Rural Research Needs and Data Sources for Selected Human Services Topics

Contents

• The Rural Context
• Review of Existing Research on the Three Focal Topics
• Data Sources Available to Conduct Research on the Focal Topics
• Implications of Study Findings
• Enhancing Rural Human Services Information

The U.S. Department of Health and Human Services (DHHS) is an essential partner in protecting and enhancing human capital and self-sufficiency throughout rural America. According to a report prepared by the DHHS Rural Task Force, the department administers some 225 programs, services, and grants in rural areas. In 2002, based on task force findings and input from rural research experts, DHHS announced its goal of conducting more and better research to inform state, local, and federal policymakers about the needs of rural communities, with a particular emphasis on human services topics.

As a first step, staff members in the Office of the Assistant Secretary for Planning and Evaluation (ASPE) formulated a study to learn more about social and economic conditions and trends in rural areas, identify high-priority family and community needs, and assess current knowledge about such needs and the services available to meet those needs in rural areas. The projects main goal was to identify data that could support empirical research, whether sponsored by DHHS or other entities, on understudied issues. ASPE contracted with Mathematica Policy Research, Inc. (MPR) to conduct the Social and Economic Conditions in Rural Areas study. This report describes the studys activities and findings and presents information on 80 data sources that could be used by the research community to study select human services topics in rural areas.

The Rural Context

For a better understanding of human services conditions in rural America, we first needed a clear picture of rural areas themselves. During the first phase of the project, MPR reviewed secondary sources describing social and economic conditions and trends in rural areas. The picture provided by the review is complex, because rural areas are neither demographically nor economically homogeneous. However, all rural areas share two distinct features. First, rural areas are changing. The second characteristic that unites rural areas is their differences from urban areas.

The rural population makes up 17 percent of the total U.S. population, or 49 million people, spread over 80 percent of the nations land mass. The proportion of the total population that is rural continues to shrink, although urban expansion, high immigration and birth rates, and in-migration of retirees have all boosted population in some rural areas. In addition to changes in population size, rural areas also are becoming more diverse racially and ethnically.

The economic picture of rural areas is mixed. Lower levels of human capital distinguish rural labor markets from urban ones. Unemployment rates have been similar in rural and urban areas since the 1990s, but residents of rural areas are more likely to be underemployed. Wages in rural areas are lower than those in urban areas. Poverty rates are higher, but the cost of living is lower, so official measures may overstate somewhat the impact of rural poverty. Rural poverty is persistent, however: 95 percent of persistent-poverty counties are rural.

Rural areas, as a whole, are more disadvantaged than urban ones across numerous physical health indicators, and access to health care facilities is more limited. Rates of mental health disorders do not differ appreciably between urban and rural areas, however. Rates of health insurance coverage are similar in rural and urban areas, though private insurance is less common in rural areas, and spells without insurance coverage are longer. About 9 percent of homeless assistance provider clients are located in rural areas, though this measure may understate rural homelessness somewhat. Data are not available to estimate rural rates of domestic violence.

Research shows the structure, access to, and use of, social supports and institutions to be different in rural and urban areas. Lack of transportation is a key barrier to employment and to accessing services in rural communities. Nearly 40 percent of rural counties have no form of public transportation. Rural schools are smaller and provide fewer resources for their students than schools in urban areas. However, rural teachers report safer learning environments, less student misbehavior, and less student alcohol and drug use than teachers in central cities. Rural families are less likely than urban ones to use formal child care, but informal child care arrangements offer advantages that may be important to rural parents, including lower child care expenditures. Rural families are more likely to report earnings, and less likely to be on welfare, though welfare recipients in rural areas face more employment barriers than their urban counterparts, including low skills, a lack of transportation, and child care problems. Since the passage of welfare reform in 1996, welfare caseloads have fallen faster in rural than urban areas.

Geographic and sociocultural factors, along with low population densities and limited organizational resources, affect service delivery in rural areas. Studies suggest that these and other rural characteristics provide benefits to recipients of social services in rural areas, as well as imposing costs.

Review of Existing Research on the Three Focal Topics

Based on the review and discussions with rural experts, three human services issues were selected as focal topics for the literature and data compilation: (1) work supports for low-income families, (2) substance abuse, and (3) child welfare services. Although many rural human services issues could benefit from additional empirical research by the research community as a whole, there were compelling reasons for focusing on these three topics for the study. Support for finding and maintaining employment is particularly valuable for low-income families in rural areas, where economic and community conditions can make it difficult to secure steady employment and achieve self-sufficiency. Recent evidence suggests that the prevalence of drug and alcohol use and abuse among youth and adults in rural areas is becoming as high as, or higher than, the prevalence in urban areas. Finally, the possible effects of child maltreatment on children, families, and communities are substantial in rural areas as well as in urban ones, but empirical research on topics related to child welfare has mostly excluded rural areas.

A review of recent empirical literature on each topic, including a discussion of research gaps, was produced during the second phase of the project.

Work Supports for Low-Income Families. In the late 1990s, Congress overhauled welfare and workforce development programs. Implementing the changes has been challenging in both rural and urban areas. Building the service network required to engage Temporary Assistance for Needy Families (TANF) recipients in work-related activities may have been a difficult challenge for rural sites, but rural areas also enjoyed some advantages in implementing and operating TANF and One-Stop workforce development programs. While employment conditions and geographic isolation of rural areas can make job placement and workforce development more difficult, rigorous studies of welfare reform programs do not support the notion that rural clients have less access to TANF employment services or participate in them at lower rates. Existing studies do not offer consistent results regarding the effects of welfare reform programs for rural clients. Welfare reform has been comparatively well studied in rural areas, but the effects of work supports are not ascertained in many existing studies. Research is needed on workforce development services offered through rural One-Stops, and on transportation and child care subsidies in rural areas.

Substance Abuse. Long thought of as an urban issue, substance abuse has emerged as an issue in rural areas as well. The rates of use and abuse in rural areas are still lower than the rates in urban areas; however, in recent years differences have narrowed. Tobacco use is more common in rural areas than in urban areas for both youth and adults. Alcohol use is also more common among rural youth than urban youth, but less common among rural adults than urban adults. Studies of youth and adult drug use indicate that the prevalence of illicit drug use is declining and remains lower in rural areas than in urban areas, although rates vary by type of drug. Increases in the production and trafficking of illicit drugs in rural areas have raised concerns about potential impacts on rural crime, drug use, and even rates of child abuse and neglect, though evidence of these impacts has not yet been found.

Empirical research confirms that substance abuse services differ between rural and urban areas, and that treatment access is limited in rural areas, although the effects of these constraints have not been examined empirically. To understand the implications of substance abuse in rural areas and to design services to meet the needs of rural clients, additional research is necessary on rates of substance use and abuse in rural areas, especially among racial/ethnic, cultural, and other population subgroups; on the availability of treatment and prevention services; and on the effectiveness of those services.

Child Welfare. Empirical research on child welfare, including maltreatment and child welfare services, has traditionally focused on urban areas where caseloads are largest. Comparisons of rates of child maltreatment in rural and urban areas over time have not been conclusive, and are based on very small samples. Foster care caseloads grew more quickly in rural than urban areas from 1990 to 1999, and the characteristics of children placed in foster care, and foster care spells and outcomes, differ between rural and urban areas. For example, compared with children in urban areas, rural children placed in foster care are more likely to return to their families to live, rather than be adopted.

Child welfare services and practices differ between rural and urban areas, but little can be said about how access to and use of child welfare services differ. The effectiveness of child welfare services has not been studied in rural areas. Further research is needed on all aspects of rural child welfare to understand better the prevalence of child maltreatment, the services available in rural areas, and the use and effectiveness of services.

Data Sources Available to Conduct Research on the Focal Topics

During the projects third phase, information about data that could be used to study the focal topics in rural areas was collected from three sources: federal, nonfederal, and state. Federal data sources include national and regional surveys and databases collected or sponsored by federal agencies. Nonfederal data sources include surveys or databases collected and sponsored by private agencies or organizations. State data sources are administrative or monitoring data collected by state agencies.

MPR collected information on 19 national data sources and one multistate data source that include rural observations and identifiers and thus can be used to study aspects of one or more of the three focal topics. Information on 60 state administrative data sources was also collected from 23 states that have at least 30 percent of their populations living in rural areas, or that have high poverty rates and high proportions of rural residents (but less than 30 percent). Volume 2 of this report describes characteristics of each data source  for example, its purpose, sample size, number of rural records and rural sampling methodology (if any), and confidentiality or other restrictions that might limit our access to the data.

Implications of Study Findings

Rural America is diverse, ever changing, and different from the urban areas in which most Americans live. The story of differences between rural and urban areas is not a simple one  rural life offers families both advantages and disadvantages. Nevertheless, in contrast to urban areas, rural areas experience at least one disadvantage: less is known about their human and social services conditions, the social services they need and use, and the effectiveness of those services.

One of the difficulties in conducting rural research is finding suitable data. The federal and nonfederal data sources described in this report are well known and have characteristics that make them valuable for rural human services research, though not ideal. They are collected mainly for the purpose of research and so are characterized by rigorous and well-defined sampling and/or data collection methodologies and instruments. Most are from ongoing studies or included multiple waves of data collection. Their data are in the public domain, are readily accessible for research, and are supported by codebooks and other published documentation. Several research gaps identified in this report could potentially be addressed by researchers using documented federal-nonfederal data sources. For example:

• The Survey of Income and Program Participation could be used to study participation in work support programs and services in rural areas, including job training, job subsidies, and transportation assistance.
• The National Longitudinal Survey of Youth (1997 panel) oversampled black and/or Hispanic respondents. It could potentially be used to examine training, participation in government assistance programs, and alcohol and drug use patterns for these understudied groups living or working in rural areas.
• The Alcohol and Drug Services Study collected information from treatment facilities, which were oversampled from rural areas, so information on treatment services is available on nearly 500 facilities located in nonmetro counties. It is thus a potentially rich source of useful descriptive information on rural substance abuse treatment needs and services.
• Data from the National Educational Longitudinal Study could be used to estimate rates of smoking and alcohol/drug use for students in rural schools, and to correlate measures of use with school, community, and family factors, to explore risk factors and support the development of rural prevention approaches.
• Longitudinal Studies of Child Abuse and Neglect data could potentially be used to construct detailed case studies to describe child and family experiences in rural child welfare systems.

State administrative data sources such as those included in this report also have both strengths and weaknesses for research. They are relatively inexpensive for researchers to obtain compared with surveys or original data collection, typically contain very large samples, and generally contain consistent data elements. Data can often be linked over time or across programs to provide longitudinal and comprehensive information on program use. An especially important benefit of administrative data for rural research is the availability of detailed geographic identifiers for each record, which facilitate the identification of rural observations and their classification into multiple typologies  though researchers would have to work closely with state agencies to use these identifiers, to ensure confidentiality and protect privacy. Because administrative data are not collected for research, however, users must invest time in understanding, cleaning, and structuring such data to prepare them for analysis. Variations in data quality may raise reliability issues, and inconsistencies in identifiers may hamper data linkages. There are several examples of rural research gaps that might be addressed using the state administrative data sources documented in this report:

• Although complete national data on the use of child care and transportation subsidies are currently unavailable, numerous states (for example, Iowa, Oklahoma, South Carolina, South Dakota, and Wyoming) maintain databases on child care assistance or include flags for receipt of child care and transportation vouchers in their TANF or family service databases.
• National data on Workforce Investment Act (WIA) services are unavailable currently, but several states collect detailed data on WIA services, which could be used to examine the types of services rural One-Stop clients receive.
• States provide child welfare data to the federal governments Adoption and Foster Care Analysis and Reporting System. Some states maintain or are developing state databases that included additional data elements not required for reporting, however. Data from these states could support more detailed analysis of child welfare systems in rural areas than can be conducted using data made available through the federal source.
• States provide information on substance abuse treatment and child welfare to the federal government, which removes (or does not collect) detailed geographic identifiers such as zip codes. By working with the states to ensure confidentiality and privacy protection, researchers can use detailed geographic identifiers to examine rural facilities and clients.

Enhancing Rural Human Services Information

Better data and more research on rural human services are needed and would be valuable. The various findings of this study suggest that entities that fund or sponsor research on rural or human services topics, as well as the organizations and individuals who plan and conduct such research, could take steps to improve the quantity and quality of rural human services information:

• Include Rural Populations, Areas, or Systems in More Studies. Entities that sponsor or conduct human services research  particularly through large national or regional studies and surveys  should more often include rural people, areas, or systems in studies.
• Incorporate Rural Sites into Program Evaluations. Since nearly one-fifth of the nations population live in rural areas, differences in the impacts and costs of programs that serve rural families could be large, both in social and in budgetary terms. Therefore, including rural sites and samples in evaluations, or conducting evaluations specifically designed for rural areas, could improve rural programs and policies.
• Oversample Rural Sites and Populations. Rural populations are small. This can make statistical analysis less precise or preclude the use of sophisticated analytic approaches. Oversampling of rural areas is an important option for improving rural research, conducting more sophisticated analyses, and better identifying significant rural findings or rural-urban differences. It is particularly important when there may be differences among racial/ethnic, cultural, or other demographic or community subgroups.
• Report Rural Findings. Many national and regional studies do include rural sites. But if rural issues are not a specific focus of the study, or if key findings do not differ between rural and urban sites, report authors generally do not include discussions of rural experiences and findings in published reports, or even provide information on the breakdown of sample members by rurality. Providing such information could help answer many important rural research questions.
• Make Better Use of Existing, Detailed Rural Classification Systems. Detailed and informative classifications of rural areas have been developed for use in demographic and economic studies. To date, however, they have been little used in research on poverty and human services issues. As a result, little information is available to study variation across diverse rural areas, or to capture the complexity of rural-urban differences. To the extent possible, rural data should include geographic identifiers that can support use of detailed rural classification typologies, and researchers should make more use of alternative rural classification approaches.
• Disclose Rural Definitions and Classifications Used in Studies. Study authors should disclose the definitions used to classify rural observations. Failure to do so makes it difficult to interpret rural research findings, as well as to summarize and synthesize findings across studies.
• Add Information to Make Small, Region-Specific Rural Studies More Generalizable. The rural human services research literature is composed largely of small, region-specific studies. Findings from such studies can be useful, in the absence of nationally representative studies. Their generalizability could be improved, if, in addition to including operational definitions of rurality, authors provided detailed descriptions of rural samples, along with descriptive and demographic information on rural study sites.