EFFECTS OF AGENCY STRUCTURE AND CLIENT CHARACTERISTICS ON REHABILITATION SERVICES

AND OUTCOMES FOR CONSUMERS WHO ARE BLIND

Brenda S. Cavenaugh

Mississippi State University

June 1999

ABSTRACT

Consumers and practitioners in blindness rehabilitation support the premise that blind persons have unique vocational rehabilitation (VR) needs and are best served in identifiable agencies, established especially for that purpose. The scarcity of empirically-based data supporting this position, however, has hampered objective dialogue within the disability community regarding the continued funding of separate (blindness-only) VR agencies. This study investigated differences in VR services (expenditures, number, and duration) and outcomes (competitive sector placement and earnings) of legally blind consumers in states with separate or combined (cross-disability) agencies.

The sample included 35,396 legally blind consumers closed in the 50 states by the state VR system in 1995 and 1996. Case data from 1995 were used to identify client disability and demographic characteristics related to competitive closure and to construct two covariates to control for these characteristics in the investigation of VR services and outcomes. The first covariate, the Index of Work Disadvantage at Referral (IWDR), was constructed using a summed weighting system applied to categories of demographic variables. The weights were based on frequencies and simple correlations with outcome. The second covariate, the Demographic Predictor (DP), was the predicted score for outcome from a step-wise multiple regression with appropriately coded disability and demographic variables entered as predictors. Both covariates were derived by applying their respective procedures to the 1996 data and then used comparatively in separate covariance analyses.

A multivariate analysis of covariance (MANCOVA) using variables aggregated by state found no significant differences in the combined set of dependent variables across agency structure types after adjusting for demographic differences using either covariate. An analysis of covariance (ANCOVA) used to investigate differences in client earnings across agency structure types showed that client earnings at closures were significantly higher in separate agency states than in combined agency states, when employing the IWDR covariate but not significantly higher when using the DP covariate. A second ANCOVA used to investigate differences in competitive sector placement determined that placement rate was significantly higher in separate agency states than in combined agency states, when employing either covariate.

Although the latest available data from RSA were used in this research, there is a strong need for additional research of RSA-911 data from previous fiscal years and of new data as it is released.

TABLE OF CONTENTS

Page

CHAPTER

I. INTRODUCTION.. 1

Statement of Problem......... 9

Hypotheses......... 10

Rationale for the Study......... 10

Limitations......... 12

Definition of Terms......... 13

II. LITERATURE REVIEW............ 18

Theoretical Support for Specialized VR Agencies 18

History of Specialized Rehabilitation Programs......... 22

The Emergence of State Commissions or Agencies......... 23

History of the State-federal VR Program......... 24

Research on the Efficacy of Specialized Agencies......... 28

The Mallas Study......... 28

The J. W. K. Study......... 29

Kirchner and Peterson......... 31

The NAC Study......... 32

Cavenaugh and Pierce......... 32

Demographic Characteristics that Predict VR Outcome......... 33

Summary of the Literature Review......... 37

III. METHODOLOGY... 39

Research Design......... 39

Advantages of Design......... 40

Restrictions of Design......... 40

Participants......... 41

Available Data......... 41

Database Cleaning and Variable Recoding......... 41

Development of Covariates......... 42

Aggregation of Variables for Hypotheses Testing......... 43

Variables......... 43

VR Structure Type......... 43

Competitive Sector Placement......... 44

Earnings At Closure......... 45

Services......... 45

Procedures......... 46

Development of IWDR......... 46

Validation of the IWDR......... 49

Reliability......... 49

Development of DP Covariate......... 50

Hypotheses Testing......... 51

Data Analysis......... 52

IV. RESULTS AND DISCUSSION.......... 55

Results of Development of Covariates......... 55

Descriptive statistics......... 55

Development of the IWDR......... 56

Development of DP Covariate......... 61

Results of Hypotheses Testing......... 64

Descriptive Statistics......... 64

Aggregation of client data......... 64

Hypothesis 1......... 65

Hypothesis 2......... 68

Hypothesis 3......... 70

Discussion......... 71

Utility of the Covariates......... 72

Differences in Services......... 73

Differences in Outcomes......... 74

Findings Across Studies......... 76

Limitations......... 77

V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS............ 80

Summary......... 80

Theoretical Basis of Research......... 80

Purpose and Hypotheses......... 81

Literature Review......... 82

Methodology......... 84

Findings......... 86

Conclusions........... 87

Recommendations for Further Research......... 91

REFERENCES 93

APPENDIX

A......... SEPARATE AND COMBINED AGENCY STATES......... 101

LIST OF TABLES

TABLE Page

4.1 Index of Work Disadvantage at Rehabilitation Referral............ 59

4.2 Correlations of IWDR Variables with Competitive

Sector Closure for 1995... 60

4.3 Point Biserial Correlations of Summed Index of

Work Disadvantage Scores with Competitive

Sector Placement for 1989, 1992, 1994, and 1995... 61

4.4 Summary of Backward Stepwise Regression Analysis

for Variables at VR Referral Predicting Placement

at VR Closure for 1995 ... 63

4.5 Table of Means and Standard Deviations for FY 1996

Aggregate Data... 65

4.6 Intercorrelation Matrix State Aggregate Data 1996 (N = 50) 66

A.1 Separate and Combined Agency States (N =50)... 102

CHAPTER 1

INTRODUCTION

The Rehabilitation Act of 1973, as amended, authorizes the allocation of federal funds on a formula basis to the states and territories for the administration and operation of a vocational rehabilitation (VR) program (commonly referred to as the state-federal VR program) to assist individuals with disabilities in preparing for and engaging in gainful employment. To be eligible for services from a state VR agency, an individual must have a disability that “requires vocational rehabilitation services to prepare for, secure, retain, or regain employment” (Rehabilitation Act Amendments of 1998). State VR agencies provide a wide range of services (e.g., physical restoration, counseling and guidance, vocational training, maintenance, job referral and placement) to assist people with disabilities in locating employment.

The state-federal VR program began in 1920 with passage of the Smith-Fess Act (P.L. 66-236). Early rehabilitation services were limited to vocational guidance and vocational education, along with occupational adjustment and placement services, and were restricted to persons with physical disabilities (Rubin & Roessler, 1995). During the 1920s and 1930s, blind persons were considered to have limited, if any, vocational potential and accordingly received little benefit from the initial VR legislation (Clunk, 1966). However, vocational opportunities for consumers who are blind began to expand with passage of the Randolph-Sheppard Act of 1936 (P.L. 74-732) and the Wagner-O’Day Act of 1938 (P.L. 75-739). (The Randolph-Sheppard Act enabled persons who are blind to operate vending facilities in federal buildings, while the Wagner-O’Day Act mandated the federal government to purchase products made by blind employees of sheltered workshops.)

As more blind people demonstrated their ability to be successfully employed, public perceptions slowly began to change. Consequently, federal support specifically directed toward the provision of VR services to blind consumers was included in the second major rehabilitation legislation, the Barden-LaFollette Act of 1943 (P.L. 78-113). This Act broadened the rehabilitation program by allowing then existing state agencies, commissions, or private agencies serving blind persons to administer the state-federal VR program for individuals with blindness (Rubin & Roessler, 1995). Subsequent legislation, including the recent Rehabilitation Act Amendments which were a part of the Workforce Investment Act of 1998, have retained language allowing states to designate a state agency, or another agency, providing assistance to adults who are blind as the “sole State agency to administer the part of the plan under which vocational rehabilitation services are provided for individuals who are blind” (PL 105-220) and to designate a different state agency as the sole agency to administer the remaining VR services. This legislation has resulted in blind consumers receiving services in (a) states with two VR agencies (one specialized agency serving only persons who are blind and one general agency serving persons with other disabilities) or (b) states with a single agency operating under one “State Plan” serving persons across all disabilities.

The administrative branch of the state-federal VR program is located in the Rehabilitation Services Administration (RSA) of the Department of Education, Office of Special Education and Rehabilitative Services. RSA provides oversight to the 82 VR agencies located in the 50 states, the territories, and the District of Columbia. Within this oversight, each state is responsible for designating a state agency to administer VR services. States may also choose to designate a second state agency to administer services for individuals who are blind (Rehabilitation Act of 1973, as amended). In 25 states, a separate VR program is authorized to provide services to persons who are blind or visually impaired (Moore, Huebner, & Maxson, 1997). Because these separate agencies may restrict services to those consumers with the most severe visual impairments (e.g., legally blind or progressive visual impairment), consumers with less severe visual impairments (e.g., those who are not legally blind) are sometimes served in the VR agency coexisting with the separate agency in the same state (Cavenaugh & Pierce, 1998).

In each of the remaining 25 states, the U. S. territories, and the District of Columbia, consumers who are blind or visually impaired are served in one combined VR agency, which provides rehabilitation services to consumers with all disabilities. In some of the combined agencies, specialized blindness staff (e.g., administrators, counselors, rehabilitation teachers, orientation and mobility instructors) are located in an identifiable subunit and are responsible for the separate administration and service delivery of all services to blind clients (e.g., Oklahoma, Tennessee). In other combined agencies, little or no specialized service delivery staff are available for the provision of services to blind clients (e.g., Georgia, Wyoming). In illustration, Lewis and Petterson (1998) found that while all separate agencies serving blind clients provided specialized rehabilitation teaching services, only 90% of combined agencies provided these services to blind clients.

Variability in the types of specialized blindness services available in the different VR agencies may be affected by individual state mandates (e.g., services to both children and adults), unique history and tradition, state financial commitments, agency order of selection (ensuring persons with the most significant disabilities are served first) and financial need policies, and availability of blindness specialized staff (e.g., orientation and mobility instructors, rehabilitation teachers, low vision specialists). Thus, differences within VR agency structure types (e.g., separate and combined) that exist across states have confounded efforts to investigate VR outcomes (JWK International Corporation, 1981; Kirchner & Peterson, 1982; Management Services Associates, 1975).

The existence of separate agencies serving blind clients has resulted in an ongoing debate within the disability community regarding the benefits from funding two VR agencies in one state--one serving blind consumers and another serving consumers with other disabilities (JWK International Corporation, 1981; Kirchner & Peterson, 1982; Management Services Associates, 1975). With the flurry of activities associated with the recent reauthorization of the Rehabilitation Act of 1973, arguments for and against continued funding of separate VR agencies serving blind consumers were, and continue to be, intensely debated and remain in the forefront of rehabilitation issues (Edwards, 1997; National Council on Disability [NCD], 1997a; NCD, 1997b). Proponents for separate VR agencies have argued that their dissolution will result in the loss of specialized blindness services critical to the rehabilitation and independent living of consumers who are blind or visually impaired (Augusto, 1997; Jernigan, 1996). At the same time, opponents have argued that administrative costs of two distinct VR agencies in one state are duplicative and possibly inequitable for persons with disabilities other than blindness (NCD, 1997a).

Although blind consumers historically have supported the dual agency model of service delivery (Gallagher, 1988; Hopkins, 1991; Joint Organizational Effort, 1994; Rusalem, 1961), other disability groups have tended to favor a cross-disability model, in which one single VR agency would serve all disability types (e.g., persons with deafness, blindness, deaf-blindness, orthopedic impairments, muscular dystrophy, multiple sclerosis, spinal cord injuries, psychological disorders, mental retardation, traumatic brain injury, cystic fibrosis, and heart conditions). The tension between these two perspectives was most recently articulated in a March 1997 statement by the National Council on Disability (1997a). As part of its involvement in the recent Rehabilitation Act reauthorization process, NCD initially recommended that the RSA discontinue funding of separate VR agencies for clients with visual impairments. Facing major resistance from blindness-related consumer and professional groups, however, NCD later withdrew its recommendation. Instead, NCD asked that the General Accounting Office (GAO) initiate a study to investigate differences in the performance, benefits, and costs of separate and combined agencies (1997b).

While NCD’s statements were congruent with its cross-disability philosophy, the Council justified its recommendations largely by pointing to the absence of conclusive empirical research to validate the claim that separate agencies are more effective. The NCD also acknowledged that its position was directly opposed to that of organizations supporting blind people and testimony of blind consumers during related public hearings.

Given the ongoing debate regarding the efficacy of separate VR agencies serving blind consumers, it is not surprising that the current number of states (25) with separate agencies is considerably less than the one-time high of 42 (Hopkins, 1991). As early as 1974, the National Council of State Agencies for the Blind (NCSAB) reacted to the decline in the number of separate agencies by commissioning the first study investigating the relationship of agency structure and program effectiveness (Management Services Associates, 1975). While results of this and subsequent studies (JWK International Corporation, 1981; Kirchner & Peterson, 1982) did not provide conclusive evidence that separate agencies were more or less effective than combined agencies in serving blind clients, JWK and Kirchner both noted that demographic characteristics of blind consumers influencing employability may differ between agency structure types. More recent studies have also reported that a higher percentage of blind consumers served in separate agencies report secondary disabilities (Cavenaugh & Pierce, 1998; National Accreditation Council, 1997), are older, have less education, receive transfer payments (e.g., Social Security disability), have more severe vision loss, and are non-White (Cavenaugh & Pierce, 1998).

While no study has investigated the efficacy of state VR agency structure types, while controlling for demographic characteristics of blind consumers, research has shown that client characteristics, singularly and in combination, rather than functional limitations associated with the medical impairment alone, are related to VR competitive employment outcomes (Wright, 1980). For example, a number of studies have found that age, ethnic background, education, and public assistance at referral are strong predictors of competitive closure outcomes (Bellini, Neath, & Bolton, 1995; Bolton, 1979; Giesen & D’Amato, 1992; Lewis & Bolton, 1986; Moriarty, Wall, & McLauglin, 1988; Vandergoot, 1987). Findings that blind consumers of separate agencies are more likely to be socially and economically disadvantaged than blind consumers of general agencies (Cavenaugh & Pierce, 1998; JWK International Corporation, 1981; Kirchner & Peterson, 1982; NAC, 1997) would suggest that differences in demographic characteristics need to be considered and accounted for in studies comparing agency structure types.

Statement of Problem

Although research has identified client disability and demographic characteristics as predictors of employment outcomes (Bellini, Neath, & Bolton, 1995; Bolton, 1979; Giesen & D’Amato, 1992; Lewis & Bolton, 1986; Moriarty, Wall, & McLauglin, 1988; Vandergoot, 1987), no study has investigated differences on service and outcome measures across agency structure types, after adjusting for differences in client characteristics. Further, research has suggested that blind consumers in separate agencies are more socially and economically disadvantaged than blind consumers in combined agencies (Cavenaugh & Pierce, 1998; JWK International Corporation, 1981; Kirchner & Peterson, 1982; NAC, 1997). In response to these findings, this proposal is directed toward determining if differences in VR services received and outcomes achieved by legally blind persons exist between separate and combined agency states, after controlling for client demographic and disability characteristics.

Hypotheses

In order to investigate whether blind consumers differ in services received and outcomes achieved in separate and combined agency states, the following null hypotheses are proposed:

H₀₁: There is no statistically significant difference in number of rehabilitation services, case service expenditures, and duration of services among legally blind VR consumers across state VR structure types (separate agency state, combined agency state), after controlling for client disability and demographic characteristics at referral.

H₀₂: There is no statistically significant difference in weekly earnings at VR closure among legally blind consumers across state VR structure types (separate, combined), after controlling for client work disability and demographic characteristics at referral.

H₀₃: There is no statistically significant difference in competitive sector placement rates among legally blind VR consumers across state VR structure types (separate, combined), after controlling for client disability and demographic characteristics at referral.

Rationale for the Study

Representatives of all major consumer groups (i.e., American Council of the Blind, Blinded Veterans Association, Canadian Council of the Blind, National Federation of the Blind) and professional organizations (Association for Education and Rehabilitation of the Blind and Visually Impaired, American Foundation for the Blind, Canadian National Institute for the Blind, National Library Service for the Blind and Physically Handicapped) in the blindness field have jointly signed a position statement indicating that it is their common experience that “specialized, comprehensive services and essential changes in social attitudes about blindness do not occur when rehabilitation services for the blind are provided through a single program which serves both blind and disabled persons” (Joint Organization Effort, 1994, p. 1). Despite this widespread belief, the paucity of supporting empirical research threatens the future existence of separate VR agencies. Organizations supporting combined agencies that serve all disability groups have called for a halt to current RSA authority permitting separate agencies for blind consumers (NCDa, 1997; Spungin, 1997).

If significant public policy changes regarding continued funding of separate agencies occur, it is imperative that those changes be based on findings of studies investigating differences in the services received and outcomes achieved realized by blind consumers served in both agency structure types. This study contributes to the existing body of knowledge regarding consumer characteristics, services, and program outcomes of VR consumers who are blind. It examines differences in the VR services provided and outcomes attained by legally blind consumers who are served in separate agency states and combined agency states, after controlling for disability and demographic characteristics. Therefore, this study can assist policymakers in their determination of the efficacy and value of separate VR agencies serving blind consumers.

Limitations

This study utilized data collected by the state VR agencies and reported in RSA-911 national case service reports. While these reports include client referral, service, and outcome information on all cases closed by the state-federal program, it does not include client information regarding other potential predictors of employment outcomes, such as powerlessness (Moriarty et al., 1988); onset of blindness (Giesen & D’Amato, 1992), and adjustment to blindness and intelligence (Bauman & Yoder, 1966).

An obvious concern in investigating VR agency structure types relates to the lack of consistency among agencies in the type and amount of blindness-specific services provided to consumers who are legally blind. As noted in previous studies investigating relationships of VR outcomes and agency structure types (JWK International Corporation, 1981; Kirchner & Peterson, 1982; NAC, 1997), the types of specialized services available to blind clients (e.g., rehabilitation teaching, orientation and mobility, low vision, computer access technology) may vary within combined and separate agency states. While aggregate data for separate agency states and combined agency states will be provided in this study, individual states within structure type may report broad differences in services and outcomes. Therefore, readers interested in comparing a specific VR program with results of this study are urged to also review individual agency data, when possible.

As with the majority of rehabilitation research (Bolton & Parker, 1998), this study will use an ex post facto design. The design is commonly used because “many of the phenomena of interest to rehabilitation researchers are not and cannot be under the control of the researcher” (Bolton & Parker, p. 455). For example, assignment to agency types cannot be manipulated by the researcher. Because of these limitations, causal relationships cannot be detected. However, this design has proved valuable in allowing researchers to study relationships in situations where experimental manipulation is impractical (Bolton, 1979; Borg & Gall, 1989).

Definition of Terms

Using the reporting manual for the RSA-911 case service report (RSA-PD-95-04, 1995) when appropriate, several key terms have been defined as follows:

Client income: Earnings, interest, dividends, and/or rent as reported on the RSA-911 to describe the individual’s largest single source of support at application and at closure.

Combined agency states: The 25 states with a single, combined VR agency operating under a single State Plan and providing services to persons with all disabilities.

Competitive employment: Work for wages, salary, commissions, tips, or piece-rates, not including work in extended employment.

Competitive sector placement: Includes Competitive employment, state managed Business Enterprise Program (BEP), or Self-Employed placements.

Computer Access Specialist: Specialized professional who provides training in computer access equipment (e.g., braille, large print, and speech computer systems for people who blind or visually impaired).

Cross-disability organizations: Organizations serving a variety of disability types.

Extended employment: Work for wages or salary in a setting conducted by a nonprofit organization for persons with disabilities unable to enter into or not ready for competitive employment (referred to as “sheltered workshop” placements in earlier RSA Case Service Reports).

Homemaker: Men and women whose principal activity is keeping house for their families or themselves, if they live alone.

Legally blind: Blindness in both eyes, with a correction of not more than 20/200 in the better eye or a limitation in field within 20 degrees (RSA major disability codes 100-119).

Non-competitive sector placements: VR consumers closed status 26 (successful) in homemaker, unpaid family workers, and extended employment statuses and all unsuccessful cases (statuses 08, 28, and 30).

Orientation and mobility instructor: Specialized professional who provides blind or visually impaired people with training in orientation and mobility skills and in use of adaptive equipment that enable them to develop or enhance their ability to travel independently.

Rehabilitation teacher: Specialized professional who provides blind or visually impaired people with training in a variety of areas, including communication (e.g., braille, writing) activities of daily living (e.g., cooking, cleaning, sewing, dressing), and low vision.

RSA-911 case service report: Client referral, service, and outcome data reported annually to RSA on all cases closed during each federal fiscal year.

Self-employed: Work for profit or fees in one’s own business, farm, shop, or office, excluding BEP.

Separate agency states: The 25 states with two VR agencies--one responsible for serving consumers with primary disabilities of blindness and another responsible for serving consumers with other disabilities.

Specialized services: Services (orientation and mobility, rehabilitation teaching, low vision, computer access technology) provided by qualified professionals in meeting the unique needs of persons who are blind or visually impaired.

State-agency-managed business enterprise (BEP): Vending facilities and other small businesses managed by persons with severe visual impairments and under the supervision of the state VR agency.

Status 08 closures: Clients not accepted for VR services from referral status (status 00), applicant status (status 02), or from extended evaluation services (status 06).

Status 26 closures: Clients accepted for services and closed “rehabilitated” (competitive employment, extended employment, self-employed, BEP, homemaker, and unpaid family worker).

Status 28 closures: Clients accepted for services and closed “not rehabilitated” after Individualized Plan for Employment (IPE) is initiated.

Status 30 closures: Clients accepted for services and closed “not rehabilitated” before (IPE) initiated.

Transfer payments: Types of public support received during the VR process, including Social Security Disability Insurance (SSDI); Supplemental Security Income (SSI-aged, SSI-blind, SSI-disabled); Aid to Families with Dependent Children (AFDC); General Assistance; and Veterans disability.

Unpaid family worker: Work status in which client performs unpaid family work that cannot be classified according to any of the Dictionary of Occupation Titles occupations.

Weekly earnings at closure: Includes total wages, salaries, tips, commissions, and profits from self-employment earned as regular income before payroll deduction in the week before VR closure.