social capital. Connections. 22: 28-51.

Lin, Nan, Walter M. Ensel, and John C. Vaughan. (1981). Social Resources and Strength of Ties: Structural Factors in Occupational Status Attainment. American Sociological Review 46: 393-405.

Lipsky, Michael. (1980). Street Level Bureaucracy: Dilemmas for the Individual in Public Service. New York: Russell Sage Foundation.

Lofland and Lofland. (1995). Analyzing social settings: a guide to qualitative observation and analysis. Belmont, Calif.: Wadsworth.

Long, D.A. (2001). From support to self-sufficiency: how successful are programs in advancing the financial independence and well-being of welfare recipients? Evaluation and Program Planning, 24: 389-408.

Meyers, M.K., Glaser, B. & MacDonald, K. (1998). On the front lines of welfare delivery: Are workers implementing policy reforms? Journal of Policy Analysis and Management, 17 (1): 1-22.

Midgley, J. & Rainford, W. (2000). Social investment in a era of welfare reform. Council on Social Work Education, Annual Program Meeting, New York.

Newman, Katherine. (1999). No Shame in My Game: The Working Poor in the Inner City. Alfred A. Knopf and The Russell Sage Foundation: New York.

Ong, P.M. (1996) Work and automobile ownership among welfare recipients. Social Work Research, 20(4): 255-62.

Sandfort, J. (1999). The Structural Impediments to Human Service Collaboration: Examining Welfare Reform at the Front Lines. Social Service Review, 73 (13): 314-334.

Shafer, M.S., J. Hill, J. Seyfarth, & P. Wehman. (1987). Competitive Employment and workers with mental retardation: Analysis of employers perceptions and experiences. American Journal of Mental Retardation, 92 (3): 304-211.

Tickameyer, Ann and Cynthia Duncan, (1990). Poverty and opportunity structure in rural America. Annual Review of Sociology, 16: 67-86.

Tonnies, Ferdinand. (1957). Community & society (Gemeinschaft und Gesellschaft) Translated and edited by Charles P. Loomis. East Lansing, Michigan State University Press.

U.S. Bureau of the Census. (1997). Bureau of Small Area Income and Poverty Estimates. Http://www.census.gov/hheswww/saipe/estimatetoc.html.

U.S. Census Bureau.(2000). 2000 Census of Population (P.L. 94-171) Redistricting Data File. Http://factfinder.census.gov.

U.S. Department of Labor. (2001). Employment and Earnings. Bureau of Labor Statistics, U.S. Department of Labor: Washington, D.C.

Vidich, Arthur and Joseph Bensman. (1958). Small Town in Mass Society: Class, Power, and Religion in a Rural Community. Princeton, NJ: Princeton University Press.

MICHELLE LIVERMORE
The Ohio State University
College of Social Work

ALISON NEUSTROM
Center for Families, Children and the Courts
Judicial Council of California
Administrative Office of the Courts