Abstract
Background
Underserved populations face many barriers to vocational rehabilitation (VR) services. However, outreach can build bridges with underserved populations.
Objective
This study compares Rehabilitation Services Administration (RSA) client data with US Census data to identify which demographic groups are underserved at the national level. This data analysis strategy can also be applied by agencies and providers to identify underserved populations at the local level.
Method
RSA 911 data from 77 agencies in the United States and its territories between the first quarter of 2021 through the third quarter of 2023 was compared to US Census data. RSA 911 data was analyzed, quantifying VR populations by demographic type. Diachronic analysis was utilized to identify service changes over time.
Results
This study found that several historically underserved demographic groups—including African Americans and Native Americans—were being served by VR at or above those groups’ representation among the general population. However, Hispanic/Latino, Asian, and female populations were underrepresented in VR applications.
Conclusion
VR agencies and providers have made significant progress with many traditionally underserved groups. However, Hispanic/Latino, Asian, and female populations are groups that require enhanced outreach efforts. In addition, agencies and providers can compare local client data with local Census figures to identify local underserved populations.
Introduction
The opportunity to work is a feature of full inclusion in society as it is a benchmark that defines adult identity (Strauser, 2020). However, among persons with disabilities, particularly those from diverse racial and ethnic background or persons living in rural areas of the United States, access and opportunity to quality employment has been beyond the reach for far too many (Tansey et al., 2023a).
To address the systemic barriers, the state-federal vocational rehabilitation (VR) system is tasked with providing "VR services to individuals with disabilities so that those individuals may prepare for and engage in competitive integrated employment consistent with their unique strengths, priorities, concerns, abilities, capabilities, interests, and informed choice." (US Department of Education, 2020).. However, despite the successes for the VR program in supporting individuals with disabilities obtaining and retaining employment, applications for services have precipitously declined over the years (U.S. Department of Education, 2020).
Far too many individuals with disabilities, including those from diverse backgrounds, are not included in the opportunity for full inclusion in the workplace. Considered the “underserved,” there is a need to identify best practices to conduct outreach to persons with disabilities from diverse racial, social, and geographic backgrounds who are among those intended to receive supports from the VR program but may have difficulties accessing theses services. Underserved communities can be wide-ranging and diverse in composure, background, culture, language, and needs. Underserved communities might be defined “as any community that is not served by VR at the same proportion as the general population.” This definition could include women, who, despite being a slight majority of the total population, may not be represented at that level in an agency's served clientele.
Underserved communities in VR services are the product of multiple factors. The lack of participation in VR programming may be due to historical discrimination or victimization of certain groups in society, systemic barriers limiting access to social and economic resources, cultural or social stigma regarding the public identification of disability, cultural and language differences when seeking services, geographical barriers, or lack of effective outreach to these communities. Persons from diverse communities have often experienced historical prejudices. For example, “redlining” practices barred certain racial groups from owning homes (Anderson et al., 2021).
Hispanic and foreign-born workers are disproportionately employed in dangerous and low-income jobs (Smith et al., 2006). A dramatic example of breach of trust between federal agencies and Native Americans occurred during widespread sterilization of native women as recently as the 1970s (Torpy, 2000). These examples highlight how barriers were established between specific groups and the wider society. The role that public agencies played in some of these historical actions should be considered as a potential barrier to overcome between state offices and minority groups.
Americans in minority communities are underserved in education. Educational attainment decreases significantly for racial minorities in the United States. While 6.1% of white Americans do not complete high school, that rate nearly doubles in African American, Asian, and Native Hawaiian/Pacific Islander communities. Among Native Americans, adults without a high school diploma rises to 22% (US Census, Table S0201, 2023b). Educational barriers cross-over to the vocational landscape. While the unemployment rate among white Americans is currently at 3.4%, this number doubles for Native Americans and African Americans (US Census, Table S0201, 2023b). Unemployment creates barriers to community living, reduces social engagement, increases health issues, creates a sense of isolation, and undermines a person's sense of self (Tansey et al., 2023b). VR acceptance rates between 1985 and 1998 show that whites were accepted at higher rates than African Americans (Rosenthal et al., 2005). A fear of stigma may hinder some racial minorities from not accepting their disability, which, in turn, will discourage such a person from seeking VR services (Groomes et al., 2011). In sum, unemployment can reduce life satisfaction (Groomes et al., 2011) as well as mire individuals in generational poverty.
Poverty rates match these educational and unemployment demographics. Poverty among white Americans is currently at 9.8%, but poverty rates among African American and Native American populations doubles to 21% (US Census, Table S0201, 2023b). Poverty, coupled with limited access to higher education that results in low educational attainment and unemployment, isolates individuals from the resources that can alleviate these issues. Native American communities, for example, experience high rates of work-related disability, coupled with high levels of poverty and unemployment, as well as more barriers to educational opportunities than the wider population (Faubion et al., 1998). In general, populations with less access to education and with limited economic resources tend to have poorer overall health (Braveman et al., 2010).
Cultural, social, educational, and economic differences can also create barriers. Migrant farmworkers may purposely avoid support services–even in the case of workplace injury–for fear of arrest or deportation (Smith et al., 2006). Cultural differences can inadvertently impede VR services (Askew et al., 2012). For instance, interviews with VR counselors to Cherokee residents in Oklahoma found that non-native counselors misinterpret native body language and do not understand the role of the spiritual healer in the VR process (Faubion et al., 1998). A VR service area may have many non-English-speaking residents, and the VR service agency may not have a counselor who can interact with this community. Thus, such populations can become “underserved,” due to legal concerns, cultural differences, or financial barriers.
Geography can create underserved communities. Rural communities generally have more poverty and more disability per capita than urban populations (Ipsen et al., 2019). Yet, many VR service areas are spread far and wide. In such situations, the agency may strive to reach these populations, but the agency does not have the personnel or resources to reach a wide geographic zone.
Thus, there are multiple ways in which a group may become “underserved.” One category of underserved population includes a pattern of prejudice that has been experienced by a specific population over a long period of time. A second category is a group who experiences current social, educational, or economic barriers. Yet another category can become underserved due to geographic and logistical limitations within a VR service area. People with severe disabilities may represent more intensive VR cases, and this population may thus become underserved.
Fortunately, despite the barriers that can create underserved communities, VR has been shown to be effective in countering the disadvantages within such communities (Anderson et al., 2021). Employment provides a way for people to become confident and more independent, while also yielding additional positive gains for a person's physical and mental health (Tansey et al., 2023b). Employment gives access to important social participation and a sense of community living (National Institute on Disability, Independent Living, and Rehabilitation Research, 2024). Funding sources exist for institutions to alleviate barriers to underserved populations. Section 21 of the 1992 Rehabilitation Act Amendments provides funds from the National Institute on Disability, Independent Living, and Rehabilitation Research for institutions that support underserved populations, including Historically Black College and Universities and tribal colleges (Administration for Community Living, 2024). In addition, the Vocational Rehabilitation Outreach Guide, developed by the Vocational Rehabilitation Technical Assistance Center for Quality Employment (VRTAC-QE), is a publicly available resource (https://tacqe.com/vr-outreach-guide/) that provides guidance for VR providers who seek to improve or expand upon outreach services. The issue remains that many individuals who might benefit from VR services are not included in this effective program.
The VRTAC-QE seeks to support the capacity of state vocational rehabilitation agencies (SVRA) to identify underserved populations within their states and develop structured, methodical approaches to addressing the barriers to participation in services among those populations. This paper asks questions relevant to VR outreach to underserved communities. The analysis investigates if VR is serving individuals from different racial or ethnic backgrounds at the rate of those groups’ composition within the general population. This study also examines the rate at which VR serves men as compared to women, and it also examines if VR service has fluctuated its caseloads in recent years with respect to different demographic groups. Rehabilitation Services Administration (RSA) 911 and US Census data is used to analyze these questions. Outreach resources, materials, and examples are also included in relation to these questions.
Methods
Research questions
RSA 911 data was collected from 77 agencies in the United States and its territories between the first quarter of 2021 through the third quarter of 2023. This data includes general agencies and blind agencies. This dataset includes demographics related race and sex. The US Census Bureau provides control data by which the RSA 911 data can be compared. US Census data (https://data.census.gov) is publicly accessible. Census table identifications are provided within the in-text citations to identify specific data sources.
The following questions provide a data-driven examination of underserved populations by race and gender. The analysis examines if current VR outreach is appropriately reaching demographic groups at a rate comparable to that group's representation in the general population. These questions shed light on the current state of VR outreach to traditionally underserved communities.
Question 1: Are VR agencies serving individuals from different racial or ethnic backgrounds at a comparable rate to each group's prevalence in the general population? Question 2: Are VR agencies serving people from each sex at an equal rate? Question 3: Are VR agencies serving clients at a stable rate, or do service volumes fluctuate over time?
The first question, “Are VR agencies serving individuals from different racial or ethnic backgrounds at a comparable rate to each group's prevalence in the general population?”, was analyzed by filtering out each demographic group from the RSA 911 dataset. The filtered dataset could then be compared to US Census figures.
RSA 911 data tabulates sex for each client, which allows for comparisons with the general population through US Census data. Thus, the research question, “Are VR agencies serving people from each sex at an equal rate?”, can be examined.
The third research question, “Are VR agencies serving clients at a stable rate, or do service volumes fluctuate over time?”, presents a diachronic examination of VR services over the three-year period between 2021 and 2023.
An important feature of RSA 911 and Census data is that both data collections account for multi-racial individuals. The RSA 911 data tabulates each individual with multiple racial identities if multi-racial. A multi-racial individual qualifies, for example, as African American and white if he or she is biracial between these two groups. For the purposes of this study, a biracial individual within the RSA 911 data—listed as both African American and white—qualifies as African American for the purposes of this study.
Results
Data analysis: Question 1
Data analysis provides insights into the question, “Are VR agencies serving individuals from different racial or ethnic backgrounds at a comparable rate to each group's prevalence in the general population?” RSA 911 served 2,283,392 clients between the first quarter of 2021 through the third quarter of 2023. Of these clients, over half a million participants were African American (Table 1), while Hispanics accounted for 380,699 clients and Native American and Asian individuals each accounted for just over 50,000 clients in each category. About 16,000 Pacific Islanders were served in this period.
VR clients served between 2021 and 2023 by demographic type.
These demographics figures can be converted into percentages to realize the ratio of that demographic's representation within VR services (Table 2). These figures can then be compared to the ratio of that demographic within the general population via US Census data. The Census ratio for Native Americans, African Americans, Asians, and Pacific Islanders was derived from US Census, Table P1, 2020a, and the Census ratio for Hispanics was derived from US Census, Table P9, 2020b.
The ratio of VR clients is juxtaposed with corresponding demographic ratios among the general population. Asian and Hispanic populations are not served at their level of representation.
Asian and Hispanic populations are not served at their level of representation. The ratio of African American, Native American, and Pacific Islander clients served by VR in this period match or exceed the ratio of that population's representation in the general population. African American clients represent nearly one in four VR clients served during this period. However, there is a discrepancy of −3.44% for Asian populations and −2.06% for Hispanic populations. These ratios are visualized in Figure 1.

The ratio of VR clients (grey bars) is juxtaposed with the ratio of that demographic within the general population (black bars). VR ratios are lower among Asian and Hispanic populations when compared to the wider society.
Table 3 examines disability proportions within different demographic groups. The data in Table 3 is derived from US Census, Table S1810, 2023c. Among the total United States population, 13.60% (nearly 45 million people) have a disability. Thus, the percentage of 13.60% can be used as a control number to compare each demographic group. Asians and Hispanics are the two demographic groups that report ratios of disability lower than the control figure (13.60%), with disability within the Asian population nearly half (8.40%) that of the general population. It is interesting that these two demographic groups that report lower ratios of disability (Asian and Hispanic) correspond to the same demographic groups that are underrepresented among VR clients.
Census figures show that the total population registers a disability ratio of 13.60%. Asian and Hispanic populations have disability figures that are lower than this control ratio.
Data analysis: Question 2
Data analysis finds a discrepancy at the national level with regards to the question, “Are VR agencies serving people from each sex at an equal rate?” Table 4 juxtaposes RSA 911 data with Census figures, comparing how the gender ratio of VR clients relates to the sex ratio of the general population. Census data in Table 4 is derived from US Census Bureau, Table S0101, 2023a. Despite the fact that women account for a slight majority of the general population, there is a discrepancy against women in VR services, with men accounting for 56% of cases and females accounting for 44% of cases.
Men are served at a higher ratio than women in RSA 911 data.
Data analysis: Question 3
The third research question, “Are VR agencies serving clients at a stable rate, or do service volumes fluctuate over time?”, presents a diachronic examination of VR services over the three-year period between 2021 and 2023. Table 5 tabulates the change in service numbers over this period. The dramatic increase in every demographic group is visualized on Figure 2. Asian and Hispanic representation increased the most in this period, both showing a five-fold increase from 2021 to 2023. It is possible that the rapid increase in VR services is due to the return of workers to the workforce following the end of the COVID-19 pandemic era.

VR cases increased in every demographic group between 2021 and 2023.
RSA 911 data shows dramatic increases in clients in each demographic category between 2021 and 2023.
Applying demographic analysis to local VR offices
VR providers can do data analysis for their local regions to determine which communities might qualify as “underserved.” The Vocational Rehabilitation Outreach Guide (https://tacqe.com/vr-outreach-guide/), developed by the VRTAC-QE, provides a publicly available spreadsheet link to assist VR professionals. This spreadsheet link can be found in Appendix A of the manual (VRTAC-QE, 2024). Instructions for the spreadsheet are provided in the outreach guide. The example VRTAC-QE outreach spreadsheet (Table 6) provides cells to add local demographic data and VR client data. Once the local data is entered, other cells will automatically compute the percentages served in that area against the percentages populating that area. VR offices can then quickly identify which demographic areas are not being served at the proportion of those living in that area.
The Vocational Rehabilitation Outreach Guide (https://tacqe.com/vr-outreach-guide/) provides a spreadsheet in appendix A that allows for users to input local numbers to identify underserved communities within their service area, as shown in this example spreadsheet below.
Discussion
Analysis of RSA 911 data from 2021 through 2023 shows areas of strength and potentials for growth. On the positive side, all examined ethnic and racial demographic groups in this study showed dramatic increases in VR application during the 2021 through 2023 timeframe. All these groups (Asian, African American, Native Hawaiian and Pacific Islander, Hispanic and Latino, and Native American) come from traditionally underserved communities. Nearly one in four VR recipients in this period was African American, demonstrating that VR providers have made tremendous progress in traditionally underserved population. This analysis is consistent with Groomes et al. (2011) that found no distinction between African Americans and European Americans in terms of acceptance of disability, potentially a product of an increasingly open society.
In addition, the RSA 911 data, relative to US Census data, shows that most of these traditionally underserved communities are served by VR at or above their representation in the general population. However, Hispanic and Asian populations were not represented in VR applications at rates corresponding to their proportion in the general population. The data shows Asian and Hispanic populations reporting lower proportions of disability as compared to the general population. Whereas the total United States population registered a disability rate of 13.6%, Asians were nearly half this rate at 8.4%, and Hispanic or Latino populations registered a disability rate of 10.6%. As such, Asian and Hispanic/Latino populations may be culturally less likely to claim disability, and thus, less likely to seek VR services.
VR services increased each year for all examined demographic groups in this study. The period 2021 to 2023 follows the COVID-19 pandemic era. The dramatic rise in VR cases in this period may reflect a large “return to work” period following the pandemic. The numbers also show that VR offices were reaching these demographic groups with increasing regularity, and in this period, no group decreased in service, at least at the national level. Women were revealed as an underserved community at the national level. Although females represent a slight majority in total population (50.52%), they represent only 43.91% of VR applicants. VR offices may consider examining their local data to see if females represent 50% of cases. The case of female underrepresentation is a good example of how a large population (even a majority population) can still be underrepresented.
Data analysis can identify what populations are underrepresented in a service area. However, the goal of the VRTAC-QE is not to admire the problem (Tansey et al., 2023a); rather it is tasked to develop systematic approaches that address the barriers to employment among persons with disabilities by supporting SVRAs and increasing the ability of these agencies to assist individuals with disabilities achieve quality employment outcomes. At the outset of technical assistance, the VRTAC-QE implemented a data-driven model for each SVRA receiving technical assistance to guide and evaluate outcomes. However, the use of data for planning and evaluation purposes was intended to model an approach as part of the technical assistance and build agency capacity to use their data to identify areas for improvement and as a metric to assess their own performance.
SVRAs can compare their client data with corresponding demographic data from the US Census Bureau. Spreadsheets, such as the one provided by the Vocational Rehabilitation Outreach Guide, developed by the Vocational Rehabilitation Technical Assistance Center for Quality Employment (https://tacqe.com/vr-outreach-guide/), can compute the ratio of VR clientele and compare it with the local general population to find disparities in service.
Outreach services offer opportunities to reach new populations. Literature provides successful examples of outreach services to underserved populations. Outreach services have found success with migrant workers, isolated rural communities, non-English-speaking populations, Native American communities, and other underserved populations.
Cultural barriers in VR services do have solutions. The Kentucky Migrant Vocational Rehabilitation Program (KMVRP) engaged with the migrant worker community. KMVRP utilized Spanish-speaking outreach workers who know the migrant culture and can operate within the community. VR materials were translated into Spanish, information was put in local Spanish-language media, and community members were included in the program's Advisory Board. Spanish-speaking outreach workers were hired to make personal connections within the migrant community. The outreach workers participated in the social life of the community. This program resulted in over 100 referrals to the Kentucky Office of Vocational Rehabilitation (Breeding et al., 2005).
Rural distances can be overcome. In a rural community, a limited use provider, such as a local teacher, can liaison with VR offices. Thus, in a rural community, somebody who has local roots and connections to the people and employers in a rural community can be a great asset to a VR agency (Ipsen et al., 2019). VR workers recommended the use of independent contractors to facilitate VR in such underserved communities. Agencies can sponsor trainings for such limited use providers (Ipsen et al., 2019). In addition, technology offers new opportunities that did not exist in previous eras. Distance and transportation barriers may be overcome by remote work or self-employment (Ipsen et al., 2019).
Patterns emerge in outreach to different communities. While migrant communities and small towns in rural areas may have many differences–including language and cultural norms–they share similar barriers to VR, as they are both isolated from VR offices and counselors, have limited employment options, and likely have higher poverty and unemployment than their urban counterparts. Nevertheless, the migrant community and the small town in a rural setting can both benefit from a local “insider” who provides a link to the agency office.
The outreach practice of having a local person who is an “insider” can be applied to other scenarios. For instance, Native American VR providers in Oklahoma have cultural knowledge that non-native providers do not have. Smaller caseloads among the Cherokee in Oklahoma grant time allowances for providers to visit client residences. Native providers in such scenarios can appropriately liaison between native clients and the state VR agency. Culturally appropriate practices can be employed; for instance, traditional medicine from a spiritual healer in a native community can be coded as “mental health services” (Faubion et al., 1998).
Urban areas also have underserved communities. Anderson et al. (2021) successfully implemented culturally responsive evaluation (CRE) methods to better consider the backgrounds of VR consumers who resided in low-income urban environments. By employing these culturally sensitive methods, the researchers learned from this community that barriers to employment involved daily life in poverty, fear of violence, lack of transportation, and the lack of job openings in poor communities. Anderson et al. (2021) noted that it was common for clients to arrive late for VR appointments, even though the client had left early, but the public transportation was late. A flexible solution to urban transportation issues is the use of phone, email, or other electronic technology in lieu of VR office visits.
People with intellectual disabilities once constituted an underserved community in VR. Apart from the societal barriers, many individuals with intellectual disability encounter low expectation barriers that impact access to quality employment. Clearly, when their voices are heard, people with intellectual disabilities reveal that they want to work and want to be treated as equals at work (Voermans et al., 2021). By the 1990s, SVRAs collaborated with other agencies and community advocates to increase the representation of people with intellectual disabilities in the labor market through practices such as supported employment (Mast & West, 2001). Much progress has been made to find vocational success for individuals with intellectual disabilities, but it required understanding of the underlying causes of why this group was underserved, data supporting those beliefs, and the development of models of practice (i.e., supported employment) capable of promoting long-term success in the workplace. Through this history, progress in VR outreach is possible, and progress with underserved populations begins with outreach.
From its inception, the VRTAC-QE recognized the importance of increasing outreach to underserved populations (Tansey et al., 2023b). Through the initial efforts to provide technical assistance to SVRAs on effective outreach strategies, we recognized that there was a need to develop a structured, data-driven approach to reach persons with disabilities and provide information regarding VR services. The VRTAC-QE outreach toolkit was designed for this very purpose and provides a mechanism to develop a sustainable outreach plan intended to identify and address the limited participation in VR of specific groups. The structure of the toolkit begins with data-driven planning. That is, before engaging in outreach, SVRAs benefit from identifying which groups, in what part of their states, are not being served in relation to their proportion of the larger population in those areas and what are reasonable targets to set for use in evaluation of any outreach efforts. Next, the outreach toolkit promotes accountability from all stakeholders in the SVRA program, beginning with identifying a coordinator in the agency down to those non-governmental organizations who these coordinators must develop relationships by which to support the distribution of information about the agencies and provide a bridge of trust between government agencies and individuals experiencing ambivalence regarding participating in the VR program (Tansey et al., 2017). Lastly, critical to the implementation of any program to increase participation is to have clear metrics of progress and successful outcomes. In the absence of these priori identified metrics, agencies seeking to increase participation may not appreciate changes that occur or highlight anecdotal evidence of improvement in participation with documented support for those claims.
As the toolkit has only recently been launched, the data on its efficacy is still unknown. However, through the program evaluation of the VRTAC-QE, we anticipate being able to provide evidence regarding the overall framework as well as a detailed analysis of effective or ineffective components toward building an evidence-based model to reverse the historical downward trend in applications to the VR program.
Conclusion
RSA 911 data shows that VR providers are reaching many historically underserved populations at or above the demographic's representation in the general population. For instance, nearly one in four VR clients between 2021 and 2023 were African American. VR client cases increased each year for every demographic group examined in the study. However, the study found discrepancies in the proportion of Hispanic and Asian participants. But data also shows that Hispanic and Asian groups also report disability at a much lower rate than other populations, which indicates there may be a cultural factor precluding those groups from initiating VR contact. Women were found to be underrepresented in VR service at the national level as well. VR offices can compare their client data with US Census Bureau data to determine which local populations are underserved. Successful patterns for outreach exist, and many involve similar techniques, including utilizing cultural “insiders” in underserved communities, a practice that has been successful with migrant communities, Native American communities, rural towns, and urban areas. Beyond admiring the current problems of engaging persons from diverse populations to participate in the VR program (Tansey et al., 2023a), an opportunity has been presented to SVRAs to address the challenges of outreach to these groups and to adopt methods that will allow them to systematically approach increasing involvement of underrepresented groups.
Footnotes
Acknowledgements
The authors thank vocational rehabilitation service agencies and providers for collecting and reporting data to RSA.
Ethics statement
This study analyzed public, de-identified data from RSA 911 and the US Census and therefore did not require institutional review board approval.
Informed consent
This study involved analyses on public, de-identified data and therefore did not require informed consent.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The contents of this manuscript were developed under a grant, the Vocational Rehabilitation Technical Assistance Center for Quality Employment, H264K200003, from the U.S. Department of Education. However, those contents do not necessarily represent the policy of the U.S. Department of Education, and you should not assume endorsement by the Federal government.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
