EMPLOYMENT NETWORKS: the SUPPLY SIDE of the TICKET to WORK-WORK INCENTIVES IMPROVEMENT ACT (PL 106-170) DISSERTATION Presented I

EMPLOYMENT NETWORKS: THE SUPPLY SIDE OF THE TICKET TO WORK-WORK INCENTIVES IMPROVEMENT ACT (PL 106-170)

DISSERTATION

Presented In Partial Fulfillment of the Requirements

for the Degree Doctor of Philosophy in Rehabilitation Services

of The Ohio State University

Judith L. Drew, ABD, CRC

********

The Ohio State University

2005

Dissertation Committee: Approved by

Associate Professor Bruce Growick, Advisor ______

Associate Professor Michael Klein Advisor College of Education Professor W. Bruce Walsh

ABSTRACT

The purpose of this project was to explore and describe the opinions and attitudes of potential Employment Networks (ENs) as factors in the successful implementation of the Ticket to Work-Work Incentives Improvement Act, PL 106-170. This groundbreaking legislation, which allows for beneficiary choice in a vocational rehabilitation provider, can be conceptualized in terms of supply and demand. The

“demand” side of the law is reflected in the number of beneficiaries who are interested in receiving services to become employed. The “supply” side is simply the number of rehabilitation providers who are equally interested in becoming ENs, and assisting beneficiaries in obtaining employment.

To date, much time and effort has been expended on notifying beneficiaries of their rights and incentives for returning to work. Meanwhile, relatively less effort has taken place in exploring and describing the attitudes of providers who have not become

ENs. The overall success of the Ticket to Work and Work Incentives Improvement Act

(TWWIIA) can be measured not only by the number of beneficiaries who take advantage of these new incentives, but also by the number and quality of the providers who become and remain an Employment Network.

This project proposed to assist the Social Security Administration (SSA) in maximizing the recruitment and maintenance of ENs by surveying a sample of the community of private, for-profit and non-profit vocational rehabilitation providers who

have elected not to participate in the Ticket to Work-Work Incentives Improvement Act

(Ticket) program as an EN. This project helped to determine what specific barriers/issues have prevented greater participation in the program by this group of rehabilitation providers.

This information was analyzed to provide SSA and Maximus, the Ticket program manager, with insight into the incentives and barriers for participation by prospective

ENs. Eight issues were assumed to be of primary importance for ENs. These issues were the following: 1) the payment system; 2) payment options/ issues; 3) delays in reimbursement; 4) the relationship between state vocational rehabilitation agencies

(SVRAs) and ENs; 5) EN reporting requirements; 6) EN requirements/credentials; 7) criteria for the evaluation of ENs; and finally, 8) capitalization/start-up costs.

The researcher also used the survey to explore three additional areas related to the potential ENs. They were 1) the respondents’ need for technical assistance and training;

2) their opinions regarding how the Ticket program could be integrated into the current

SSA disability determination process; and 3) what effect, if any, did their geographic location, size of their organization, or number of services they provided, impact their decision to participate in the Ticket program. The eight issues and additional areas of exploration were identified based on the following factors: 1) literature review;

iii

2) professional experiences of the researcher; 3) input of an expert panel; and 4) on anecdotal information collected informally from ENs around the country as the Ticket program has been implemented.

This survey research accomplished two primary objectives. The first objective was to determine which factors most heavily influenced the decision of potential ENs not to participate in this program. The second objective was to determine the relationship between (a) the Ticket program components, and the demographic characteristics of participants, and (b) the decision of rehabilitation providers not to participate as an

Employment Network. These objectives were accomplished primarily by the use of a web-based survey.

The overall results of this study reinforce the fact that major regulatory changes are needed to ensure the success of the Ticket program. Unfortunately, 63% of the respondents to the survey indicated that they were not interested in becoming ENs and participating in the Ticket program. The following reasons were cited for this lack of interest: 1) structure of the payment system for ENs, (2) the nature of the relationship between ENs and the State-Federal VR system (SVRA), (3) the administrative burdens of the program for ENs, and (4) the lack of defined criteria for evaluating ENs and the nature of the dispute resolution process. Within each of these areas of concern, the study uncovered reasons for the unwillingness of ENs to participate in the Ticket program.

The payment system as currently structured for ENs is overly complex and complicated. Almost one-third (30%) of the respondents did not understand the payment system, and cited it as a barrier to their participation. The majority of respondents preferred a payment system that is a combination of hourly payments, and a lump sum for

placement and follow-along. Furthermore, 71% agreed that ENs should be paid for services that lead beneficiaries to employment at a level less than the current SGA.

Respondents to the survey also cited the inequities in payment to SSI recipients versus

SSDI beneficiaries as a barrier to participation. A majority (55%) felt that the payment structure for ENs should be the same for the two groups, because the current inequity creates a 'financial and service dilemma'. Finally, the concern of ENs on the potential of overpayments to beneficiaries after return-to-work, which causes a delay in payment to them for their services, was also cited as a barrier to participation.

Interestingly, 64% of the ENs felt that they should be treated in the same fashion as SVRAs, both in terms of overall payment and review by SSA. More specifically, prospective ENs reported that when they refer Ticket holders to SVRA for collaborative services, they are concerned about how they will get paid eventually by SSA for their services. Moreover, many ENs are concerned about competing with SVRAs for ticket holders, and are afraid of losing funding for many of the other services they provide in the community, which are paid for by SVRA. Overall, the nature of the relationship between the ENs and the SVRAs needs to be modified so that there is more cooperation and less competition. In this way, ENs would feel at less of a disadvantage, and the program would serve more beneficiaries.

On a positive note, most ENs (66%) felt that the application process and requirements for participating as an EN in the Ticket program were not burdensome.

However, many providers (63%) viewed the requirement of tracking pay stubs of beneficiaries who had returned to work as excessively burdensome. Consequently, the

SSA’s decision to remove that administrative burden from providers was welcomed.

Many ENs (42%) suggested that SSA develop a computer tracking system for ENs so that their participation in the program can be simplified.

A majority (74%) of the respondents felt that the criteria for evaluating ENs should go beyond earnings of the SSA beneficiary, and include type and level of job placement. This response seemed to imply that the nature and severity of the ticket holders’ disability should be considered when EN performance is being evaluated.

Additionally, a fair number of respondents (38%) were not comfortable with the current rules surrounding dispute resolution.

In order to improve the current Ticket program, respondents were also asked to make suggestions for system change. Forty percent agreed that they would be more likely to participate in the Ticket program if there was an established referral process with information available about beneficiaries. The most important pieces of information to receive at the time of referral were medical and work history. Fifty-six percent of the respondents agreed with the statement that “SSA should consider time-limited benefits as an motivator for program participation,” while only 26% disagreed with the statement.

Also, 62% of the respondents felt that a ticket should be provided to beneficiaries at the time of application for disability rather than at the determination of disability. Finally, only 25% of the respondents felt that the Ticket program should be mandatory, not voluntary, for all beneficiaries.

In conclusion, the most pervasive result of this survey was the fact that many ENs are simply unaware of the many nuances and complexities of the Ticket program and are extremely reluctant to participate in the program as presently structured. Therefore, it is strongly recommended that SSA consider modifying the Ticket program so that it is more

'supply-side' oriented, and engage in a concerted effort at educating the provider community on ways in which the Ticket program can be successful for them. In this way, both SSA and its beneficiaries can have the access to return-to-work services that they deserve.

vii

DEDICATION

This dissertation is dedicated to the two men in my life whose love and support helped make this accomplishment possible. First, to my husband, Rob, my technology guru, I owe many thanks. Your sacrifices touched my heart, and your continual encouragement uplifted me even though you were often hundreds of miles away. Then, many thanks go to my son, Jonathan, whose belief in my dream gave me the courage to keep going during my most challenging moments. I give my heartfelt love and thanks to you both.

viii

ACKNOWLEDGMENTS

First and foremost, I want to thank my faculty advisor, Dr. Bruce Growick, for encouraging me many years ago, to pursue my lifelong dream of getting my PhD. His support, enthusiasm, and belief in my abilities and scholarship helped me spread my scholarly wings. I have grown as a scholar and have discovered a love of research because of this opportunity.

I would be remiss if I did not thank the rest of my dissertation committee, Dr.

Michael Klein , and Dr. W. Bruce Walsh. They provided invaluable insight into both the content of the dissertation and the in and outs of the bureaucracy. In addition, they encouraged my growth as an educator and researcher. Dollie Curry, department secretary, also deserves many thanks for her assistance with the formatting of this document, constant encouragement, humor, and prayerful support.

A special thank you goes to Angel Rhodes for our friendship and scholarly conversations that were always interspersed with fun and frolic. She is the one person I know with whom I could engage in stimulating conversations about research and statistics one minute, and the next, be cheering the Buckeyes on to victory. She will always be my research buddy, and fellow Buckeye. Go Bucks!

I also want to thank my special friends Ginny Perelson, Kristin DeKuiper, and

Mary McMullen for their encouragement and listening ears. They never doubted my

ability to achieve my dream, and stood by me throughout the last three and one-half years with cards, phone calls and wonderful words of wisdom and encouragement.

Finally, a huge thank you goes to my parents. First to my Dad, who, through his passing, gave me the freedom to relocate hundreds of miles away from home and pursue my dream. I have continually felt his love and presence. Lastly, to my Mom, who through her own struggles with Alzheimer’s, continues to encourage and cheer me on as her “Queen” when she can not remember I earned my doctorate. Thank you for your love and support.

x VITA

2004 ...... ABD, Rehabilitation Services, The Ohio State University

1985 ...... M.A. Rehabilitation Counseling R.I. College, Providence, R.I.

1976 ...... B.A. Special Education R.I. College, Providence, R.I.

1991 – present ...... Vocational Expert/Rehabilitation Counselor, VocWorks, Cumberland, R.I.

2005 – present ...... Adjunct Faculty, Salve Regina University, Newport, R.I.

2001- 2004 ...... Graduate Teaching Assistant, The Ohio State University

2002 - 2004 ...... Research Assistant, The Ohio State University

1990 – 2001...... Adjunct Faculty, Assumption College, Worcester, M.A.

2001 – present ...... Career Counselor, Delta Consultants, Providence, R.I.

1990 – 1991 ...... Vocational Rehabilitation Counselor Rehabilitation and Re-Employment, Providence, R.I.

1988 – 1990...... Vocational Rehabilitation Counselor Conservco, Swansea, M.

xi 1984 – 1988 ...... Director, Vocational Services Mental Health Services, Johnston, R.I.

1981 – 1983 ...... Trainer, USDOL, Job Corps Washington, D.C.

1976 – 1980 ...... Special Education Teacher Harmony Hill School, Chepachet, R.I.

PUBLICATIONS

1. Drew, J. & Growick, B. (2004). The Role of the Vocational Expert in Social Security. Journal of Forensic Vocational Analysis, Vol. 7, p. 51-67.

2. Growick, B., & Drew, J. (2003). The Ticket to Work: The Unintended Consequences of an Imperfect Law. Journal of Forensic Vocational Analysis, Vol. 6, p.49-54.

FIELDS OF STUDY

Major field: Education Rehabilitation Services Minor field: Research

xii

TABLE OF CONTENTS

ABSTRACT...... ii DEDICATION...... viii ACKNOWLEDGMENTS ...... ix VITA...... xi LIST OF TABLES ...... xv LIST OF FIGURES...... xviii CHAPTER 1 ...... 1 1. INTRODUCTION ...... 1 1.1 Statement of the Problem ...... 1 1.2 Importance of the Study ...... 6 1.3 Design of the Study...... 20 1.4 Assumptions of the Study...... 21 1.5 Limitations of the Study...... 22 1.6 Research Questions/Hypotheses...... 26 CHAPTER 2 ...... 29 2. LITERATURE REVIEW...... 29 2.1 Overview to Employment as a Part of Disability Legislation...... 29 2.2 Overview to Ticket to Work and Work Incentives Improvement Act...... 30 2.3 Historical Antecedents of the Law...... 33 2.4 Highlights of the Ticket Legislation ...... 34 2.5 Use of the Internet in Survey Research...... 40 CHAPTER 3 ...... 45 3. METHODOLOGY ...... 45 3.1 Survey Population...... 45 3.2 Survey Design...... 45 3.3 Instrument Reliability...... 49 3.4 Survey Implementation ...... 50 3.5 Data Analysis...... 52 CHAPTER 4 ...... 55 4. RESULTS...... 55 4.1 Population and Response Rates ...... 55 4.2 Initial Survey Response Rate...... 55 4.3 Extension Period ...... 56 4.4 Response Rate Following the Extension Period...... 57 4.5 Overall Response Rate for the Initial and Extension Period ...... 58 4.6 Statistical Analysis...... 59 CHAPTER 5 ...... 141

xiii

5. CONCLUSIONS AND IMPLICATIONS...... 141 5.1 Overall Implications...... 141 5.2 Interest in Becoming an EN...... 143 5.3 Primary Factors for Non-Participation for Prospective ENs...... 144 5.4 Unanticipated Findings of This Study...... 155 5.5 Other Disability Policy Issues Related to the Ticket to Work...... 156 5.6 Additional Recommendations for Ticket Program Changes...... 158 5.7 Summary...... 164 REFERENCES...... 165 APPENDICES...... 171 APPENDIX A List of Panel of Experts...... 172 APPENDIX B Cover Letter to Panel of Experts...... 173 APPENDIX C Survey Notification Letter...... 174 APPENDIX D Cover Letter for Mailed or Faxed Surveys...... 176 APPENDIX E Drawing Postcard ...... 177 APPENDIX F Reminder Postcard...... 178 APPENDIX G Tables and Figures ...... 179 APPENDIX H Welcome to the Prospective Employment Network Survey...... 199

xiv

LIST OF TABLES

Table 4.1 Response Rate for the Initial Survey ...... 56 Table 4.2 Response Rate for the Extension Phase ...... 58 Table 4.3 Overall Response Rate for the Initial and Extension Phase...... 59 Table 4.4 Geographic Location and Interest in Being an EN...... 62 Table 4.5 Chi Square Test for Geographic Location and Becoming an EN...... 63 Table 4.6 Chi-Square Test for Interest in Becoming and ...... 65 EN and Number of Services Offered ...... 65 Table 4.7 Interest in Becoming an EN and Number of Services Offered ...... 66 Table 4.8 Interest in Becoming an EN and Payment System as a Barrier ...... 69 Table 4.9 Chi-squared test for Interest in Becoming and EN and Payment ...... 70 System as a Barrier...... 70 Table 4.10 Interest in Being an EN and Payment Issues and Options as Barrier ...... 78 Table 4.11 Chi-Square Test for Interest in Being an EN and Payment Issues and Options ...... 79 Table 4.12 Chi-Square Test for Interest in Becoming an EN and Waiting 45 Days for Payment...... 93 Table 4.13 Interest in Becoming an EN and Waiting 45 Days for Payment as Barrier..94 Table 4.14 Chi-Square Test for Interest in Becoming an EN and Loss of Funding ...... 98 Table 4.15 Interest in Becoming an EN and Loss of Funding and Competition with SVRAs ...... 98 Table 4.16 Chi-Square Test for Interest in Becoming an EN and Tracking Pay Stubs 105 Table 4.17 Interest in Being an EN and Tracking Pay Stubs as a Barrier...... 105 Table 4.18 Chi-Squared Test for Interest in Becoming an EN and Evaluation of ENs 106 Table 4.19 Interest in Becoming an EN and Evaluation of ENs...... 107 Table 4.20 Chi-Square Test for Interest in Becoming an EN and Need for Upfront Funding ...... 117 Table 4.21 Interest in Becoming an EN and Need for Upfront Funding ...... 117 Table 4.22 Chi-Square Test for Interest in Being an EN and Total Number of Staff...118 Table 4.23 Interest in Being an EN and Total Number of Staff ...... 119 Table 4.24 Frequency Table for Sources of Information About the Ticket Program .121 Table 4.25 Frequency Table for Overall Knowledge of the Ticket Program...... 122 Table 4.26 Chi-Square Test for Interest in Being an EN and Overall Knowledge of the Ticket Program...... 123 Table 4.27 Interest in Being an EN and Overall Knowledge of the Ticket Program.....123 Table 4.28 Frequency Table for Knowledge about Payment Options ...... 124

Table 4.29 Chi-Square Test for Interest in Being an EN and Knowledge about Payment Options...... 125 Table 4.30 Interest in Becoming an EN and Knowledge About Payment Options ...... 125 Table 4.31 Frequency Table for Beneficiary Status Reporting Requirements...... 126 Table 4.32 Frequency Table for Knowledge of Record Keeping Requirements...... 127 Table 4.33 Frequency table showing overall knowledge about requirements to become an EN...... 128 Table 4.34 Frequency table showing percentage of annual referrals...... 128 from SVRAs and the number of providers in each group...... 128 Table G.1 Frequency Table Worksite Locations ...... 179 Table G.2 Chi-Square Test of Independence for Worksite Location...... 180 Number of Services...... 180 Table G.3 Measures of central tendency for number ...... 180 of services with interquartile ranges...... 180 Table G.4 Frequency table for number of services offered...... 181 Table G.5 Contingency table for chi-square for interest in being an EN and number of services offered...... 181 Table G.6 Chi square test of independence for number ...... 181 of services and interest in becoming an EN...... 181 Table G.7 Correlation matrix for payment system and issues...... 182 Table G.8 Payment system and issues KMO and ...... 182 Bartlett’s Test of Sphericity...... 182 Table G.10 Varimax rotation of principal components for payment system ...... 184 and issues...... 184 Table G.11 Correlation Matrix for payment system factor analysis ...... 184 Table G.12 Payment system and issues KMO and Bartlett’s Test ...... 185 of Sphericity...... 185 Table G.13 Total variance explained for payment system factor analysis...... 185 Table G.14 Unrotated component matrix for payment system...... 186 Table G.15 Regression factor score for payment system...... 187 Table G.16 Correlation matrix for payment issues factor analysis...... 187 Table G.17 KMO and Bartlett’s Test of Sphericity...... 187 Table G.18 Total variance explained for payment issues/options factor analysis ...... 188 Table G.19 Component Matrix for payment issues ...... 189 Table G.20 Rotated component matrix...... 189 Table G.21 Regression factor score for payment...... 190 issues/options including mean and median values...... 190 Table G.22 Correlation matrix for EN reporting requirements factor analysis ...... 190 Table G.23 KMO and Bartlett’s Test of Sphericity ...... 191 Table G.24 Total variance explained for EN reporting requirements factor analysis....191 Table G.25 Component matrix for EN reporting requirements...... 192 Table G.26 Rotated component matrix...... 193 Table G.27 Correlation matrix for relationship between SVRAs and ENs...... 193 Table G.29 Total variance explained for SVRA and competition factor analysis...... 194 Table G.30 Unrotated component matrix for Competition ...... 195

xvi

Table G.31 Regression factor score for SVRA-EN ...... 196 factor analysis, separated into equal appearing thirds...... 196 Table G.32 Cramer’s V statistic for interest in being an EN and ...... 196 need for upfront funding...... 196 Table G.33 Frequency table for position held by respondents ...... 197 Table G.34 Frequency table for type of organization ...... 197 Table G.35 Frequency table for number of services provided...... 198

xvii

LIST OF FIGURES

Figure 4.1 Interest in Becoming an EN...... 60 Figure 4.2 Number of Direct Services Offered...... 64 Figure 4.3 Structure of the Payment System ...... 71 Figure 4.4 Structure of the Payment System ...... 71 Figure 4.5 Payment Structure is Clear and Easy to Understand ...... 71 Figure 4.6 ENs Are Adequately Reimbursed for SSI Beneficiaries ...... 72 Figure 4.7 ENs Are Adequately Reimbursed for SSDI Beneficiaries ...... 72 Figure 4.8 Not Participated because Calculation of Payment Options is Unclear.. 73 Figure 4.9 Understand the Difference between Milestone/Outcome Payments..... 74 Figure 4.10 Understand How Payments are Calculated ...... 75 Figure 4.11 Type of Payment Option that Would Encourage Participation as EN . 76 Figure 4.12 Providers Should be Reimbursed for RTW at Less Than SGA...... 80 Figure 4.13 Not Participating Because Many Clients Never Reach SGA...... 81 Figure 4.14 Payments Should be Based on Sliding Scale not 0 Cash Benefits...... 82 Figure 4.15 Ticket Program Not Useful Many Clients in Supported Employment.. 83 Figure 4.16 Differences in Amount of Payment for SSI/SSDI Prevent Serving SSI ...... 84 Figure 4.17 Payments Should be the Same Regardless of SSI/SSDI...... 84 Figure 4.18 ENs are Not Adequately Reimbursed for SSI Clients...... 85 Figure 4.19 Serving Ticket Holders Need Long Term Supports is a Barrier ...... 86 Figure 4.20 Difference in Payment is Financial and Service Delivery Dilemma.... 87 Figure 4.21 Possibility of Overpayment is Issue ...... 88 Figure 4.22 ENs Should be Reimbursed on Cost Reimbursement Basis Like SVRA ...... 89 Figure 4.23 Agree Payment for Services Should be Shared...... 90 Figure 4.24 Waiting for 45 Days or More for Reimbursement is Hardship...... 91 Figure 4.25 Not Participated Because Concerned Re: Length of Time for Payment ...... 91 Figure 4.26 Not Participated Because Concerned if More than 60 Days for Reimbursement ...... 92 Figure 4.27 Concerned about Competing for Clients with State Agency ...... 95 Figure 4.28 Competing for Ticket Clients May Result in Losing SVRA Funding . 96 Figure 4.29 SVRA has Required Ticket Must be Deposited with State Agency .... 99 Figure 4.30 Not Participating Because Will Lose Other Funding ...... 100

xviii Figure 4.31 Not Participated Because Staff Do Not Have Credentials to Serve Ticket Holders...... 101 Figure 4.32 Organization Employs Staff Who are Qualified to Serve Ticket Holders...... 101 Figure 4.33 Not Participated Because Application Cumbersome/Time Consuming ...... 102 Figure 4.34 Collecting Pay Stubs for 60 Months is Barrier to Participation...... 103 Figure 4.35 Not Participated Because Difficult to Track Pay Stubs...... 104 Figure 4.36 Not Participated Because No Computer Tracking Program ...... 108 Figure 4.37 Would Participate if Could Use Software Program for Tracking Participants...... 109 Figure 4.38 Not Participated Because No Program to Track Financial Records..... 109 Figure 4.39 Organization Can Provide Service to Any Ticket Holder Who Requests It ...... 110 Figure 4.40 Not Participated Because Do Not Understand Program...... 111 Figure 4.41 Criteria for EN Evaluation: Placement, Level of Work, Population Served ...... 112 Figure 4.42 Criteria for Evaluating ENs Should be Number of Placements ...... 113 Figure 4.43 Dispute Resolution Process is Unfair to ENs...... 114 Figure 4.44 Third Party Should be Responsible for Dispute Resolution ...... 114 Figure 4.45 Upfront Funds Enable Participation as EN...... 115 Figure 4.46 Become an EN if Low Interest Loans Were Available ...... 116 Figure 4.47 Would Participate if SSA Implemented Referral Process ...... 129 Figure 4.48 Most Preferred Method for Ticket Holder Referrals...... 130 Figure 4.49 Important Information at Time of Referral...... 131 Figure 4.50 SSA Should Provide Time-Limited Benefits as Incentive/Bridge to RTW ...... 132 Figure 4.51 Provide Ticket at Time of Application of Benefits if Presumed Eligible ...... 133 Figure 4.52 SSA Should Refer Beneficiary for RTW with Ticket Application ...... 133 Figure 4.53 Ticket Program Should Be Mandatory for All Disabled SSA Beneficiaries ...... 134 Figure 4.54 Do Not Have Ability to Market to Ticket Holders...... 135 Figure 4.55 Need Help Identifying and Marketing to Ticket Holders ...... 136 Figure 4.56 Need Help Developing Individual Work Plan ...... 136 Figure 4.57 Support Services Offered...... 137 Figure 4.58 Transportation Services Available ...... 138 Figure G.1 Scree Plot for Payment Systems and Issues...... 181 Figure G.2 Scree Plot for Payment System PCA...... 185 Figure G.3 Scree Plot for Payment Issues/Options PCA...... 187 Figure G.4 Scree Plot for EN Reporting Requirements PCA...... 191 Figure G.5 Scree Plot for Competition and Loss of Funding PCA...... 194

xix CHAPTER 1

1. INTRODUCTION

1.1 Statement of the Problem

The Social Security disability rolls have experienced substantial growth in the past ten years. According to the Social Security Administration (SSA), the number of beneficiaries receiving Supplemental Security Income (SSI) and Social Security

Disability Insurance (SSDI) has increased 65% in the last ten years (Bovbjerg, 2001). As of 2000, an estimated 3.8 million people received benefits based on disability from SSI, and another 4.9 million received benefits from SSDI (SSA, 2001). At the present rate, the Government Accountability Office (GAO) predicted the insolvency of the Social

Security Disability Trust Fund by 2023 (Bovbjerg, 2001). In response to this concern, the SSA and Congress implemented a variety of measures and incentives over the past

10-15 years to return SSI and SSDI beneficiaries to work. Unfortunately, these incentives and programs have not reduced the number of SSA beneficiaries on the disability rolls.

The SSA has long been concerned with the increase in the number of the beneficiaries, but this concern has escalated over the years as the percentage of applicants who have been awarded benefits has increased. More importantly, the SSA is concerned with the fact that people are not returning to work after they have started receiving benefits. A GAO study conducted in the 1990’s revealed that, under the current system of

1 rehabilitation, less than one-half of 1% of the beneficiaries returns to work. In fact, this study documented that more beneficiaries leave the disability rolls because of retirement or death than because of a return to work (Fallavollita & Bordelon, 1997).

The state vocational rehabilitation agencies (SVRAs), which have been generally viewed as the primary providers of vocational services to SSA beneficiaries, have a return to work rate of fewer than 1 in 500 for SSDI beneficiaries. The return to work numbers are even less for SSI beneficiaries (Fallavollita & Bordelon, 1997; Tenney &

McCray, 1997). Consequently, SSA began to look toward the private sector as an alternative to the traditional vocational rehabilitation offered by SVRA. The SSA was attracted to the prospect of enhancing the return to work rate of disabled beneficiaries based on cost-effective, timely, and outcome-oriented measures (Forgiel & Growick,

1997; Growick, 2000b).

Consequently, SSA initiated Project NetWork in 1991 and the Alternative

Providers program in 1995. Private practitioners, as well as the traditional non-profit organizations such as Goodwill Industries and Easter Seals, were approved to deliver rehabilitation services to SSA beneficiaries. The implementation of these demonstration projects and the passage of the Ticket to Work-Work Incentives Improvement Act

(TWWIIA) legislation reflected SSA’s concern that an alternate means of rehabilitation was needed (Growick, 2000a).

1.1.1 Project NetWork

The SSA commissioned Project NetWork in the early 1990’s. It was initiated by the SSA to examine the effectiveness of private sector rehabilitation as a means of encouraging and facilitating a return to work for people with disabilities who were

receiving SSA disability benefits (Fallavollita & Bordelon, 1997). This program, which specifically targeted SSI and SSDI beneficiaries, offered intensive outreach, work incentive waivers, and case management and referral services. Several findings of this project are particularly pertinent to the payment issues of the Ticket to Work (Ticket) program. In spite of an overall 11% increase in earnings, no substantial decline in receipt of SSI and SSDI benefits resulted. In fact, 60% of the workers in this program were earning wages at $6.50/hour or less, which is a level of wages that never would have allowed these workers to move off the disability rolls to self-sufficiency, even in the

1990’s (Kornfield & Rupp, 2000).

The Project NetWork sites, which were in Arizona and Nevada, found that beneficiaries who had been out of the work force longer required more services, such as retraining, to re-enter the work force. They noted that SSDI beneficiaries may require more services to qualify them for higher paying jobs in order to encourage them to leave the benefit rolls. Additionally, SSI recipients with little work history had greater fear regarding the viability of working (Tenney & McCray, 1997). Across the eight sites in this demonstration project, researchers found that of the 150,000 eligible beneficiaries who were contacted by mail, only 5% chose to participate in the program with an average expenditure of $3,660 per beneficiary (Kornfield & Rupp, 2000). However, 100 beneficiaries were employed by the end of the project, at a potential savings of $647,000 to the Disability Trust Fund, if they remained employed (Tenney & McCray, 1997).

Three findings from this study were particularly cogent, and have direct implications for the current Ticket program. The first is that the costs to serve beneficiaries were based on the needs of the individual. Although costs were higher to

serve some clients, Project NetWork case managers (CMs) were reimbursed based on their proposed pricing structures, which were negotiated at the time of application to participate as a CM. The payment structure was not prescribed by SSA, but was market driven. The second factor was that a great deal of up front information, training and support was provided to organizations that agreed to participate. This level of support facilitated recruitment and is not presently available to prospective ENs in the current

Ticket program. The final factor is that the project provided for the assessment of employment potential by the CM prior to providing services. Under the present Ticket program, reimbursement for initial employability screenings is not included in the payment structure. Testimony at the Ticket Advisory Panel meetings and anecdotal information from current ENs have cited the extensive cost of upfront screening of Ticket holders as a major financial concern for the organizations.

1.1.2 The Alternative Providers Program

Although Project NetWork resulted in limited success, the SSA believed the results were promising and the rehabilitation efforts of the private sector could be beneficial in SSA’s return to work efforts for their beneficiaries. Consequently they established the Alternative Providers (APs) program in 1995. The SSA initially contacted private providers via a survey implemented by the Gallup organization. If providers indicated an interest in the program when they responded to the survey, they were enrolled as APs in the program. Over 300 providers were enrolled across the country.

Unfortunately, only one formal survey of APs was conducted. In 1999, Marini and Stebnicki undertook a survey to determine the experience of the APs with the

program. The investigators reported some barriers to success. The APs cited the following issues as barriers to their participation: 1) reimbursement schedules; 2) need for medical information about the claimant prior to contacting the individual to begin services; 3) improved communication between SSA and the APs; and 4) the need for time-limited benefits with mandatory participation in the program for disability beneficiaries. Also, the Bulletin Board method of notifying the APs of beneficiaries in their area was inadequate. Often the list was not regularly up-dated and available, making marketing to and contacting these individuals prohibitive (Marini & Stebnicki, 1999).

This study documented that in the four years from 1995-1999, only five beneficiaries returned to gainful employment out of over 1,100 individuals contacted by the 347 APs. The open-ended comments from the APs were of particular concern. This information, collected by the authors, documented the most frequently identified barriers to success: the length of time without reimbursement, slow payments from the SSA, and the cost of up-front services. At the time of the survey, two-thirds of the providers indicated they either were considering, or were dropping out of the AP program (Marini

& Reid, 2001). Although some of these issues were addressed in the initial Ticket program regulations, anecdotal evidence from potential ENs who are currently private providers continue to mirror these concerns.

Consequently, the SSA convened a group of subject matter experts to review these concerns and to determine effective means to overcome some of the barriers to participation in the Ticket program. The Adequacy of Incentives Advisory Group

(AOI) was brought together in 2003 to make recommendations to the SSA for changes in regulations or needed changes in legislation to facilitate the effectiveness of the Ticket

program. The AOI group specifically evaluated EN payments options and issues, and

payment structures. They subsequently recommended changes to the Ticket program that

are addressed in a later portion of this document.

1.1.3 Ticket-to-Work Legislation

In 1999, Congress passed the Ticket to Work-Work Incentives Improvement Act.

In this legislation, Congress formalized the Alternative Providers program by establishing the Employment Networks. Although recruitment of ENs has been ongoing, a relatively small number of ENs are currently enrolled (slightly more than 1300) and some of the same problems/barriers continue to exist. Among these concerns, significant barriers remain unresolved: the provider payment schedule, timely reimbursements, lack of up front funding, and the reporting and accounting documentation required by ENs

(Growick, 2000b). The Social Security Advisory Board echoed these concerns in their preliminary report on the adequacy of incentives for the ENs. In this report, their design committee encouraged the study of perceived incentives to ENs, especially as it relates to the harder to serve populations (Social Security Advisory Board, 2001). More recently, both the Ticket Advisory Panel (Ticket to Work and Work Incentives Advisory Panel,

2003) and the AOI Group (Adequacy of Incentives Advisory Group, 2003) issued reports in which similar issues were raised and recommendations were made for revising the payment system for ENs and modifying some reporting requirements.

1.2 Importance of the Study

In order for the Ticket program to be successful in returning disabled SSA

beneficiaries to work, current and potential barriers to implementation for prospective

ENs must be identified, evaluated and remedied. Additionally, the ENs who have chosen

to participate need to remain involved in the program. By increasing the number and

range of ENs, consumer choice will increase, and more SSA beneficiaries will return to

work (Growick, 2000a). The final step in the supply side of the Ticket program is that

the prospective ENs must be identified and surveyed to determine their knowledge of,

and interest in, the program. The success of the Ticket program is not only dependent

upon beneficiary participation, but also on a willing and open labor market, as

characterized by EN participation.

Efforts are already underway to address barriers to success on the “demand

(beneficiaries) side” through the implementation of the Benefits Planning and Assistance

Outreach Program (BPAO). This research project on the supply side of the Ticket

program proposed to assist SSA in maximizing the recruitment and maintenance of

potential ENs by soliciting the providers’ opinions regarding barriers to the success of the

Ticket program in order to determine what changes, if any, may encourage increased

participation.

1.2.1 Issues for ENs

The researcher proposed to survey potential ENs to identify which of the issues that are assumed to be of primary importance are actually potential barriers to participation in the Ticket program. These issues are listed below, and are discussed in further detail in this section.

• Payment system • Payment options/issues • Delays in reimbursement • Relationship between ENs and the SVRAs • EN credentials/requirements • EN reporting requirements • Criteria for the evaluation of ENs 7

• Upfront funding/Capitalization

The researcher used the survey to explore three additional areas related to the potential ENs. They are 1) the ENs’ need for technical assistance and training; 2) their opinions regarding whether the Ticket program could be integrated into the current SSA disability determination process, and how that might be structured; and 3) what effect, if any, the geographic location, size of their organization (number of staff), or number of services they provided, impacted their decision to not participate in the Ticket program.

These issues were identified from the following sources: 1) literature review; 2) professional experiences of the researcher; 3) input of an expert panel; and 4) anecdotal information collected informally from prospective ENs around the country as the Ticket program has been implemented.

Payment System

From the inception of Ticket program, the adequacies and complexity of the payment system, and the basis for the calculation of payments have been areas of ongoing controversy. Prospective ENs were asked to comment on the nature and structure of the two methods of payments, the outcome and milestone/outcome options.

The understanding of the payment design has generated many ongoing questions from potential ENs since the implementation of the program. Consequently, consideration was given to the following issues: Is the structure of the payment options a barrier to participation in the Ticket program? If so, what payment structure would increase participation in the Ticket program? Do prospective ENs understand the payment options or does the complex nature of this structure prevent potential ENs from participating in the program? (See the survey instrument in Appendix H for survey

questions 1, 3, 4, 5, 6, and 13, 15, 16, 23, 28 about payment options and program

structure.)

Payment Options/Issues

Multiple payment issues have been identified. The current payment system does

not allow ENs to collect reimbursement until the beneficiary is receiving no cash

benefits. Additionally, the adequacy of the actual payment options in terms of SSI and

SSDI beneficiaries has raised concerns. A series of questions in the survey addressed the

relationship between payment inequities, and payment options, and the willingness of

rehabilitation providers to become ENs.

Use of Substantial Gainful Activity as a Criterion for Payment of Services

Based upon the current rules and regulations, payments to ENs become available only when the beneficiary attains earnings at the level of substantial gainful activity

(SGA)1 and moves off cash benefits. Payments for serving SSDI or SSI beneficiaries

vary because many of the harder to serve population are engaged in supported

employment, which typically does not reach the SGA level. Questions related to this

issue included: Should ENs collect payment for employment services provided to people

with disabilities who are working, but have not reached SGA? What incentives are

necessary for ENs to increase their willingness to serve the harder to serve clients?

Would payment for services that result in work at less than SGA encourage EN

participation? (See the survey instrument in Appendix H for survey questions 11, 12, 14,

and 36 about SGA.)

1 Earnings equal to or above $830/month are the 2005 current level of SGA

Inequities in Payments for Serving SSI and SSDI Beneficiaries

A difference exists between the cash benefits paid to SSI beneficiaries and SSDI beneficiaries. Individuals on SSDI receive cash benefits that are based on their earnings over the past forty quarters, while SSI beneficiaries typically have little or no work experience. Consequently, the SSI beneficiaries receive the amount set by Congress, which is presently approximately $538/month. The difference in the beneficiaries’ cash benefits impacts the level of payments to ENs for serving them. Under the current payment design for the Ticket program, reimbursements for returning SSDI beneficiaries to work is often much higher than for SSI beneficiaries. The impact of this difference was explored to determine to whether it may be a barrier to participation, and to what extent it affects the provision of services to those beneficiaries in the harder to serve categories. (See the survey instrument in Appendix H for survey questions 2, 7, 8, 9, 24,

32, and 40 about SSI/SSDI inequities.

Overpayments

When SSA beneficiaries return to work, within the present SSA system, overpayments continue to be a major problem. The complexity of the rules and regulations of the SSI and SSDI programs makes the processing of earnings information difficult. Recently, SSA found new evidence of overpayments to SSI beneficiaries totaling nearly $2 billion in 2001 (SSA, 2003a). The numbers are harder to determine for SSDI beneficiaries. However, one government source reported that SSA’s inability to process earnings information in a timely manner might account for approximately $350 million in SSDI overpayments in 2001 (GAO, 2002a). Since 1997, the SSI program has been designated as high risk because over $1 billion has been paid in benefits

overpayments (SSA, 2003a). If SSA determines a beneficiary has been overpaid, a notice is sent to the beneficiary indicating the amount of the overpayment and requesting reimbursement to SSA within 30 days. The beneficiary has the option of appealing the decision and requesting reconsideration. However, there is considerable cost to the individual in terms of psychological distress and future financial obligations if the overpayment decision is upheld.

A senior staff attorney at the Disability Law Center in Boston, Massachusetts testified before the Ticket Advisory Panel that experience has shown that beneficiaries who receive overpayment notices were most likely to discontinue their work efforts

(Cebula, 2003). He also noted that at a recent focus group at the Disability Law Center, the Ticket program was referred to as the “ticket to overpayments.” He reported that under the current regulations related to overpayments, a beneficiary who has begun a return to work effort may face wage garnishment by the very same agency (SSA) that is encouraging returning to work.

According to testimony from a SSA union representative provided to the Ticket

Advisory Panel (Fehner, 2003), the field staff are already overwhelmed and understaffed in meeting the demands of their normal workload. Another SSA union representative reiterated this belief. She testified that increasing evidence indicated that the workload was much higher than originally anticipated, and will continue to grow as the Ticket program provisions are used by beneficiaries. The number of potential ticket users who require additional processing adds to their burden and raises the possibility that under the

Ticket program overpayments will occur (Einschlag, 2003). When overpayments do occur, beneficiaries remain on the rolls and ENs are unable to collect reimbursement for

services, even though the beneficiary is employed. The impact of overpayments for providers was reflected in the following question in the survey. Does the prospect of overpayments prevent prospective ENs from participating because overpayments impact the reimbursement for services? (See the survey instrument in Appendix H for survey question 19 about overpayments.)

Sharing Payments and Method of Payments.

The difference between the payment structure for providers and SVRAs has been a source of controversy with current ENs. Based on anecdotal information, it was assumed that this concern might also exist for prospective ENs. Under the Ticket program, the SVRAs presently have two funding options available to serve Ticket holders. One option is cost reimbursement, which allows the SVRAs to recoup the actual dollars spent for services or rehabilitation technology. The second option is the payment structure of the Ticket program, which is bifurcated in two alternatives: outcome or milestone-outcome method of payments. Under the Ticket program’s present payment structure, cost reimbursement is not available for providers. Consequently, many prospective ENs are unable to offer extensive supports or services to ticket holders because there is a lack of funding outside of state agency dollars for the “harder to serve population”. For these providers, often the only option is to refer the beneficiary to the

SVRA so the appropriate services and supports can be offered. This defeats the purpose of the Ticket program because it was conceived to increase consumer choice, not reinforce the same system of provision of rehabilitation services.

Anecdotal information from providers has revealed that many are reluctant to refer a ticket holder to SVRA. They believe referring the ticket holder to the SVRA may

result in the ticket holder depositing the ticket with the SVRA and the potential of sharing the payment for the services. They also believe that under the Ticket program, the provider may have to pay the SVRA for services provided, in advance of receiving payment from SSA for the ticket holder returning to work. These issues were addressed in the survey by asking the following questions. Do they agree they should share payment for services if they have referred the beneficiary to the SVRA or another agency? Do prospective ENs believe they should be allowed reimbursement for services on a cost reimbursement basis like the SVRA? (See the survey instrument in Appendix H for survey questions 23 and 28.)

Delays in Reimbursement

Because of their ability to “float” funding, it was anticipated that larger private sector providers would be more likely to become ENs. Larger providers typically have multiple sources of funding that might allow them to serve ticket holders in the event that reimbursements were slow. Due to a history of slow processing of paperwork, many providers assume that it will take longer than 30 days to be reimbursed after the paperwork has been submitted. This is a major area of concern for the Ticket program because it will limit the choice of providers. Smaller “mom and pop” organizations and rural programs with less access to funding may be negatively impacted. This also impacts the availability of services to beneficiaries. Problems with delays in reimbursements have raised many questions. To what extent will anticipated delays in payments prevent a broader range of potential ENs from participation or continuing with the Ticket program? Is the potential of delays in reimbursement a barrier to participating

in the program? (See the survey instrument in Appendix H for survey questions 15, 16, and 34 about delays in reimbursements.)

Relationship between the State-Federal Vocational Rehabilitation System and ENs

Under the regulations, when two ENs request payment for the same milestone, or same outcome payment month, the payment is allocated based upon the services provided by each EN (Maximus, 2002). Under this system, payment to the SVRA under the traditional cost reimbursement method in reference to a ticket precludes any subsequent payment to another EN. Additionally, anecdotal information was collected by the researcher regarding the agreements state vocational rehabilitation agencies (SVRAs) have developed and required ENs to sign. In some areas, this agreement effectively ended consumer choice, and was developed without input from the providers (Ticket to

Work and Work Incentives Advisory Panel, 2003).

Recently the Ticket Advisory Panel expressed concern regarding the agreements between SVRAs and ENs (Ticket to Work and Work Incentives Advisory Panel, 2003).

Specifically, Transmittal 17 has received a great deal of attention due to two features of the document. Under SSA’s Transmittal 17, guidance is provided to state agencies regarding the assignment of a ticket. According to this document, the client’s signature on the Individual Plan for Employment (IPE) is considered to be the assignment of the ticket. The assignment of the ticket in this manner can take place without the knowledge and expressed consent of the beneficiary, and raises concerns regarding uninformed consent. The second feature is that Transmittal 17 indicates that SVRAs may use cost reimbursement as a payment option for ticket holders who have assigned their ticket to the state agency. Having this option available to SVRAs and not ENs has created an

inequity in the reimbursement for services for Ticket holders. In addition to Transmittal

17, the Ticket Advisory Panel also voiced concern regarding problems with cooperative agreements between SVRAs and ENs that are unfavorable. The Lewin Group noted that there are four types of agreements, which are problematic for current and prospective

ENs (Lewin Group, 2003):

• Agreements that require ENs to reimburse SVRAs for all vocational rehabilitation

(VR) costs to serve a ticket holder who has assigned the ticket to a non-SVRA

EN.

• Agreements that require all tickets in the SVRA/EN partnership to be assigned to

the SVRA.

• Agreements that require ENs to share income with SVRAs over and above the

VR cost to serve a ticket holder.

• Agreements that require ENs to repay the SVRAs for the costs of services

provided before ENs have received any payments through the Ticket program.

The issues related to the SVRA and EN agreements have generated many questions. Have the SVRAs in the states that were surveyed required that a ticket be deposited with the agency? Is there concern that becoming an EN will result in competition with the SVRAs and potentially jeopardize referrals and revenue from the

SVRA for other clients who may require longer-term supports? Additionally, is there concern about the loss of other sources of funding, such as Medicaid. (See the survey instrument in Appendix H for survey questions 20, 21, 26, and 49 about the relationship between SVRAs and ENs.)

Employment Network Credentials/Requirements

The survey also explored the current EN program requirements that may be potential barriers to participation. The EN application describes the staff credentials, which are necessary for EN participation. The potential that these credentials may be barriers to participation for traditional and non-traditional providers raised many questions. Do the current credentialing requirements for staff prohibit the participation of otherwise appropriate rehabilitation providers or non-traditional providers such as employers or temporary agencies? Do the current credentialing requirements for agencies/companies prohibit participation in the Ticket program? Is the application for becoming an EN time-consuming and overwhelming? Is the application a barrier to participation in the program? (See the survey instrument in Appendix H for survey questions 22, 25, 27, and 39 about EN credentials.)

Employment Network Reporting Requirements

For small and large rehabilitation providers, the reporting requirements as promulgated in the rules and regulations of the Ticket program are a burden, especially when some providers may lack support staff and automated systems for case management and accounting. Requirements for providing employment verification for 60 months essentially precluded smaller providers from participation. Prospective ENs were given the opportunity to voice their opinions and suggestions regarding the reporting requirements and structures. Issues addressed include the following. Does the structure of the reporting system regarding the tracking and implementation of tickets present a barrier to participation in the program? Is requirement to provide verification of employment for 60 months a barrier to participation? Would case management software

specifically designed for the Ticket program be effective in increasing EN participation?

Do potential ENs have the financial accounting systems necessary to accurately track

Ticket program revenue and billing? Would software specifically designed for this purpose facilitate the reporting process and encourage EN participation? (See the survey instrument in Appendix H for survey questions 29, 30, 31, 33, and 50 about EN reporting requirements.)

The Criteria for the Evaluation of ENs

Under the Ticket program, SSA will periodically evaluate the performance of an

EN for quality assurance purposes. The SSA’s evaluation process is believed to be a deterrent to the participation of some prospective ENs. This is especially true because this process has not been clearly articulated, even though the Ticket program is now in its fourth year of implementation. Prospective ENs have voiced concerns regarding the potential that exists for SSA’s comparison of programs to be across disability groups, which may not take into consideration the population served, and the types of placements based on the needs of the individual. This may result in criticism of programs that work with the harder to serve populations. Consequently, several issues related to EN evaluation were explored. Additionally, concerns about the dispute resolution process were addressed. Questions generated about these issues included: What criteria should be used to evaluate ENs? How should employment outcomes be evaluated to encourage increased EN participation? Should types of placements and levels of placements be part of the evaluation criteria based on the population served? Who should be responsible for resolving disputes between ENs or ENs and beneficiaries? Is the dispute resolution

process a barrier to participation? (See the survey instrument in Appendix H for survey questions 37, 38, 43, and 46 about criteria for evaluating ENs.)

Upfront Funding/ Capitalization

Much of the earlier writings regarding the Ticket program addressed concerns related to the initial costs for participation. Because one of the goals of the Ticket program was to expand the range of employment service providers who are available to serve SSA beneficiaries, large start up costs were anticipated to be a significant barrier to participation for smaller providers. This research study explored this issue through a poignant question related to the prospective ENs’ need for capitalization and start up funding. (See the survey instrument in Appendix H for survey question 10 about upfront funding and capitalization.)

1.2.2 Other Issues

Policy Changes

The opinions of prospective ENs were also solicited about several critical disability policy changes. The first aspect of policy changes addressed ways in which employment services could be integrated earlier into the Social Security disability determination process. Creating an expectation of working from the time of application is an important concept for the successful implementation of the Ticket program for two primary reasons. First, it may increase the number of beneficiaries who participate in services leading to employment because they do not have the opportunity to adjust to a disabled lifestyle. Second, beneficiaries would not have to prove they are permanently and totally disabled, only that they are unable to work at the present time.

Another area that was explored related to the difference between the reimbursements for services for employment of SSI versus SSDI beneficiaries. Finally, prospective ENs were also asked their opinions regarding improvements that could be made to SSA disability policies that would facilitate increased provider and beneficiary participation in the Ticket program. What type of payment option or changes to the payment schedule would increase the willingness of potential ENs to serve SSI beneficiaries? Should applicants with disabilities be given the opportunity to participate in the Ticket program at the time of application? Should the Ticket program be mandatory rather than voluntary for beneficiaries? Should SSA implement a process to refer beneficiaries directly to ENs? Should SSA provide beneficiaries with time-limited benefits as a bridge to return to work and encourage employment? (See the survey instrument in Appendix H for survey questions 17, 18, 41, 45, and 47 about policy changes.)

Need for Technical Assistance and Training

Testimony before the Ticket Advisory Panel has addressed the need for technical assistance and training for prospective ENs (Harles, 2003; Webb, 2003). Some rehabilitation professionals have taken the position that prospective ENs are not participating because they do not have the expertise necessary to market to SSA beneficiaries, and may not have the necessary experience to develop individual work plans or individual plans for employment (IPEs). The researcher included questions in the survey to measure the perceived need for support and training in these areas. (See the survey instrument in Appendix H for survey questions 42, 44, and 48 about the need for technical assistance and training.)

Work Setting, Number of Staff, Number of Clients Served and Non-Participation

One of the major areas of concern, according to SSA, is representation of a range of providers of services to ticket holders (Growick and Drew, 2003). This survey explored whether geographic locale, work setting, size of the organization, or number of services provided had any influence on participation in the Ticket program.

1.3 Design of the Study

1.3.1 Overview to Population

The survey population was identified by combining two databases of known potential ENs from six of the initial thirteen roll out states for the Ticket program. The six states were chosen on the basis of geographic diversity. They included Arizona and

Colorado representing the West; Oklahoma and Wisconsin representing the Midwest, and

Florida and New York representing the East. The geographic distribution of these states enabled the researcher to identify providers who represented urban, suburban and rural programs, in addition to representing a range of size of businesses from large, national organizations to sole proprietors.

A comprehensive mailing list of organizations located in the six states was compiled. This list was developed by combining the community rehabilitation providers

(CRP) database from the University of Wisconsin-Stout with the membership list from

International Association of Rehabilitation Providers (IARP). For the initial survey phase, this combined list was cross-referenced against the list of providers who had become ENs as of June 30, 2003. The known population of providers that met the definition of a CRP (according to the Stout project) or were IARP members as of June

2003 was included. Duplicate organizations, providers who were already ENs as of June

30, 2003, and national providers (all of whom were ENs) were also removed from the

population.

1.3.2 Study Design

The design of this study was non-experimental, descriptive research (Campbell &

Stanley,1963). Determination of the sample size and design of the survey for this phase

was accomplished through consultation with the Ohio State University’s Center for

Survey Research(CSR). CSR’s role in this process encompassed the following areas:

a) to provide expertise in survey design related to skip patterns within

surveys;

b) to provide consultation related to how response rates were to be measured;

c) to provide consultation regarding wording of questions and a priori

assumptions;

d) and to provide consultation with the researcher regarding sampling issues,

sample size, and sample size effect.

A decision was made to conduct a census because it was both affordable and

feasible due to the relatively small size of the population (n=1,732) for the initial phase of

the survey. Taking a census was the optimal sampling method because it eliminated the

potential for sampling variance within the six selected states since very little was known

about the characteristics of the organizations that were potential survey participants.

1.4 Assumptions of the Study

Little was known about the characteristics of the population of potential ENs

included in this study. Consequently, parametric statistical procedures were not

appropriate for data analysis. Instead, a non-parametric statistic, the chi-square test for

independence was used to analyze the data in this study. The researcher wanted to

determine if a relationship existed between categorical variables. The chi-square statistic

is the most appropriate measure because it assesses whether random chance might

produce differences that are equivalent to the differences between the data that has been

gathered (the observed frequency, fo) and the distribution that should occur theoretically

(expected frequency, fe). This statistic was used to analyze the data to determine if differences existed between the respondents’ actual (or observed) responses to the categorical variables and the responses that were expected.

There were five assumptions made for the use of the chi-square test in this research analysis. These assumptions were the following:

1. The data to be analyzed should come from the census population with randomized

assignment to survey instruments.

2. Data to be used for chi-square must be restricted to counts of organizations, and

categorical variables, not scaled or continuous scores.

3. Given the n size of 100, when the df is greater than 1, average expected frequency

per cell can be as low as two (Hays, 1994; Hopkins, Hopkins & Glass, 1996)

4. The sum of the observed frequencies should equal the sum of the expected

frequencies.

5. No organization can appear in more than one cell.

1.5 Limitations of the Study

Over the past 30 years, Dillman has published many articles and books on the characteristics of a good survey. He noted that in spite of the researcher’s best intentions,

every study will have limitations. He summarized his thoughts on this topic in a 1991 article.

A good sample survey, by whatever method, is one in which all members of

a population have a known opportunity to be sampled for inclusion in the survey

(non-coverage error is avoided); the people to be surveyed are sampled by

random methods and in sufficiently large numbers to provide a desired level

of precision (sampling survey error is limited); questions are selected and phrased

in ways that result in people providing accurate information (measurement error

avoided); and everyone who is included in the sample responds (non-response

error is avoided). Seldom, if ever, does a sample survey accomplish all of these

goals (Dillman, 1991).

The researcher attempted to manage and address some of the errors typically associated with survey research as identified by Dillman in this quote. However, since no research project is perfect and without error, the following potential limitations to this study are offered and an explanation is given for what contingencies were implemented to attempt to control for these potential errors.

1.5.1 Sampling Error/ Non-coverage Error

The databases of non-profit and for profit organizations provided for this study encompassed a constantly changing population. These organizations, like any business, can go out of business, be acquired or bought out by other organizations, or become established during the course of the survey timeframe. However, extensive efforts were made to update the lists prior to the beginning of data collection, as will be explained in the methodology section.

In order to avoid sampling error, a census of the population of known rehabilitation providers in the six targeted states was used. As previously discussed, this decision was made since little was known about the characteristics of the population to be survey. In addition, it was determined that a random sample of non-respondents would be conducted, as well. In this survey, non-coverage error was anticipated because the census used was based on data provided by external organizations and was based upon known rehabilitation providers at the time that the data was provided. During the course of the project it was determined that both the database from the University of Wisconsin and the IARP database contained names and addresses of organizations or individuals who were out of business or could not be found.

1.5.2 Non-response Error

Non-response error can be simply described as the discrepancy between the non- respondents and the respondents. This type of error occurs when a significant number of potential respondents do not respond. The ability to determine if the characteristics of the non-respondents differ in some way from the respondents is helpful in determining non-response error. However, according to Dillman (1991), “ a low response rate does not necessarily entail non-response error…Those who respond to a survey may not differ in any measurable way from those who do respond” (p. 229). There was non-response error in this research since non-respondents did not respond in spite of multiple contacts over several months. Further details are provided in the methodology section.

Factors that may have increased the non-response rate for this survey included the following: 1) the databases used to establish the population; 2) technical problems with the College of Education’s server; 3) difficulty with web browsers; and 4) the apathy and

misinformation that is present among rehabilitation providers regarding the Ticket program. In addition, a bias existed in the response rate in relation to the use of the

Internet as a primary tool for data collection.

1.5.3 Technical problems/Difficulty with Web Browsers

There were technical problems with the College of Education server that could not be predicated when a hacker accessed the server. The server was subsequently shut down for the first four weeks of initial data collection. Many non-respondents that were contacted for follow-up indicated that they had attempted to respond or had responded prior to the server going down and were not going to complete the survey again. Over

100 potential respondents contacted the researcher via email or telephone calls to indicate they could not access the web-site. The researcher maintained a log of the emails and calls, and kept in regular contact with these organizations to advise them regarding the status of the server and to encourage participation. Only seven of those organizations eventually completed the survey, in spite of multiple follow-up contacts. According to the technology department, any data that was collected prior to the server crashing was lost and could not be recovered.

Additionally, it was anticipated that some respondents would have difficulty accessing the survey on-line due to their browser or out-dated computer processor issues.

In order for web-based surveys to be an effective means of data collection, Smith points out that the “target population must be technologically savvy enough to use it [the

Internet]” (Smith, 1997). In order to control for this potential error in response rate, the pre-notification letter advised respondents that an alternative format for the survey was available. The alternative format was an exact replica of the survey instrument as it

appeared on the web-site in order to control for survey bias due to a different instrument format.

1.5.4 Apathy and Misinformation

At the Ticket Advisory Panel’s November 2003 quarterly meeting, Maximus reported they made over 500,000 contacts with potential providers. According to their report, they found the response rate for recruitment for the program was one new EN for every 400 contacts (Maximus, 2003). At this time, approximately 1,300 providers have applied to become ENs, while fewer than 300 are actively taking Tickets. The results of the prospective EN survey and the analysis of open-ended comments made by respondents, demonstrated that apathy and misinformation might be profound regarding this program. If so, this negative attitude had the potential of impacting the response rate to the survey, and more importantly, the overall success of the Ticket program.

1.6 Research Questions/Hypotheses

Based on the literature reviewed, the personal experience of the researcher, anecdotal information gathered and feedback from a panel of experts, the following research questions were explored in this study. Each of these questions addressed specific Ticket to Work program components and the potential role they may play in an organization’s decision to not participate as an EN in the Ticket program. Research questions are followed by the statistical method used and the rationale for each method.

1.6.1 Research Hypotheses

1 H 0: There is no relationship between interest in becoming an EN and work setting.

Dichotomous nominal variable and multichotomous variables: chi-square, 2 X 5 contingency table

2 H 0: There is no relationship between geographic locales and interest in becoming an

EN. Multichotomous and dichotomous nominal variables: chi-square, 2 x 4 contingency table

3 H 0: There is no relationship between number of services provided and interest in becoming an EN. Ordinal variables treated as nominal variables and dichotomous nominal variables: chi-square, 2 x 4 contingency table

4 H 0: There is no relationship between attitudes toward the payment system and the decision not to participate as an EN. Dichotomous nominal variable, and ordinal variables that were treated as nominal variables: chi-square, 2 x 3 contingency table

5 H 0: There is no relationship between attitudes toward payment options/issues and the decision not to participate as an EN. Dichotomous nominal variable, and ordinal variables that were treated as nominal variables: chi-square, 2 x 3 contingency table

6 H 0: There is no relationship between attitudes toward delays in reimbursement and the decision not to participate as an EN. Dichotomous nominal variable, and ordinal variables that were treated as nominal variables: chi-square, 2 x 3 contingency table

7 H 0: There is no relationship between attitudes toward the contractual relationship between ENs and SVRAs and the decision not to participate as an EN. Dichotomous nominal variable, and ordinal variables that were treated as nominal variables: chi-square, 2 x 3 contingency table

8 H 0: There is no relationship between EN reporting requirements of pay stubs, or the evaluation of ENs and the decision not to participate as an EN. Dichotomous nominal variables and ordinal variables that were treated as categorical variables: chi-square, 2 x3 contingency table

9 H 0: There is no relationship between the lack of upfront funding and the decision not to participate as an EN. Dichotomous nominal variables and ordinal variables that were treated as categorical variables: chi-square, 2 x3 contingency table

10 H 0: There is no relationship between the number of staff and interest in becoming an

EN. Dichotomous nominal variable and ratio variable that was treated as a categorical variable: chi-square, 2 x 4 contingency table

1.6.2 Research Questions

1. What is the level of interest in becoming an EN? Dichotomous variable (yes/no):

Descriptive statistic, frequency count

2. What proportion of urban, rural, suburban, small town or inner city organizations are

interested in becoming an EN? Categorical variable: Descriptive statistic, frequency count

3. What is the level of knowledge about the Ticket program? Categorical variable:

Descriptive statistic, frequency count

CHAPTER 2

2. LITERATURE REVIEW

2.1 Overview to Employment as a Part of Disability Legislation

In order to fully appreciate the development of the Ticket to Work- Work

Incentives Improvement Act, it is important to understand it within the context of disability insurance legislation. As early as 1938, SSA planners saw the need for vocational retraining as a critical partner to disability insurance programs. Over the next

20 years, SSA officials unsuccessfully advocated for encouraging rehabilitation while providing cash benefits to disabled beneficiaries. In 1956, a disability insurance law was passed but without the financial support for return to work services. No financing was made available for rehabilitation services for beneficiaries until the passage of the

Beneficiaries Rehabilitation Program (BRP) in 1965.

Unlike other federal funds received by the SVRAs, the BRP funds did not require matching monies from the states. Initially the belief was that the program was working well based on anecdotal information. However, concerns began to be raised that the state-federal agencies were "creaming", that is to say, serving beneficiaries who would have been expected to leave the rolls due to medical improvement. The average cost of providing rehabilitation services to beneficiaries through the BRP program increased to twice the amount of the general vocational rehabilitation (VR) program. As a result,

29 1981, SSA replaced the BRP with a different payment structure. Under this design, SSA

would reimburse the SVRAs for services provided after the beneficiary returned to work

at earnings greater than $300/month, the level of SGA at that time. The beneficiary had

to remain employed for nine months at this level before the payment was issued. This

substantially decreased the monies paid by SSA because so few beneficiaries could meet

this standard (Berkowitz, 2003).

2.2 Overview to Ticket to Work and Work Incentives Improvement Act

The basic premise behind the Ticket program was straightforward. If some number of beneficiaries could be moved off the SSA disability rolls, the Disability Trust

Fund could realize savings. The design of the present program was simple, but the provision for payment to VR providers was complex, requiring a higher level of risk for the providers. Upon implementation of the program, beneficiaries would receive a ticket

(or voucher), which they would be under no obligation to use because beneficiary participation in this program is voluntary. Beneficiaries could deposit the ticket with a provider of choice who could also choose whether or not to accept the ticket. If VR providers returned beneficiaries to work, the providers would be paid a percentage of the benefits beneficiaries would have received if they had remained on the rolls. Payment to

VR providers would occur only after beneficiaries have returned to work and remained there for a specified period.

From the early conceptualization of the Ticket program, questions were raised as to whether the program would be profitable and cost-effective for potential providers of employment services. The common belief was that only large organizations would be able to participate due to the level of risk, which smaller providers would most likely be

unable to assume. During the early deliberations, providers indicated they preferred

some money up front to share the risk with SSA, rather than a payment schedule that was

based on outcomes only (Berkowitz, 2003).

Throughout the development of the program, specific issues were identified and

recommendations were made for implementation. These recommendations focused on

four areas. The first area was the beneficiaries’ disincentives to return to work.

Beneficiaries were concerned about the loss of cash benefits (the “cash cliff”) and the

loss of medical coverage if they returned to work. The Ticket program attempted to

address these concerns by extending medical coverage through the Medicaid Buy-In

process and by halting the Continuing Disability Review (CDR) for beneficiaries who

were using the ticket and making progress in their rehabilitation.

A second issue, provider participation, was addressed by recommending the

provision of a payment process that was seen as a viable business model for service

providers. The belief was that if ENs could share some risk up front through the

milestone payment process, and SSA would provide reimbursements in a timely fashion, it would be sufficient to attract a wide range of providers as Employment Networks.

The structure and efficiency of the SSA system was another concern as the Ticket program was developed. With the increase in the number of beneficiaries applying for benefits, and in addition to managing the claims of the non-disabled beneficiaries, concerns were raised regarding increasing the workload in the SSA field offices.

Consequently, recommendations for the SSA return to work system included making the process as streamlined and efficient as possible. Since the implementation of the Ticket program in 1999, SSA has explored a variety of computerized methods to provide

benefits, track payments and monitor work activities for disabled beneficiaries. The SSA has established and trained Area Work Incentives Coordinators (AWICs) who have been trained on the new work incentives. The AWICs are responsible for training local Work

Incentives Liaisons (WILs) who are specialists in promoting an understanding of work incentives in the SSA field offices under the Ticket program. However, as late as

February 2003, two years into the rollout of the Ticket program, the SSA field offices reported a substantial backlog in processing work incentives paperwork and training had only taken place for a small number of WILs (Einschlag, 2003).

The final recommendation for implementation of the Ticket program involved the missing link in the partnership of returning individuals with disabilities to work, the employer. The SSA was encouraged to develop incentives for employers to hire its beneficiaries. The belief was that this could happen in concert with other agencies and organizations such as the U.S. Department of Labor, the Chamber of Commerce and the

Business Leadership Network. These partnerships have been expanded, but the employer response has been skeptical and slow, as evidenced by the fact that to date, only two employers have elected to become Employment Networks (Maximus, 2003).

Additionally, no business partnership model has been developed with the Ticket program as occurred with Welfare to Work partnership.

The Ticket program took several gigantic steps to removing barriers for SSA beneficiaries who want to work. First, it increased choice in vocational rehabilitation for

SSA beneficiaries. It is the first time people with disabilities can decide which provider is the “best fit” for them for employment services and supports. Second, the Ticket program continued access to health care while working. This provision allows

beneficiaries to either continue Medicare or to buy into Medicaid coverage in order to continue to receive medical services necessary to become and remain employed. Lastly, because the legislators recognized that SSA would need assistance in implementing and operating the Ticket program, the legislation established the Ticket to Work - Work

Incentives Advisory Panel (Growick, 2000b).

2.3 Historical Antecedents of the Law

One of the major reasons this legislation was passed was to rescue the Disability

Trust Fund of the Social Security Administration from bankruptcy. In 1995, the SSA had to transfer $500 billion from its retirement fund to its disability fund in order to maintain solvency. The cost of cash benefits alone increased from $34.4 billion in 1989 to $62.9 billion in 1998. The lackluster performance of the SVRA system in assisting SSA beneficiaries in obtaining employment was cited as one of the reasons for this fiscal crisis

(Fallavolita & Bordelon, 1997; GAO, 1996). Alternate providers of rehabilitation services were needed so that more beneficiaries would be employed. Without a greater return on investment for the money SSA was spending on rehabilitation, the Disability

Trust Fund would continue to operate in arrears (Growick, 2000b).

Before passage of the Ticket program, the SVRAs were the sole providers of rehabilitation services for SSA beneficiaries. In this exclusive capacity, they were returning less than one-half of one percent of the eligible beneficiaries to work.

According to the GAO, there was a greater likelihood that a SSA disability beneficiary would either reach retirement age or die than return to work (GAO, 1996). This was simply unacceptable, and not fiscally prudent for SSA. Understandably, the SSA has recently established the Office of Employment Support Programs to oversee its delivery

of employment services through the Ticket program and to demonstrate its commitment

to employment for its beneficiaries with disabilities.

The Ticket program was created to encourage alternatives for the delivery of

rehabilitation and return to work (RTW) services for SSA beneficiaries so that more

individuals with disabilities would become employed. The SSA looked to the return on

investment in private-sector rehabilitation as a model for cost-effective rehabilitation

(Forgiel & Growick, 1997). The intention was that an increase in providers would rescue

the Disability Trust Fund of the SSA from financial insolvency. In fact, when the final

draft of the legislation was sent to the U.S. Office of Management and Budget for determining its monetary impact on the SSA Disability Trust Fund, the law was conservatively estimated to save 3.5 billion dollars. This savings was based on 75,000

(or just 1%) of the 7.5 million Americans of working age receiving SSA benefits going to work and leaving the disability rolls.

2.4 Highlights of the Ticket Legislation

In general, the law directed SSA to establish the Ticket to Work and Self-

Sufficiency (Ticket) Program with three goals in mind: 1) to provide disabled Social

Security beneficiaries with a ticket that allows them the freedom to select the

employment provider of their choosing for vocational rehabilitation services; 2) to

expand the number of employment service providers (ENs) to increase beneficiary

choice; and 3) to reduce the numbers of disabled beneficiaries on the rolls to maintain the

solvency of the Disability Trust Fund.

2.4.1 The Role of the Social Security Administration

The legislation, and the rules and regulations promulgated by SSA, delineated the rights and responsibilities of each of the stakeholders in the program: SSA, the Program

Manager, SVRAs, and Employment Networks. The SSA was given the responsibility of establishing the program by selecting one or more program managers (PMs) to assist in administering the program. In addition, the SSA was authorized to engage in the following activities:

• Terminate agreements with PMs who failed to meet the required performance

standards.

• Preclude PMs from providing direct services to beneficiaries.

• Select and enter into agreement with prospective ENs who choose to provide

vocational services under the program.

• Terminate agreements with ENs who fail to meet the required performance

standards.

• Provide for periodic reviews of ENs to ensure quality assurance in the provision

of services.

• Provide a process for resolution of disputes between beneficiaries and ENs, and

between PMs and ENs.

• Perhaps most importantly, TWWIIA (1999) also prohibits SSA from initiating

continuing disability reviews while a beneficiary participating in the program and

is making progress.

2.4.2 The Role of the Program Manager

Under the law, the responsibility of the Program Manager (PM) related to the recruitment and monitoring of the ENs, and ensuring the rights of beneficiaries were maintained and respected. Included in those responsibilities were the following:

• Recruit and recommend ENs for selection by SSA that can provide employment

and related services under the Ticket program.

• Monitor ENs to ensure that beneficiaries have adequate choice of services and

reasonable access to the services they need to return to work.

• Ensure that ENs comply with the terms of their EN agreement and that payment

by SSA is warranted.

• Ensure beneficiaries are allowed to change ENs for good cause.

2.4.3 The Role of the ENs

The role of ENs in the Ticket program was outlined in the legislation, as well. In general, they were responsible for providing services, meeting specified requirements, and implementing work plans. These responsibilities included the following:

• Assume responsibility for coordination and delivery of services to beneficiaries

who have assigned their tickets to the EN, either by providing services directly, or

by entering into agreements with other providers.

• Develop and implement an Individual Work Plan (IWP).

• Ensure that services provided under the Program meet the requirements of the

IWP.

• Meet the financial reporting requirements as prescribed by SSA.

• Prepare periodic reports as prescribed by SSA 36

The law also incorporated the initial design of the EN payment systems. It authorized SSA to pay an EN under one of two payment methods: outcome only or milestone-outcome. In addition, it allowed SSA to adjust the payment system to ensure that the incentives to participate for the ENs and beneficiaries were adequate. The law required that SSA report on the adequacy of the incentives to Congress within 36 months of the enactment of the legislation. The adequacy of the incentives was of particular concern to the drafters of the legislation who hypothesized that the harder to serve populations may find it to be more difficult to receive employment services under the

Ticket program with the proposed payment scheme.

Under the current outcome method, an EN is paid a percentage (not to exceed

40%) of the national average SSDI or SSI payment for each month the beneficiary works at or above $830/month (SGA) and does not receive a cash benefit. These payments cannot exceed 60 months duration. The milestone-outcome method, as indicated, is a combination of milestone and outcome payments. ENs can elect the milestone-outcome method of payment to be reimbursed for milestones that are achieved toward assisting beneficiaries to return to work. Presently, there are four milestone payments, after which the EN is then paid for providing services by the outcome method. The total amount of payments made to an EN under this payment system must be less than the outcome only payments because funding has been provided up front (Ticket to Work-Work Incentives

Improvement Act, 1999).

2.4.4 The Role of the SVRAs

The role of the state vocational rehabilitation agencies was addressed, and the

Ticket to Work Advisory Panel and the Benefits, Planning and Outreach (BP&O)

Program were established in the legislation. According to the law, SVRAs could elect to participate in the program as an EN, or could continue to operate under a cost reimbursement method of payment on a case-by-case basis. The law also required a written agreement between the SVRA and an EN before a state agency could accept a referral of a beneficiary. The responsibility for proscribing the regulations related to the terms of these agreements rested with SSA and the Secretary of Education, Office of

Special Education and Rehabilitation Services (OSERS). The law prohibited the payment to an EN if it referred a beneficiary to the SVRA without entering into an agreement, and required SSA to establish a dispute resolution process if the SVRA and an EN failed to reach an agreement regarding the cross-referring of beneficiaries.

2.4.5 The Role of the Advisory Panel

The legislation also established the twelve-member Ticket to Work-Work

Incentives Advisory Panel to assist the President, Congress and SSA in the implementation of the Act. All twelve members must represent the interests of beneficiaries, service providers, employers and employees. Beyond providing counsel with respect to the initial roll-out of the tickets, their role incorporated advising in five additional key areas: 1) adequacy of the EN payment systems; 2) work incentive programs for the beneficiaries; 3) benefits planning and assistance for individuals with disabilities; 4) evaluating the work incentives provisions under Titles II, XI, XVI, XVIII, and XIX of the Act; and 5) monitoring the SSA management information systems to facilitate the successful implementation of the program. Another primary purpose of the

Ticket Advisory Panel was to assure the success of the Ticket program by ensuring that beneficiaries have adequate access to ENs, and that the most effective designs for

research and demonstration projects related to the Ticket program are conducted. The

Ticket Advisory Panel must furnish annual reports to the President and Congress, and provide progress reports on the Ticket program to SSA and Congress, which it has done on a quarterly and annual basis.

2.4.6 The Role of the Benefits, Planning and Outreach Organizations (BPAOs)

The BPAOs were created by the law to advise SSA beneficiaries regarding the impact of returning to work on their cash and medical benefits. The major changes to health care coverage for those beneficiaries who returned to work were, perhaps, the most critical provisions of the legislation, in terms of reducing barriers to employment for disabled SSA beneficiaries. The first of these provisions was the expanded availability of health care services that allowed each state to develop a Medicaid Buy-In program. The

Medicaid Buy-In permitted states to liberalize limits on resources and income, and allowed individuals with medical impairments to buy into Medicaid although they may no longer be eligible for SSDI or SSI cash or medical benefits. The second provision expanded the continuation of Medicare coverage by extending Medicare Part A for an additional 48 months for individuals with disabilities who returned to work. Anecdotal and research data has recorded that the BPAOs have been in great demand and have consistently provided benefits counseling services to ticket holders.

2.4.7 Research Related to the Employment of People with Disabilities

A final major aspect of the legislation requires the GAO to conduct a number of studies related to the employment of people with disabilities. One study was targeted to evaluate the effectiveness of existing tax credits and the other, employer incentives for hiring people with disabilities. Other studies were to focus on the adequacy of work

incentives and expanded medical coverage for disabled beneficiaries, and the merits of increasing SGA levels as a further incentive to employment.

2.5 Use of the Internet in Survey Research

This survey research project was accomplished through a web-based data collection process. The use of the Internet in survey research has transitioned from a rarely employed method of social science research to a methodology that has become widely recognized and researched. This transition has occurred largely due to the phenomenal increase in the use of the Internet on a world-wide basis (Zhang, 1999;

Schmidt, 1997). Many studies have cited the positive aspects of using the Internet for survey research. These benefits include a faster response, protection against the loss of data, easy transfer of data into a database for statistical analysis, cost savings, and the possibility of reaching a wider audience (Mertler, 2002; Schillewaeert, et al., 1998).

Studies on the use of the Internet for research have also cited numerous methodological concerns: low response rates, self-selectivity of Internet users, technological issues related to the dissemination and use of the survey, and concerns over Internet security

(Sills, 2001; Schmidt, 1997). These concerns are included in external threats to validity along with the issues of sampling error, coverage error, non-response error, and measurement error (Dillman, 2000).

2.5.1 Sampling Error

Sampling error occurs when a subset of a heterogeneous population is selected to represent the whole population. This type of error is generally a result of attempting to survey a representative number of a population, but not all of the population. Using randomized sampling methods may minimize the variance that may occur. Increasingly,

market researchers are using purposive and convenience samples to target their

audiences on the Internet in order to measure consumer attitudes and satisfaction. In

these cases, the market researchers have concentrated their efforts on the quality of the

responses gathered rather than the generalizability of the data (Sills, 2001). Those

researchers who have used traditional probability sampling methods have found that their

attempts to increase generalizability have resulted in inherent biases of the data due to the

self-selection of the respondents (Yun, 2000). Non-coverage error occurs when the

sampling frame does not cover all members of a population. As a result, not all members

have an equal chance of being selected to participate (Dillman, 1991). This issue is a

major concern when using probability sampling.

2.5.2 Non-response error

Non-response error occurs when a significant number of people in a survey do not respond and differ in some important way from those who did respond. It is a challenge that faces many researchers who chose the Internet or mixed methods, including the

Internet, to conduct research. Studies on the use of the Internet for survey research reference the lack of representative samples and the self-selection of respondents as critical areas of concern and relevance for survey data collection. (Schmidt, 1997;

Couper, 2000). However, if the population is known and can be readily and cost- effectively accessed, issues of generalizability are not major considerations in non- response error. Within the context of Internet survey research, non-response error can be minimized by the use of mixed mode methods of data collection, according to Dillman

(2000).

2.5.3 Response rate

Response rate is predicated on two factors. The first factor is the willingness of the participant to engage in the cognition necessary to respond. From both a behavioral and cognitive perspective, great cost is involved on the part of the respondent to carefully consider each question and to answer accordingly. This is especially true if the survey has been constructed to minimize the possible order effect of the questions.

Within the social exchange theory of human behavior, it is believed that the actions of an individual are a consequence of the response that person expects to get as a result of this action. Dillman defines these factors as rewards, costs and trust (Dillman,

2000). Token incentives provided for responding to a survey have been extensively studied by a variety of researchers. Most found that incentives provided as an immediate reward with the survey resulted in a higher response rate, while those provided as a promise for participation resulted in a lower response rate (Dillman, 2000; Bosnjak,

2003). Rewards as social influencers include showing positive regard, saying thank you, asking for advice and supporting group values.

The second factor is the commitment of the researcher to engage in multiple attempts to contact the respondents to increase the response rate. Multiple mailings, and reminder telephone contacts, postcards and emails have all been found to be effective means to increase response rate and reduce non-response error (Dillman, 2000).

Importance is attached to making the survey look professional, making the task easy to understand, and making the process more personalized to increase the response rate

(Fowler, 1995).

Many studies have documented the variability of response rates in Internet survey

research. Response rates as high as 70% have been found when using mixed methods of

data collection. In contrast, lower response rates occurred when Internet surveys used

listservs or e-mail addresses to target potential respondents (Bosnjak, 2003; Couper,

2000; Sills, 2001; Schmidt, 1997).

2.5.4 Technological issues

Technological issues can be significant factors in non-response errors for web- based surveys. Differences in web browsers, downloading time and the capabilities of a respondent’s server can effect the willingness and desire to participate. A survey designed for the web, although constructed in an HTML language may appear differently on different browsers. In order to minimize this effect the survey designer must check the survey instrument on as many browser types as possible to ensure that its appearance is as consistent as possible across web platforms. Downloading time has also been shown to be a significant deterrent to response rate for web-based survey instruments.

Heerwegh and Loosveldt (2002) found that increased downloading time decreased the response rate of participants by 25% in survey of web users on a college campus. This finding was significant because the assumption was made that the students participating in the survey were Internet proficient and familiar with variations in downloading times.

2.5.6 Measurement error

Measurement error occurs when a respondent’s answer to a question is imprecise, inaccurate, or cannot be compared to the responses of other participants in a survey

(Dillman, 2000). With self-administered surveys measurement error could occur as a result of lack of motivation, comprehension problems, lack of knowledge, or deliberate

misrepresentation of opinions or attitudes (Couper, 2000). Some aspects of measurement error can be controlled for web-based survey research by identifying and enlisting participation from organizations or individuals who may have an inherent interest in the topic and using specific software programs that will not allow the survey to be submitted without all appropriate fields completed (Schmidt, 1997).

CHAPTER 3

3. METHODOLOGY

3.1 Survey Population

The design of this study was non-experimental, descriptive research (Campbell &

Stanley, 1963). A secured, web-based site was utilized, supplemented by mailed questionnaires, if requested, and mail and telephone follow-up. This survey site was designed to meet the web accessibility requirements as defined by the Web Accessibility

Initiative (WAI) 2000 and Section 508 standards. Permission was requested and received from The Ohio State University Institutional Review Board to proceed with the project as proposed.

3.2 Survey Design

The survey instrument was designed by the researcher with the assistance of the

Ohio State University Center for Survey Research (CSR). The sample survey was submitted to a panel of experts in the field of vocational rehabilitation, and the Social

Security Administration for review and comments. The panel was advised about the objectives and purpose of the study, and the survey was posted to the website. Panel members were asked to evaluate the survey questions for clarity and content, to provide

45 feedback on the survey items, and recommend changes or additions to the instrument.

They were also asked to provide feedback regarding the survey for ease of use in terms of downloading time and formatting on their browser. Both face and content validity were established with the assistance of the expert panel. Comments related to clarity and content were solicited. (See Appendices A and B for panel of experts’ cover letter and the list of experts). (See Appendix H for the complete survey.)

In addition, a pilot test was conducted with a random sample of forty providers who were selected from the population. The organizations that were selected for the pilot resembled the general population of organizations because of their status as non- participants in the Ticket program. Modifications to the survey instrument were made based on the feedback of the panel of experts and the results from the pilot test. Wording of some items, and directions for the survey were clarified or modified.

The pilot study demonstrated that some respondents were not interested in participating in the survey. It also identified potential problems with the population to be surveyed. Eleven of the forty organizations were either out of business or could not be found. Of the 29 organizations contacted, five responded after a three-week period of data collection and follow-up phone calls. Those non-respondents reached for comment

(n=10) indicated they were not interested in the Ticket program and did not want to take the time to respond to the survey. Following this pilot, a sample analysis of the data was conducted to ensure accuracy of the data collection and efficiency of the process utilizing the data dumping method into SPSS. Minor changes were made to the data coding to streamline data collection.

The survey instrument was divided into three separate sections: General

Knowledge of the Ticket to Work Program, Ticket to Work Program Components, and

Demographic Information. All items were to be answered by all respondents. No skip patterns were used in the design. Section A, General Knowledge of the Ticket, began with a four-point Likert scale to determine the level of general knowledge regarding five areas: 1) overall Ticket program; 2) payment system; 3) beneficiary status reporting requirements; 4) record-keeping requirements; and 5) requirements for providers to become ENs. The respondents were asked to rate their knowledge of the program as

“very knowledgeable,” “somewhat knowledgeable,” “not at all knowledgeable,” and

“don’t know.” Respondents were also asked to respond to four additional questions: 1) what percentage of people they served were referred by the state vocational rehabilitation program; 2) was their organization interested in becoming an EN; 3) what were all the source(s) of their knowledge regarding the Ticket program; and 4) their current position within the organization.

Section B of the survey, entitled “Ticket to Work Program Components,” addressed specific program components and possible barriers to participation as an EN.

For ease of reading and responding, this section was divided into 10 blocks with five questions in each block. In total, 50 items were contained in the ten blocks. All 50 items included a don’t know option.

The Section C, “Demographic Information”, asked for demographic data regarding the type of organization, number of clients and staff, services provided, and geographic location. In addition, in two separate areas, respondents were given the opportunity to share additional comments in an open-ended format. In Section C, the

respondents were asked “Please tell us under what circumstances your organization would want to participate in the Ticket program”. In Section D, they were told they had an opportunity to tell the Social Security Administration what they wanted it to know about the Ticket program. They were assured their comments would be held strictly confidential. A separate analysis of the text was conducted on these responses to identify major themes of the comments.

Based on the researcher’s professional experience, the feedback of the panel of experts, and the results of the pilot, the researcher recognized that many of the questions might elicit strong responses. A decision was made to control for order effects such as value-based responses, anchoring issues, and addition (carryover) and subtraction effects.

In addition, the researcher developed three instruments with identical sections for A & C, but randomizing items in Section B. With the assistance of the CSR, questions related to the specific issues and program components were randomly listed in order to control for the potential order effect of the questions and to ensure that the survey design had no effect on the responses. Each of the three web-based instruments was constructed in an

HTML format (Microsoft Front Page) to allow input of the responses only by the use of a case ID number, thus reducing the potential for error from multiple submissions.

The prospective ENs were randomly assigned to one of the three survey instruments using the computerized random numbers generated by the CSR.

Respondents with random numbers beginning with one were assigned to survey one; numbers beginning with two were assigned to survey two, and numbers beginning with three were assigned to survey three. All three surveys were posted on the College of

Education server with dedicated web-site addresses. Based on the Dillman (2002) tailored design method for survey research, each organization received a pre-notification letter explaining the purpose of the survey, and asking for their participation in the survey and (See Appendices C and D for copies of the notification letters). The letter contained a randomly generated ID number, along with the link to the corresponding survey website. This method limited access only to authorized participants, and ensured confidentiality.

Data collection was streamlined by eliminating the need for coding. The data was directly “dumped” into Excel for review of the analysis for any formatting issues. It was then “dumped” into SPSS, a statistical analysis program that was utilized to analyze and describe the results. This reduced errors that might have arisen from manually coding and transcribing information. A small amount of data clean-up was required in order to successfully transfer the raw data from Excel into SPSS.

3.3 Instrument Reliability

Several factors have influenced the reliability testing of this instrument. First, because the constructs to be measured by this instrument were unique to a new federal program, no other instrument existed that could be utilized for measures of comparison to establish reliability. Second, this type of survey instrument is not conducive to the test- retest method of reliability because this study was designed to measure the respondents’ attitudes at the point in time the survey was taken. Third, a parallel form reliability measure was inappropriate due to the potential for sampling error inherent in the construction of two different forms, the length of time required to administer the instruments, and the large sample size needed. Consequently, a measure of internal

consistency for a single administration of an instrument with multiple response categories was chosen as the appropriate measure of internal consistency. The preferred reliability procedure was Cronbach’s alpha to produce a coefficient of internal consistency (Ary,

Jacobs, & Razavieh, 2002). This procedure was conducive to a Likert-type scale and allowed the researcher to split the 50 questions in Section B to analyze the consistency of responses. The Cronbach’s alpha for this instrument was .9318, thereby demonstrating a high degree of internal consistency.

3.4 Survey Implementation

In order to increase survey response, the pre-notification letter followed the principles of the tailored design method as described by Dilman (2000). The respondents were advised that their opinion on the Ticket program was valued, and needed, in order to provide SSA with information that could be used to improve the program. The letter contained the following information: website URL, unique case ID number, notification of the availability of the paper survey, and an assurance of confidentiality of the survey information. (See Appendices C and D for the notification letters). In addition, each organization was given the opportunity to win $50 in free gas for completing the survey. A return postcard to be entered into the drawing was included in the letter. (See Appendix E.) All respondents who participated in the survey and returned the postcard were entered into the drawing. Twenty respondents were selected to receive the $50 gas card.

Undeliverable letters were collected and telephone contact was initiated with those organizations to up-date their contact information for a repeat mailing. There were five purposes for this telephone contact:

a) to advise the subjects about being chosen for the study;

b) to identify the correct contact person and contact information;

c) to determine the willingness to participate;

d) to clarify the most appropriate survey format for the participant from a

range of options, including accommodations for visual impairments, etc.,

e) to provide current information to update both the University of Wisconsin-

Stout and IARP’s databases.

Identification of the contact person for each provider with undeliverable letters

took place to determine who was responsible for making the decision regarding

participation as an EN. The sole criterion for responding to the survey was the

individual’s role as the responsible party for making the decision to participate in the

Ticket program. The telephone contact, which was conducted by the researcher,

confirmed the name of the person to be contacted, contact information, confirmation of

web-based accessibility and determined if an alternative format, such as large print or a

paper survey, was required. This process was designed to reduce non-response error by

ensuring that the respondents received the survey in the most user-friendly format. It also ensured equal ability to participate regardless of ability to access the Internet. The use of a web- based survey enabled the researcher to reach potential respondents in a diverse range of geographic areas. Organizations that were out of business or could not be found at the time of the initial survey were removed from the database and placed into an out of business category.

Reminder/thank you postcards were sent at three, five and seven weeks. After 12 weeks, an attempt was made to contact the non-respondents from each state. The non-

respondents were identified by selecting every 7th organization from the randomized list of organizations for each state. During the initial phase of the survey, the non- respondents were called over an eight- week period to solicit their responses. They were given the option of completing the survey by phone or filling it out and returning it by fax. The researcher conducted the follow-up to non-respondents. Those organizations that declined to participate were counted and reasons for declining were collected, when possible. Surveys completed with significant missing data were not included in the final data analysis.

The period for data collection was extended due to the problems with the College of Education’s server during the initial survey, and the subsequent low response rate.

During the extension, attempts were again made to contact non-respondents by telephone.

In each state, every 7th organization was selected beginning at a random point on the list

in order to reduce sampling bias. Following the protocol of the initial data collection,

reasons for non-participation were compiled whenever possible.

3.5 Data Analysis

The statistical methods of choice for the survey results were descriptive statistics and non-parametric tests. Non-parametric tests were used to analyze the data because little was known about the characteristics of the population of prospective ENs in the six states. Frequency tables, contingency tables, and chi-square tests for independence were used to evaluate the responses of the participants. The alpha level for significance for the

chi-square tests was set a priori at p=.05.

An exploratory factor analysis of the more critical items related to Ticket program

components was also conducted using principal components analysis in SPSS. The

researcher hypothesized that some of the observed variables were a combination of underlying factors associated with payment issues, payment options, and EN programmatic issues based on the literature, knowledge of the researcher, and anecdotal information. Consequently, the survey was intentionally constructed to measure those factors. The goal of the factor analysis was to reduce the data set for parsimony, and for additional, and more meaningful statistical analysis.

Factor analysis is designed to identify a small set of factors that represent an underlying relationship among a group of variables. According to Stevens (2002), there might be a greater opportunity for chance in finding statistical significance for factor loadings when the sample size is small (less than or equal to 100). He opined, “Because of this capitalization on chance, the formula for the standard error of correlation can seriously underestimate the actual amount of error in the factor loadings” (Stevens, 2002, p.393). Stevens recommended a two-step conservative approach to determining statistical significance. He advised using two more stringent measures. First, set the alpha level at a more conservative .01 for a two-tailed test. Second, he also recommended checking whether a factor loading is statistically significant by doubling the critical value required for significance for a correlation (Stevens, 2002). In order to avoid finding statistical significance in factor loadings when there might be none (Type I error), the decision was made to follow Stevens’ approach.

In general, a multi-step decision process was used to determine the principal components. The first step was to determine the suitability of the data for factor analysis using three methods. This involved inspecting the correlation matrix for coefficients of

.3 and above, and calculating the Kaiser-Meyer-Olkin Measure of Sampling Adequacy

(KMO MSA) and Bartlett’s Test of Sphericity. Ideally the sample size should be 150+, or at least 5 cases for each variable in order to conduct a viable factor analysis. Because this is not the case in this study, Stevens’ approach to identifying significant factor loadings was applied. The alpha level for a significant factor loading was set at .01, a more conservative measure than the .05. The critical value for a correlation coefficient at α = .01 for a two-tailed test is .256. This number was doubled, requiring a critical value of .512 in order to be considered a statistically significant loading, as recommended by Stevens.

The following basic assumptions were made regarding the factor analysis process:

• Correlation Matrix: In order to be suitable for factor analysis, the correlation

matrix must have some correlations of r=. 512 or greater;

• The KMO MSA value must be statistically significant at .06 or above; and

• The Bartlett’s Test of Sphericity must be statistically significant at p<.05. This

test was particularly critical given the sample size.

The second step involved determining how many underlying factors there were in the set of variables and which ones to retain. Using the Kaiser’s criterion, only components that had an eigenvalue of 1 or greater were retained. The scree plot was used to identify the break point in the data by plotting the eigenvalue on the y axis against its ordinal number on the x axis. (The total variance explained tables, correlation matrixes and scree plots can be found in Appendix G under the specific group headings.)

CHAPTER 4

4. RESULTS

4.1 Population and Response Rates

The combined population databases received from the University of Wisconsin-

Stout and IARP totaled 2,441 organizations within the six states. A random sample of 40 organizations was selected for the pilot study, thus reducing the population (n=2,401). At the beginning of the study in June 2003, the list of organizations was cross-checked against the current EN list on the Maximus web-site. The 125 organizations that were identified as current ENs were removed from the list. An additional 102 organizations were found to be duplicate entries. Both through returned undeliverable mail and through initial telephone contact, 442 organizations were found to be either out of business or could not be located. Consequently, the population to be surveyed was reduced by 669 to a census for the initial phase (n=1,732).

4.2 Initial Survey Response Rate

Of the census (n=1,732), 176 organizations responded. This resulted in an overall initial response rate of slightly over 10% of the population. The breakdown of the 176 respondents was as follows: declined to participate, no employment program –47; identified as current ENs- 21; cases missing substantial data- 8: final n=100. The

55 decision was made to remove the additional eight cases due to a substantial amount of missing values, rendering the data unusable. The Table 4.1 further breaks down the response and non-response rate by state.

State Participate Missing Declined Already Non- d Data No Employment EN Respondents Valid data Program

AZ 10 8 89 CO 9 10 201 FL 28 7 400 NY 18 11 390 OK 8 3 113 WI 27 8 336 TOTAL 100 8 47 21 1556 S NOTE. EN = employment networks.

Table 4.1 Response Rate for the Initial Survey

4.3 Extension Period

Due to the initial low response rate and the problems with the College of

Education server, the data collection phase was extended to increase the response rate.

During the extension, the decision was also made to further determine reasons for non- participation of the providers. Potential reasons for not participating in the survey were separated into three sub-categories: 1) declined no employment program, 2) declined due to lack of interest, 3) out of business or no listing. The follow-up contacts took place from July to October 2004. During that time 1,057 calls were made to non-respondents across the six states following the protocol described in the methodology section.

4.4 Response Rate Following the Extension Period

The non-respondents from the initial study totaled 1556 organizations. Of that number, there was no telephone number or other current contact information that could be found for 160 organizations. These organizations may be out of business, have changed names or have been acquired or subsumed by another entity. An additional 76 organizations were duplicate entries with different names but affiliated with another organization. Consequently, of the non-respondents (n= 1556) for the extension phase, only 1320 were credible entities that could be contacted. Of that number, 508 organizations responded during the extension, either by completing the survey or declining to participate. The breakdown of the 508 respondents during the extension was as follows:

• 7 new respondents to the survey; however, they all were currently ENs;

• 414 declined to participate because they had no employment program;

• 60 declined to participate due to lack of interest; and

• 27 were new ENs and did not appear on the Maximus web-site at the time of the

initial survey, or at the beginning of the extension.

The decision was made to remove the seven new respondents since they were already

ENs, rendering the data unusable in this project. The 414 organizations with no employment programs typically provided physical rehabilitation services or other social services such as counseling, medical case management, or transportation and housing supports.

Those organizations that declined due to no interest stated a variety of issues.

Themes from these statements included the perceived inadequacy of reimbursements, 57

especially with the harder to serve population. The perception that the reimbursements would not be timely was also an issue. Two of the organizations that declined to participate during this period were former ENs who referenced problems with the timeliness of reimbursement, and the low rate of reimbursement as major barriers to their participation. The following table breaks down the response rate by state for the extension.

No No Already Responded Total Non State No EP Interest Listing Duplicate EN ENs Respondents AZ 42 6 14 7 8 12 CO 72 15 27 21 4 62 FL 74 20 23 9 11 263 NY 121 14 30 18 4 203 OK 16 * 3 22 1 0 71 WI 89 2 44 20 0 181 TOTALS 414 60 160 76 27 7 808 Note. *Includes two former ENs. EP = employment programs; EN = employment networks.

Table 4.2 Response Rate for the Extension Phase

4.5 Overall Response Rate for the Initial and Extension Period

In order to determine the combined response rate for the initial survey and the extension, organizations were eliminated that were duplicates or could not be found.

This reduced the known population from 1732 in the initial survey to 1320 for the extension. The resulting combined response rate for both the initial survey and the extension period was slightly over 46.6 % of the known population with 616 total

respondents. Table 4.3 summarizes the disposition of the response rate for the population

for both the initial survey and extension.

Respondents Respondents Declined Declined Already Total Non- W/Missing No No ENs Respondents Data Employment Interest Program TOTALS 100 8 461 83 55 808

Note. ENs = Employment networks.

Table 4.3 Overall Response Rate for the Initial and Extension Phase

4.6 Statistical Analysis

Each of the research hypotheses measured the interest in becoming an EN with critical Ticket program issues or organizational characteristics. For each of the hypotheses the interest in becoming an EN was represented by a dichotomous, nominal variable with values of 0 for No and 1 for Yes. The decision matrix used to determine if interest in becoming an EN was independent of organizational characteristics or Ticket program issues involved a two-step process. For the chi-square tests of independence it included comparing the calculated X2 to the critical X2 value, and a comparison of the p

value to the alpha level of .05.

4.6.1 Interest in Becoming an EN

One critical question of this study was what is the level of interest in becoming an

EN. A simple yes/no response was requested. As shown in Figure 4.1, the majority of

respondents, 63%, indicated they had no interest in participating in this program.

Although this information must be interpreted with caution given the low response rate, it does reflect the attitudes of those who responded. If it is also indicative of the attitudes of prospective ENs in other areas of the country, then substantive changes to the Ticket program are needed to attract more providers.

Interest in Becoming an EN

No Response 4.0%

YES 33.0%

NO 63.0%

Note. Percentages of interest are shown for yes, no and don’t know categories

Figure 4.1 Interest in Becoming an EN

4.6.2 Working Settings and Interest in Becoming an EN

The first area of exploration was the relationship between interest in becoming an

EN and work setting. For the purposes of this survey, work setting was separated into 21 categorical levels (See frequency table in Appendix G). The frequency count for these variables showed that the largest number of organizations were located in urban settings

(17 or 18.1%). Small town locations totaled 11 (or 11.7%) of the organizations.

Organizations located in two work settings were the next largest groupings. They were urban/suburban settings and small town/rural at 9.6% and 8.5%, respectively. Seven organizations were located exclusively in rural areas, while six organizations had work sites exclusively in suburban areas, and four in inner city sites. Combinations of worksite locations made up the remainder of the groupings.

The contingency table for the chi-square was used to determine the interest in becoming an EN based on geographic location of the organizations. The chi-square test for independence could not be interpreted because 54% of the cells had frequency counts less than 2.

Organizations located exclusively in urban settings were the largest proportion of respondents that were interested in becoming an EN and represented 16% of the total organizations interested in becoming an EN. Organizations located in small town or urban/suburban settings were the second largest proportion of respondents interested in being an EN.

4.6.3 Geographic Locale and Interest in Becoming an EN

The second research hypotheses addressed whether the interest in becoming an

EN was independent of geographic locale. In addition to the yes, no choice for interest in being an EN, the other categorical variables for this chi-square were East, Midwest and West. Each of the categories included two states. Because one respondent failed to provide the complete ID number, the organization’s geographic locale could not be determined. Consequently, that response has its own category of “geo” in the following

2 table. The null hypothesis (H 0) was there is no relationship between geographic location and interest in becoming an EN.

As shown in the frequency count in Table 4.4, the East group included organizations from NY and FL. This was the largest group not interested in becoming an

EN. Their no responses comprised over 40% of the total organizations indicating a lack of interest in being an EN. However, they were also the largest group indicating an interest in becoming an EN.

Because it was a 2 x 4 contingency table, a Yates correction was not required. In this chi-sqaure test, X2(df=3, n=98). As shown in Table 4.5, the calculated value of

X2=1.431 at the .05 α level. The critical value of X 2 = 7.82. Additionally, the p value of

.698 is greater than the α value of .05. Therefore, at an alpha level of .05 with a two- tailed test, the statistical hypothesis of independence was accepted. It was concluded that the interest in becoming an EN was not dependent on geographic location.

Be An EN No Yes Total Geoa 1 0 1 East 28 16 44 Midwest 21 12 33 West 15 5 20 Total 65 33 98 a. This value reflects one organization that responded without a full ID number. Therefore the geographic locale could not be determined.

Table 4.4 Geographic Location and Interest in Being an EN

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 1.287a 3 .732 Likelihood Ratio 1.627 3 .653 N of Valid Cases 96 a. 2 cells (25.0%) have expected count less than 5. The minimum expected count is .34.

Table 4.5 Chi Square Test for Geographic Location and Becoming an EN

4.6.4 Numbers of Direct Services Offered and Interest in Becoming an EN

3 The third research hypotheses (H 0 ) addressed whether there was a relationship between interest in becoming an EN and the number of services offered by the organizations. The data regarding services was collected for two reasons. One was to determine the range of services that may be available to Ticket holders if these providers were recruited to become ENs. The second was to assess whether the number of services provided influenced the decision to not participate as an EN.

As shown in Figure 4.2, the majority of providers offered evaluation and assessment, job development, job readiness, and career counseling (CC). All of these services are critical to assisting SSA beneficiaries in returning to work. Other services less frequently offered included rehabilitation plan development, employment support and maintenance, job training, supported employment (SE), and benefits planning and assistance (BPAO). Only one organization reported offering occupational certificates or licensing training (CERT/LIC).

Direct Services Offered

90 80 70 60 50 40 30 20

Number of Providers of Number 10 0 r v rt T ic e o E L D lan p CC S her b P eady p ob rt/ h R J BPAO e Ot Jo e Su C R Job E Eval & Assmt Types of Services

Note. Number of providers offering specific types of services.

Figure 4.2 Number of Direct Services Offered

The number of services offered was tallied by a frequency count and then grouped by quartiles determined by SPSS. It was then cross-tabulated to determine if the interest in becoming an EN was independent of the number of services provided by an organization. The average number of services provided by the respondents was 5.85, which was close to the median number of services provided at 6.0. Forty-one percent of the organizations offered from one to five services, eighteen percent six services, nineteen percent offered seven services, with the remainder (23%) offering more than seven services (See Appendix G for frequency tables).

The null hypothesis for this question was there is no relationship between number of services provided and interest in becoming an EN. Because it was a 2 x 4 contingency table, a Yates correction was not required. As shown in Table 4.6, X2(df=3, n=84). The

calculated value of X2=1.953 at the .05 α level. The critical value of X2 = 7.81.

Additionally, the p value of .582 is greater than the α value of .05. Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence was accepted. Interest in becoming an EN was independent of the number of services an organization offered.

As shown in Table 4.7, organizations that offered from zero to five services were least likely to be uninterested in becoming an EN, as a percent of the total responses.

However, of the organizations that expressed an interest in becoming an EN, they were the largest group. Those that offered more than seven services were the next largest group expressing interest. This may indicate that SSA has fertile ground for marketing to the smaller providers who may be seeking multiple revenue streams, and to larger, multi-service providers who can absorb the current upfront financial risks of the Ticket program.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 1.953a 3 .582 Likelihood Ratio 1.945 3 .584 Linear-by-Linear Association .663 1 .415 N of Valid Cases 84 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.71.

Table 4.6 Chi-Square Test for Interest in Becoming and EN and Number of Services Offered

Interest in Becoming an EN and Number of Services Offered

Number of Services Offered 0 - 5 6 7 >7 Total Be an No Count 23 8 11 10 52 EN Expected Count 21.0 9.3 9.9 11.8 52.0 % within been 44.2% 15.4% 21.2% 19.2% 100.0% % within Number of 67.6% 53.3% 68.8% 52.6% 61.9% Services Offered % of Total 27.4% 9.5% 13.1% 11.9% 61.9% Yes Count 11 7 5 9 32 Expected Count 13.0 5.7 6.1 7.2 32.0 % within been 34.4% 21.9% 15.6% 28.1% 100.0% % within Number of 32.4% 46.7% 31.3% 47.4% 38.1% Services Offered % of Total 13.1% 8.3% 6.0% 10.7% 38.1% Total Count 34 15 16 19 84 Expected Count 34.0 15.0 16.0 19.0 84.0 % within been 40.5% 17.9% 19.0% 22.6% 100.0% % within Number of 100.0% 100.0% 100.0% 100.0% 100.0% Services Offered % of Total 40.5% 17.9% 19.0% 22.6% 100.0%

Note. Contingency table showing range of number of services and percent of providers interested in being an EN in each range.

Table 4.7 Interest in Becoming an EN and Number of Services Offered

4.6.5 Ticket to Work Program Components

Section B of the survey contained 50 questions that are related to specific Ticket program components. These components were identified as issues for prospective ENs based on the following factors: 1) literature review; 2) professional experiences of the researcher; 3) input of the expert panel; and 4) on anecdotal information collected informally from prospective ENs around the country as the Ticket program has been implemented. Respondents were asked to indicate to what extent they agreed or disagreed with each of the statements. 66

A seven-point scale was constructed to provide greater variability in the responses for statistical purposes based on the assumption that there would be a large number of respondents. Because only 100 organizations actually completed the survey, the use of a seven-point scale created data that was too dispersed to provide meaningful analysis.

Consequently, the Likert-type scale was collapsed from seven categories to three in order to simplify the reporting of the results and for data analysis. “Strongly disagree,”

“somewhat disagree,” and “disagree” were combined into one variable, disagree.

Similarly, “strongly agree,” “somewhat agree,” and “agree” were combined into the variable, agree. Thus, the results are reported in three categories: “disagree,” “agree,” and “don’t know.” In each of the figures that follow, the question number in the figure title is the number of the question on the actual survey. This is provided as a reference so that the reader may refer to the actual question in the survey located in Appendix H.

An initial exploratory factor analysis was conducted to separate components related to the payment system and payment options/ issues using SPSS. Nineteen questions were included in this principal components analysis (PCA). They are the following questions that can be found in the survey in Appendix H:: 1, 3, 4, 5, 6, 7, 8, 9,

11, 12, 13, 14, 15, 16, 19, 23, 28, 34, 36. Inspection of the correlation matrix revealed many coefficients of .3 or higher. The KMO was .790 and the Bartlett’s Test of

Sphericity reached the level of statistical significance at .000, supporting the factorability of the correlation matrix. The component matrix revealed the presence of many coefficients of .512 or larger. The PCA revealed the presence of five factors with eigenvalues greater than one. These eigenvalues explained 60% of the variance. An inspection of the scree plot revealed a clear break after two of the components. To aid in

the interpretation of these two components, a Varimax rotation was performed. The rotated solution revealed the presence of a simple structure. The two components retained were labeled payment system and payment issues/options. (See Appendix G for tables).

Interest in Becoming an EN and Attitudes Toward the Payment System

Based on the initial PCA, questions 4,5,6,7, and 8 were retained as the principal component of payment system. A test was done to determine the suitability of the data for factor analysis. (See Appendix G for tables.) The correlation matrix showed many coefficients of .3 or higher. The KMO was .762 and the Bartlett’s Test of Sphericity reached the level of statistical significance at .000, supporting the factorability of the correlation matrix. The component matrix showed the presence of many coefficients of

.512 or larger. The PCA revealed the presence of one factor with an eigenvalue greater than one. This eigenvalue explained 60% of the variance. An inspection of the scree plot revealed a clear break after the first component. Since there was only one component, the simplest structure had been identified and a Varimax rotation could not be performed.

4 The fourth research hypotheses (H 0) explored whether interest in becoming an

EN was independent of the attitudes toward the payment system. Two aspects of the payment system were the following: 1) structure of the payment system; and 2) the overall understanding of the payment system. The contingency table (Table 4.8) shows the level of agreement with the concept that the Ticket payment system is a barrier to participation. Thirty-five percent of the respondents who agreed with this statement also indicated their organization was not interested in being an EN, while 25% who disagreed

expressed interest in being an EN. Overall, the distribution of the responses was fairly evenly divided between the three levels of agreement.

A chi-square test for independence was performed (Table 4.9). In this chi square test, X2(df=2, n=92), the calculated value of X2= 2.393 at the .05 α level. The critical value of X 2 = 5.99. Additionally, the p value of .302 was greater than the α value of .05.

Therefore, at an alpha level of .05 with a one-tailed test, the null hypothesis of no relationship was retained. The interest in becoming an EN was found to be independent of the attitudes regarding the Ticket payment system.

Interest in Becoming an EN and Payment System as Barrier

Level of Agreement Don't Disagree Agree Know Total Be an No Count 23 21 16 60 EN Expected Count 20.2 20.9 18.9 60.0 % within been 38.3% 35.0% 26.7% 100.0% % within Payment 74.2% 65.6% 55.2% 65.2% System as Barrier % of Total 25.0% 22.8% 17.4% 65.2% Yes Count 8 11 13 32 Expected Count 10.8 11.1 10.1 32.0 % within been 25.0% 34.4% 40.6% 100.0% % within Payment 25.8% 34.4% 44.8% 34.8% System as Barrier % of Total 8.7% 12.0% 14.1% 34.8% Total Count 31 32 29 92 Expected Count 31.0 32.0 29.0 92.0 % within been 33.7% 34.8% 31.5% 100.0% % within Payment 100.0% 100.0% 100.0% 100.0% System as Barrier % of Total 33.7% 34.8% 31.5% 100.0%

Note. Contingency table showing level of agreement with the statement payment system is a barrier to participation.

Table 4.8 Interest in Becoming an EN and Payment System as a Barrier

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 2.393a 2 .302 Likelihood Ratio 2.402 2 .301 Linear-by-Linear 2.359 1 .125 Association N of Valid Cases 92 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 10.09.

Table 4.9 Chi-squared test for Interest in Becoming and EN and Payment System as a Barrier

Although it appears the Ticket payment system is not a barrier to participation at a statistically significant level, the participants’ responses for individual items are important. Overall, they indicated the payment structure is a problem for their organization in terms of their understanding of the payment system, encouraging them to participate, and adequately reimbursing them for their services.

As shown in Figure 4.3, it was clear that the majority of respondents agreed that the current payment system is not adequately reimbursing providers. Equally clear was the fact that 40% of the respondents were not sure. Taken together, 91% of the organizations that responded had either a negative attitude towards the payment structure, or were not sure it is a feasible design. The majority (50%) of respondents (Figure 4.4) indicated that the present payment structure will not encourage their participation as an

EN. The large numbers of don’t know responses may indicate a lack of knowledge about the Ticket program and a potential opportunity for education and training for prospective

ENs. Additionally, Figure 4.5 shows that 41% of the respondents disagreed that the payment structure is clear and easy to understand.

Question 6 - Payment Structure Question 5 - Payment Structure for RTW Will Adequately Compensates Providers Encourage Participation as EN 60 60

50 50

40 40

30 30

20 20

10 10

0 0 Percent Percent Disagree Agree Don't Know Disagree Agree Don't Know Level of Agreement Level of Agreement

Figure 4.3 Structure of the Figure 4.4 Structure of the Payment System Payment System

Question 4 - Payment Structure is Clear and Easy to Understand 50

0 Percent Disagree Agree Dont' Know

Level of Agreement

Figure 4.5 Payment Structure is Clear and Easy to Understand

Providers were asked if they were adequately reimbursed for serving SSI and

SSDI populations. As indicated in Figures 4.6 and 4.7, the majority of the respondents 71

indicated that they are not adequately reimbursed for services to SSI and SSDI beneficiaries. A full one-third of providers expressed this opinion, while the majority in both questions (about 44%) indicated they did not know. The numbers of “don’t know” responses may demonstrate that many of the respondents are not currently serving

SSI/SSDI beneficiaries.

Question 7 - ENs Are Adequately Reimbursed for Returning SSI Beneficiaries to Work 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.6 ENs are Adequately Reimbursed for SSI Beneficiaries

Question 8 - ENs Are Adequately Reimbursed for Returning SSDI Beneficiaries to Work 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.7 ENs Are Adequately Reimbursed for SSDI Beneficiaries

Interest in Becoming an EN and Attitudes Toward the Payment Options and Issues

Payment Options

A major concern raised by the original designers of the Ticket program was that the payment system would not be attractive enough to recruit prospective ENs.

Therefore, one of the objectives of this research was to assess the providers’ attitudes and opinions regarding the payment options to determine if the design is a barrier to provider participation.

One theory explaining why prospective ENs were not participating is that they do not understand the payment options, and the difference between outcome only and milestone-outcome payments. As shown in Figure 4.8, while most of the organizations

(42%) reported that the clarity of the payment options was a not a barrier to participation,

30% did agree that the lack of understanding of the payment options is a barrier to their participation.

Question 13 - Not Participated Because Way Payment Options Are Calculated Are Unclear 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.8 Not Participated because Calculation of Payment Options is Unclear

Overall, 52% of the respondents responded they understood the difference between the two payment types, as seen in Figure 4.9. More importantly, the respondents were asked if they understood how the milestone-outcome payments are calculated because it is the most complex payment option. The responses were fairly evenly distributed between disagree, agree and don’t know at approximately 33% each. As shown in Figure 4.10, 67% of the respondents either disagreed, or indicated they do not know if they understand how these payments are calculated, while 33% reported they do understand. This is an indication that the payment options should be made simpler and easier to understand in order to increase provider interest, and possibly participation, in the program.

Question 1 -Understand the Difference Between Milestone/Outcome Payments 60

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.9 Understand the Difference between Milestone/Outcome Payments

Question 3 - Understand How Payments Are Calculated Under Milestone/Outcome Options 40

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.10 Understand How Payments are Calculated under Milestone/Outcome Options

Type of Payment Option That Would Encourage Participation

In a follow-up question, prospective ENs were asked to identify what type of payment option would encourage their participation as an EN. It is clear from Figure

4.11 that the majority of providers preferred a payment system that is a combination of hourly payments for services and a lump sum for milestones, such as employability evaluations and return to work plans (individual work plans). The second most preferred method for payment was reimbursement based only on hourly rates. Interestingly, no respondents selected the option that the current payment system is adequate.

Based on these responses to the payment options questions, SSA needs to take action to modify the milestone-outcome payment option to enable ENs to recoup the costs for initial services, such as screening and employability assessments that are staff intensive. This action was also supported by the final report of the Adequacy of

Incentives Advisory Group, which proposed a series of changes to the payment structure

to increase potential EN participation (AOI, 2004). At the present time the AOI’s proposed changes are under consideration at the Office of Management and Budget. It is

SSA’s hope that this regulatory change may be implemented by late spring of 2005

(Gerry, 2005).

Type of Payment Option that Would Encourage Participation as EN

70 60 50 40 30 20 Percentage 10 0 Hourly e Lump Sum Payment OK Both Hrly and Hrly Both Outcome Only Milestone/Outcom

Type of Payment Option

Figure 4.11 Type of Payment Option that Would Encourage Participation as an EN

5`- The fifth research hypotheses (H 0) addressed whether there was a relationship between attitudes towards payment issues/options and interest in becoming an EN. The initial principal components analysis was performed to identify the factors associated with payment issues/options from a combination of Ticket component questions

7,8,9,11,12,13, 14,15,16,19, 28,34,36.

Before performing the PCA, a test was done to determine the suitability of the data for factor analysis. The correlation matrix showed the presence of many coefficients of .3 or larger. The KMO was .729 and the Bartlett’s Test of Sphericity reached the level of statistical significance at .000, supporting the factorability of the correlation matrix. 76

The PCA revealed the presence of four factors with eigenvalues greater than one. These factors explained a total of 60% of the variance. An inspection of the scree plot revealed a clear break after the first component. To aid in the interpretation of these four components, a Varimax rotation was performed. The rotated component matrix revealed a simple structure that had many factors loading quite strongly (>.512) on the first component. (See Appendix G for tables). The interpretation of this component is consistent with assumptions of the relations of these questions to payment issues and options. Consequently, the decision was made to retain this variable and to conduct a chi-square test for independence.

The level of agreement with this statement was equally divided between disagree and agree as shown in Table 4.10. However, Table 4.10 also shows that 21% of the organizations that disagreed with this statement expressed no interest in becoming an EN.

Only 13% of the organizations both disagreed and expressed interest in being an EN.

In real numbers, this 21% represents approximately 300 potential ENs who are not participating in the program because of this issue. Although the following chi-square test of independence resulted in retaining the null hypothesis for this research question, the practicality of the numbers in this contingency table tell a story of a program that needs revision to its payment options to attract more providers.

Interest in Being an EN and Payment Issues and Options as Barrier

Level of Agreement Don't Disagree Agree Know Total Be an No Count 18 21 16 55 EN Expected Count 19.0 19.0 17.0 55.0 % within been 32.7% 38.2% 29.1% 100.0% % within Payment Issues 62.1% 72.4% 61.5% 65.5% and Options as Barrier % of Total 21.4% 25.0% 19.0% 65.5% Yes Count 11 8 10 29 Expected Count 10.0 10.0 9.0 29.0 % within been 37.9% 27.6% 34.5% 100.0% % within Payment Issues 37.9% 27.6% 38.5% 34.5% and Options as Barrier % of Total 13.1% 9.5% 11.9% 34.5% Total Count 29 29 26 84 Expected Count 29.0 29.0 26.0 84.0 % within been 34.5% 34.5% 31.0% 100.0% % within Payment Issues 100.0% 100.0% 100.0% 100.0% and Options as Barrier % of Total 34.5% 34.5% 31.0% 100.0%

Note. Contingency table showing level of agreement with statement payment issues and options are a barrier to participation.

Table 4.10 Interest in Being an EN and Payment Issues and Options as Barrier

Because this was a 2 x 3 contingency table, a Yates correction was not required.

In this chi-square test, X2(df=2, n=84). The calculated value of X2= .945 at the .05 α

level. The critical value of X 2 = 5.99. Additionally, the p value of .624 is greater than

the α value of .05 (Table 4.11). Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence was retained. It was concluded that the interest in becoming an EN was not related to attitudes towards the payment issues and options.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square .945a 2 .624 Likelihood Ratio .963 2 .618 Linear-by-Linear .000 1 .992 Association N of Valid Cases 84 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 8.98.

Table 4.11 Chi-Square Test for Interest in Being an EN and Payment Issues and Options

Although the interest in becoming an EN was found to be independent of concerns regarding payment issues, respondents expressed clear concerns in their responses to several important questions. These questions addressed Ticket program areas that were identified as critical issues based on a review of the literature, personal communication with the researcher, and the knowledge and experience of the researcher.

Payment Issues

Substantial Gainful Activity (SGA) as a Criteria for Payment

The level of cash benefit that triggers payments to ENs has also been a topic of discussion among providers and policy makers. The present system requires that the beneficiary be at a zero cash benefits status in order for payments to be made to ENs.

Prospective ENs were asked to respond questions regarding what criteria should be used for triggering payments to ENs, and whether they were not participating because of the use of SGA as a criteria for payment.

As shown in Figure 4.12, providers agreed that using SGA as the level for reimbursement is not optimal. In fact, 71% agreed that they should be reimbursed for

return to work at less than the current SGA level. Only 4% disagreed, while approximately one-quarter indicated they don’t know.

Question 14 - Providers Should Be Reimbursed for RTW at Less Than SGA 80

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.12 Providers Should be Reimbursed for RTW at Less Than SGA

Some prospective ENs have reported they are not participating in the Ticket program because it is not financially feasible for their clients, who are harder to serve, and never reach SGA. As anticipated, 44% agreed that they are not participating because many of their clients never reach SGA, as seen in Figure 4.13. Their responses to this question were surprising because almost 34% disagreed with this statement. These responses demonstrate that SGA level wages are not a standard that will enable providers to benefit financially from this program because many of their clients never reach the

SGA level of income. For the one-third of respondents who disagreed, it is clear that

SGA is not a factor in their decision not to participate in the Ticket program. Over 20% of the respondents indicated they do not know if they are not participating because their

clients never reach SGA, and do not know if they should be reimbursed for return to work at less than an SGA level of earnings. These responses may indicate that many of the prospective ENs are not currently serving SSI/SSDI beneficiaries or they do not have much knowledge about the Ticket program.

Question 12 - Not Participating Because Many Clients Never Reach SGA 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.13 Not Participating Because Many Clients Never Reach SGA

Whether ENs felt that a sliding scale would be more financially feasible versus zero cash benefits or SGA was also explored. As shown in Figure 4.14, responses to this question were interesting in that 39% agreed with this statement, and 42% did not know.

There may be several explanations for the “do not know” response to this question. It could be because the prospective ENs may not familiar with a sliding scale payment system. It may be that since it was not defined it was unclear, or they may not be working with SSI/SSDI beneficiaries so they have no knowledge of SGA issues.

Question 36 - Payments Should Be Based on Sliding Scale Not 0 Cash Benefits or SGA 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.14 Payments Should be Based on Sliding Scale not 0 Cash Benefits or SGA

Beneficiaries who are working in supported employment programs are usually earning less than SGA. Their ability to benefit from the Ticket program may be adversely affected by the SGA criteria. Consequently, providers were asked if the Ticket program was useful to them if their clients were involved in supported employment. As shown in Figure 4.15, 42% of the respondents disagreed with this statement, while 34% agreed. The majority of providers believe the Ticket program would be useful, but one- third feels that their organization may not benefit from the current payment system if they primarily serve clients who are engaged in supported employment.

Question 11 - Ticket Program Not Useful Many Clients in Supported Employment 50

Percent 0 Disagree Agree Don't Know Level of Agreement

Figure 4.15 Ticket Program Not Useful Many Clients in Supported Employment

Inequities in Payment for Serving SSI versus SSDI Beneficiaries

The Adequacy of Incentives Advisory Group (2003) was initially established to address the issue of increasing employment of hard to serve populations through the

Ticket program. Included in the AOI group’s work was an investigation into the inequities in the adequacy of payments to ENs, and concerns about the difference in reimbursement levels for SSI and SSDI beneficiaries. Anecdotal information from providers and disability advocates raised questions regarding the inequities in reimbursement between the two populations. Disability advocates and legislators voiced concern about the potential for “creaming” to take place, meaning that providers would choose to serve SSDI beneficiaries over SSI beneficiaries because the reimbursement was larger for SSDI beneficiaries, and with a prior work history, they are typically easier to re-employ at lower costs for services. Five questions were asked soliciting the attitudes of ENs toward the difference between SSI and SSDI payment for services.

It is clear from Figure 4.16 that 31% of the providers agreed that the difference in reimbursements prevents them from serving the harder to serve clients. Equally 83

compelling is the fact that 44% of the providers indicated they do not know if the differences between the two groups would prevent them from providing services. This information indicated that SSA has a window of opportunity to implement changes that will encourage serving SSI beneficiaries, and perhaps, greater participation by providers.

In addition, as seen in Figure 4.17, 55% of the providers agreed that the payment structure for serving SSI/SSDI clients should be more equitable. The numbers of don’t know responses may be an indication that those respondents are not currently serving

SSI/SSDI beneficiaries

Question 2 - Difference in Amount of Payment for SSI/SSDI Prevent Serving SSI 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.16 Differences in Amount of Payment for SSI/SSDI Prevent Serving SSI

Question 32 - Payments Should Be The Same Regardless of SSI/SDI 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.17 Payments Should Be the Same Regardless of SSI/SSDI

Many providers serve Ticket holders who require long-term services or supports.

This has been perceived as a barrier to participation because they cost more to serve.

Consequently, providers were asked if ENs were adequately reimbursed for SSI clients who typically require intensive services. (This question was a restated version of question 7 to measure consistency of responses.) As shown in Figure 4.18, a plurality of respondents (48%) agreed that ENs are not adequately reimbursed for serving SSI clients, while a small proportion (13%) disagreed. The fact that 40% of the respondents do not know if ENs are adequately reimbursed may be an indication that they are not presently serving SSI beneficiaries. Under the current payment structure, the don’t know group are unlikely to be enticed to participate and serve this harder to serve population as long as providers believe that serving SSI ticket holders is not financially feasible.

Question 9 - ENs Are Not Adequately Reimbursed for SSI Clients 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.18 ENs Are Not Adequately Reimbursed for SSI Clients

To further clarify their position, providers were also specifically asked if serving ticket holders who need long-term supports was a barrier to participation. As shown in

Figure 4.19, 52% of the providers indicated they agreed with this statement. However, those who disagreed and those who did not know were almost equally divided at 23% and 25% respectively. The level of agreement with this statement confirms that for these respondents, the payment options available for serving the harder to serve population are a barrier to participation. The numbers who don’t know may indicate that these respondents are not familiar with serving SSI beneficiaries because they are not currently working with them. For those who disagreed, other funding sources may be supporting the services they provide to this population.

Question 24 - Serving Ticket Holders Who Need Long Term Supports is a Barrier 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.19 Serving Ticket Holders Need Long Term Supports is a Barrier

The results of this survey revealed that the differences in payments for SSI and

SSDI beneficiaries are perceived as a financial and service delivery dilemma for providers. As shown in Figure 4.20, 49% of the respondents agreed with the statement

that the discrepancy between SSI and SSDI reimbursements creates a financial and

service delivery dilemma, while only 13% disagreed. Nearly 40% did not know if the

difference in the payment structures is a financial and service delivery dilemma. The

“don’t know” responses may indicate that these respondents are not presently serving

SSA beneficiaries and are not familiar with the payment structure.

Question 40 - Difference in Payments is Financial and Service Delivery Dilemma 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.20 Difference in Payment is Financial and Service Delivery Dilemma

Overpayments

The issue of overpayments is a pertinent topic because payment to ENs cannot take place if a beneficiary is in an overpayment status. As anticipated, many of the organizations (almost 43%) agreed that the possibility of overpayment was an issue, while 11% disagreed (Figure 4.21). The fact that 46% indicated they did not know may have shown that many of the respondents may not be serving SSA beneficiaries, or that they didn’t know at the time of the survey that this aspect of the Ticket program is problematic.

Question 19 - Possibility of Overpayment is Issue 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.21 Possibility of Overpayment is Issue

Sharing Payments with SVRAs and Other Providers.

The differences between the payment structure for providers versus the SVRAs, and the controversy surrounding sharing payments if a beneficiary is referred to another provider for other support services were also explored. As demonstrated in the results related to SGA and SSI/SSDI payment differences, providers believe they are at a financial disadvantage when providing services to the harder to serve populations. In order to explore attitudes toward the need for cost reimbursement as a funding mechanism for ticket holders, providers were asked if they should be reimbursed for the cost of services like the SVRA. As shown in Figure 4.22, 64% of the respondents agreed with this statement. The numbers of the don’t know responses may indicate these respondents may not be serving clients who are SSA beneficiaries or may not be serving clients who have been referred by the SVRA.

Question 28 - ENs Should Be Reimbursed on Cost Reimbursement Basis Like SVRA 70

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.22 ENs Should be Reimbursed on Cost Reimbursement Basis Like SVRA

A related problem is the sharing of reimbursements if a provider refers a ticket holder to another agency for ancillary services. The current guidelines indicate that reimbursement is calculated based on the percentage of services provided by each entity.

However, it is unclear how the percentage will be determined by Maximus, who has been given that responsibility by SSA. Consequently, prospective ENs reported (both in anecdotal comments and testimony) they are confused and apprehensive about the need to refer to outside vendors for additional services under the Ticket program, and how they will be reimbursed for these services. Rather than encouraging the development of an employment network to serve the ticket holder, this confusion may be preventing providers from participating, and more importantly, reaching out to other service providers who can provide important supplemental services. Consequently, providers were asked if they agreed that payments for services should be shared if a ticket holder is referred to other agencies for ancillary services related to employment. Although 23% of

the respondents indicated they did not know, 47% agreed that it was appropriate to share reimbursements if a ticket holder is referred for services the provider cannot offer, as shown in Figure 4.23.

Question 23 - Agree Payment for Services Should Be Shared 60

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.23 Agree Payment for Services Should be Shared

Interest in Becoming an EN and Delays in Reimbursement

Prospective ENs have reported that many believe they cannot serve ticket holders because it will take too long to be reimbursed for services. Two questions were asked which explored this issue. The first asked if waiting 45 days or more for reimbursement would be a financial hardship. The second asked if the perceived length of time to be reimbursed was a barrier to participation.

These two questions not only measured the consistency of responses, but also determined to what extent this issue was a barrier to participation. As shown in

Figure 4.24, 58% of the respondents agreed that waiting for reimbursement was a problem, while the same percentage agreed they have not participated because of their concern regarding delays in reimbursement, as shown in Figure 4.25.

Question 15 - Waiting for 45 Days or More for Reimbursement is Hardship 70

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.24 Waiting for 45 Days or More for Reimbursement is Hardship

Question 16 - Not Participated Because Concerned Re: Length of Time for Payment 70

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.25 Not Participated Because Concerned Re: Length of Time for Payment

The responses to these questions clearly demonstrated that SSA might be able to increase participation if these organizations can be assured of timely reimbursements.

This issue is critical because 48% of the respondents were non-profits, while 16% were sole proprietors. Positive cash flow for these organizations is essential to their continuing viability. As a result, they are not able to carry thousands of dollars in receivables due to delayed payments as are being reported by current ENs to the Ticket Advisory Panel

(Forand, 2004; Webb, 2003). 91

When organizations were asked directly if they had not participated because of anticipated delays in reimbursement, the respondents were evenly divided (Figure 4.26).

One-third of the respondents disagreed, indicating that anticipated delays were not a barrier to participation. One-third agreed that anticipated delays are a barrier, and one- third reported they “don’t know.” The fact that 33% of the respondents felt delays in reimbursement are an issue is a situation that SSA can readily address in order to increase recruitment of ENs. The number of prospective ENs who replied don’t know may be because they are not currently serving SSI/SSDI beneficiaries. Delays in reimbursement will make it difficult for this program to be viable because it is competing for private and non-profit vendors with systems such as Workers’ Compensation and Long Term

Disability where reimbursements for services usually occur within 30-45 days.

Question 34 - Not Participated Because Concerned if MoreThan 60 Days for Payment 40

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.26 Not Participated Because Concerned if More than 60 Days for Reimbursement

6 The sixth research hypotheses (H 0) addressed whether interest in becoming an

EN was independent of attitudes toward delays in reimbursement. The null hypothesis

was that there was no relationship between attitudes towards delays in reimbursement as a barrier to participation and the interest in becoming an EN.

Because this was a 2 x 3 contingency table, a Yates correction was not required.

In Table 4.12, X2(df=2, n=94). The calculated value of X2= 2.651 at the .05 α level. The critical value of X 2 = 5.99. Additionally, the p value of .266 is greater than the α value of

.05. Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence was accepted. It was concluded that the interest in becoming an EN is not related to attitudes toward delays in reimbursement.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 2.651a 2 .266 Likelihood Ratio 2.868 2 .238 Linear-by-Linear Association .223 1 .637 N of Valid Cases 94 a. 1 cells (16.7%) have expected count less than 5. The minimum expected count is 4.21.

Table 4.12 Chi-Square Test for Interest in Becoming an EN and Waiting 45 Days for Payment

As shown in Table 4.13, 23% of those who agreed that waiting for payment was a barrier were in interested in becoming an EN. Conversely, 20% of those who disagreed were also not interested in becoming an EN. From a standpoint of practical significance, if 23% is representative of the rest of the population surveyed, improving the timeliness of reimbursements may add approximately 300-400 additional ENs to the Ticket program. Although no statistical significance was found, Figures 4.24 and 4.25 did show

that respondents do feel this issue was a barrier to participation. Explanations for this apparent disparity are discussed in detail in the conclusions and implications section.

Interest in Becoming and EN and Waiting 45 Days for Payment as Barrier

Level of Agreement Don't Disagree Agree Know Total Be an No Count 19 32 10 61 EN Expected Count 18.2 35.0 7.8 61.0 % of Total 20.2% 34.0% 10.6% 64.9% Yes Count 9 22 2 33 Expected Count 9.8 19.0 4.2 33.0 % of Total 9.6% 23.4% 2.1% 35.1% Total Count 28 54 12 94 Expected Count 28.0 54.0 12.0 94.0 % of Total 29.8% 57.4% 12.8% 100.0%

Note. Contingency table showing level of agreement with statement waiting 45 days for reimbursement is a barrier to participation.

Table 4.13 Interest in Becoming an EN and Waiting 45 Days for Payment as Barrier

Interest in Becoming and EN and the Relationship between SVRAs and ENs

Anecdotal information, public testimony at the Ticket Advisory Panel Meeting in

August 2002 (Expediter, 2002; Zwirn, 2002; Rehabilitation Partnerships, 2002), and the

Ticket Advisory Panel (2003) have recently expressed concern regarding Transmittal 17 and its impact on the recruitment of ENs. Three areas of concern were raised regarding this issue: 1) were ENs fearful of losing SVRA funding for clients they currently serve;

2) were ENs concerned they were competing with SVRAs for the same client pool; and

3) was it a barrier to participation if SVRAs mandated that a ticket be deposited with the state agency if the client is referred for support services.

Competing with SVRAs for Ticket Holders and Loss of Funding

The responses of the participants were evenly dispersed between those who agreed (43%) and those who disagreed (41%) with concern about competing with the state agency for clients, as shown in Figure 4.27. Respondents who indicated that they did not know about this potential conflict totaled only 14%. The number of respondents who agreed with this statement demonstrated that some concern still lingers about the ramifications of competing with a funding source for the same clients.

Question 21- Concerned About Competing for Clients with State Agency 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.27 Concerned about Competing for Clients with State Agency

Prospective ENs have voiced concerns about competing with SVRAs for ticket holders. Some providers fear they may lose referrals, and thus revenue, if they are in competition for the same clients. When they were asked about competition with the

SVRAs and the possible loss of funding, the responses were almost equal between those potential ENs (38%) who were not concerned about losing SVRA funding, and those

(35%) who were concerned, as shown in Figure 4.28. Any level of concern about this

issue in the Ticket program should be viewed as unacceptable, and is a major barrier to

participation for providers.

Question 26 - Competing for Ticket Clients May Result in Losing SVRA Funding 40

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.28 Competing for Ticket Clients May Result in Losing SVRA Funding

7 Based on the response shown to these questions, the research hypotheses (H 0) addressed whether attitudes toward the contractual relationship between ENs and SVRAs influenced the interest in being an EN. The null hypothesis was attitudes toward the contractual relationship between ENs and SVRAs was independent of the interest in becoming an EN.

A principal components analysis was performed to identify the factors associated with the contractual relationship between ENs and SVRAs. Before performing the PCA,

a test was done to determine the suitability of the data for factor analysis. The correlation

matrix showed many coefficients of .3 or larger and the component matrix showed four

factor loadings at .512 or greater. The KMO was .650 and the Bartlett’s Test of

Sphericity reached the level of statistical significance at .000. The PCA revealed the

presence of only one factor with an eigenvalue greater than one. This eigenvalue explained 46% of the variance. The scree plot showed a break at the first component. A simple structure had been identified,. Therefore, a Varimax rotation could not be performed. (See Appendix G for tables.)

The null hypothesis for the related research was there is no relationship between concern about competing with the SVRA for clients or loss of funding from SVRAs and the interest in becoming an EN. The principal component that was retained includes both variables. Because this was a 2 x 3 contingency table, a Yates correction was not required. In this chi-square test, X2(df=2, n=96). The calculated value of X2=1.083 at the

.05 α level. The critical value of X 2 = 5.99. Additionally, the p value of .582 is greater than the α value of .05 (Table 4.14). Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence could not be rejected. Since the null hypothesis was retained, it was concluded that the interest in becoming an EN was independent of attitudes towards the concerns about competing with, or loss of funding from, SVRAs.. Interestingly, Table 4.15 shows that the responses to this issue were very evenly distributed across the question. The normalcy of this distribution further supports the finding from the chi-square test of independence.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 1.083a 2 .582 Likelihood Ratio 1.072 2 .585 Linear-by-Linear .010 1 .920 Association N of Valid Cases 96 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 9.97.

Table 4.14 Chi-Square Test for Interest in Becoming an EN and Loss of Funding

Interest in Becoming an EN and Loss of Funding and Competition with SVRAs

Level of Agreement Don't Disagree Agree Know Total Be an EN No Count 23 20 20 63 Expected Count 21.7 22.3 19.0 63.0 % within been 36.5% 31.7% 31.7% 100.0% % within Loss of Funding and Competition 69.7% 58.8% 69.0% 65.6% % of Total 24.0% 20.8% 20.8% 65.6% Yes Count 10 14 9 33 Expected Count 11.3 11.7 10.0 33.0 % within been 30.3% 42.4% 27.3% 100.0% % within Loss of Funding and Competition 30.3% 41.2% 31.0% 34.4% % of Total 10.4% 14.6% 9.4% 34.4% Total Count 33 34 29 96 Expected Count 33.0 34.0 29.0 96.0 % within been 34.4% 35.4% 30.2% 100.0% % within Loss of Funding and Competition 100.0% 100.0% 100.0% 100.0% % of Total 34.4% 35.4% 30.2% 100.0%

Note. Contingency table showing level of agreement with statement that loss of funding and competition with SVRAs are barriers to participation.

Table 4.15 Interest in Becoming an EN and Loss of Funding and Competition with SVRAs

Assignment of Tickets to the SVRA.

As shown in Figure 4.29, the majority of respondents (52%) indicated that they did not know that a ticket must be deposited or shared with SVRA if the beneficiary is referred for services by the EN. This finding further illustrates the fact that many potential ENs are not fully aware of the rules and regulations governing the referral of

SSI/SSDI beneficiaries to SVRA by ENs. To increase participation in the Ticket program, extensive education and training of potential ENs needs to take place so that lack of knowledge or misinformation about the program is not a barrier to participation.

Question 20 - SVRA Has Required Ticket Must Be Deposited with State Agency 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.29 SVRA has Required Ticket Must Be Deposited with State Agency

Potential Loss of Other Funding Sources

The majority of providers (59%) disagreed with the statement that participating in the Ticket program would result in the loss of other types of funding such as Medicaid,

US Department of Labor, or other vocational rehabilitation funding such as Veteran’s

Administration dollars. Many advocates, especially those who serve people with chronic psychiatric disabilities, have raised this issue because their clients need community supports, such as subsidized housing and transportation. For the respondents to this survey, the fear of losing other sources of funding does not appear to be a factor in the decision not to participate as an EN, as shown in Figure 4.30.

Question 49 - Not Participating Because Will Lose Other Funding 70

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.30 Not Participating Because Will Lose Other Funding

Interest in Being an EN and EN Credentials/Requirements

Staffing Requirements

The issue of staffing requirements and credentials as required by the Ticket program were explored to determine if they are barriers for some providers. Providers were surveyed relative to staffing credentials and requirements as specified in the EN application. As Figure 4.31 shows, 66% of the respondents disagreed with this statement, indicating that the qualifications of their staff were not a barrier to participation. Only a small proportion, less than 10%, reported that the staff qualifications were a barrier. As

100

shown in Figure 4.32, most of the respondents agreed that they employ professionals who are qualified to serve ticket holders under the stipulations of the EN application. The numbers of “don’t know” responses reinforced the need for further education of prospective ENs.

Question 22 - Not Participated Because Staff Do Not Have Credentials to Serve Ticket Holders 70

0 Percent Disagree Agree Don't Know

Level of Agreement Figure 4.31 Not Participated Because Staff Do Not Have Credentials to Serve Ticket Holders

Question 27 - Organization Employs Staff Who Are Qualified to Serve Ticket Holders 70

0 Percent Disagreed Agreed Don't Know

Level of Agreement Figure 4.32 Organization Employs Staff Who are Qualified to Serve Ticket Holders

101

EN Application

The EN application is an issue that was identified through anecdotal information as a potential barrier to participation. As seen in Figure 4.33, prospective ENs were asked if the application was too time consuming or too cumbersome to complete. Thirty- seven percent of the providers agreed that the application is time consuming and too cumbersome, while 28% disagreed. The number of providers who responded “don’t know” equaled 35%, and represented approximately 400 providers in this study. This result indicated that they have not examined the application, which may be a demonstration of a lack of interest, or lack of information about the program.

Question 25 - Not Participated Because Application Cumbersome /Time Consuming 40

0 Percent Disagree Agree Don't Know . Level of Agreement

Figure 4.33 Not Participated Because Application Cumbersome/Time Consuming

Interest in Being an EN and EN Reporting Requirements

Tracking Pay Stubs for 60 Months

Many providers have expressed concerned about the administrative burdens that the Ticket program may impose on their organizations. Statements made in the open-

102

ended sections of the survey also corroborated this problem. Both disability advocates and providers believed that collecting pay stubs for 60 months following placement was an administrative burden for the providers and an invasion of privacy for the ticket holder. As shown in Figure 4.34, nearly 63% of the providers agreed that this requirement of the Ticket program is a barrier to participation. The second highest response, don’t know, may be an indication that these providers are not familiar with the specific details of documenting employment for SSA beneficiaries who have returned to work under the Ticket program.

Question 50 - Collecting Pay Stubs for 60 Months is Barrier to Participation 70

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.34 Collecting Pay Stubs for 60 Months is Barrier to Participation

When the SSA changed this regulation, this concern became less prevalent according to anecdotal information provided by both current and prospective ENs.

However, SSA recently replaced this requirement with an employment verification procedure that is just as onerous. It requires ENs to provide more detailed information such as year to date earnings (Harles, 2005; Petkauskos, 2005). With this procedural change, the number of providers becoming ENs will not significantly increase.

103

Providers were asked if they had not participated in the Ticket program because it

is difficult to track pay stubs. This question was included to measure the consistency of

responses related to this topic. Nearly 53% of the respondents indicated that tracking pay

stubs was a barrier to participation (Figure 4.35). in addition, public testimony at the

Ticket Advisory Panel Quarterly Meetings has revealed that beneficiaries find the

tracking of employment information as intrusive, and at times, demeaning (Nelson, 2005;

Petkauskos, 2005).

Question 29 - Not Participated Because Difficult to Track Pay Stubs 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.35 Not Participated Because Difficult to Track Pay Stubs

The eighth research hypotheses further explored whether the relationship between

interest in becoming an EN was independent of attitudes toward EN reporting

requirements for pay stubs, or evaluation criteria. Because of anecdotal information and

the literature review, the decision was made to evaluate these two variables separately.

8a The first hypothesis (H 0) was there is no relationship between the attitudes toward EN reporting requirements for pay stubs and the interest in becoming an EN. In

Table 4.16, X2(df=2, n=92). The calculated value of X2=.910 at the .05 α level. The 104

critical value of X 2 = 5.99. Additionally, the p value of .634 is greater than the α value of

.05. Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence could not be rejected. Since the null hypothesis was retained, it was concluded that the interest in becoming an EN was not related to attitudes toward the tracking of pay stubs. However, 42% of the respondents who expressed no interest in becoming an EN agreed that tracking pay stubs for 60 days was a barrier to their participation in the program, as shown in Table 4.17.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square .910a 2 .634 Likelihood Ratio .886 2 .642 Linear-by-Linear Association .563 1 .453 N of Valid Cases 92 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.38.

Table 4.16 Chi-Square Test for Interest in Becoming an EN and Tracking Pay Stubs

Interest in Being an EN and Tracking Pay Stubs as a Barrier

Level of Agreement Don't Disagree Agree Know Total Be an No Count 8 39 12 59 EN Expected Count 9.6 37.8 11.5 59.0 % of Total 8.7% 42.4% 13.0% 64.1% Yes Count 7 20 6 33 Expected Count 5.4 21.2 6.5 33.0 % of Total 7.6% 21.7% 6.5% 35.9% Total Count 15 59 18 92 Expected Count 15.0 59.0 18.0 92.0 % of Total 16.3% 64.1% 19.6% 100.0%

Note. Contingency table showing level of agreement with statement that tracking pay stubs is a barrier to participation.

Table 4.17 Interest in Being an EN and Tracking Pay Stubs as a Barrier

105

8b The second null hypothesis (H 0) was related to the criteria for evaluating ENs.

The null hypothesis stated that interest in becoming an EN was independent of the the

attitudes toward the evaluation criteria of ENs. Because this was a 2 x 3 contingency

table, a Yates correction was not required. In Table 4.18, X2(df=2, n=95). The calculated value of X2=3.716 at the .05 α level. The critical value of X 2 = 5.99. Additionally, the p

value of .156 is greater than the α value of .05. Therefore, at an alpha level of .05 with a

one-tailed test, the statistical hypothesis of independence could not be rejected. It was

concluded that the interest in becoming an EN was not related to attitudes toward EN

evaluation criteria. It should be noted that while 40% of the respondents agreed that the

evaluation criteria for ENs is a potential barrier to participation (Table 4.19), 49%

disagreed with this statement. However, of the 42 respondents who disagreed, only 17

(18%) expressed an interest in becoming an EN.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 3.716a 2 .156 Likelihood Ratio 4.215 2 .122 Linear-by-Linear 2.796 1 .095 Association N of Valid Cases 95 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.21.

Table 4.18 Chi-Squared Test for Interest in Becoming an EN and Evaluation of ENs

106

Interest in Becoming an EN and Evaluation of ENs

Level of Agreement Don't Disagree Agree Know Total Be an EN No Count 25 24 13 62 Expected Count 27.4 24.8 9.8 62.0 % of Total 26.3% 25.3% 13.7% 65.3% Yes Count 17 14 2 33 Expected Count 14.6 13.2 5.2 33.0 % of Total 17.9% 14.7% 2.1% 34.7% Total Count 42 38 15 95 Expected Count 42.0 38.0 15.0 95.0 % of Total 44.2% 40.0% 15.8% 100.0%

Note. Contingency table showing level of agreement with statement that the criteria for evaluating ENs is a barrier to participation.

Table 4.19 Interest in Becoming an EN and Evaluation of ENs

Many additional factors were addressed by this survey in terms of the reporting, staffing and support needs for potential ENs. These factors are summarized in the following discussion and bar charts.

Computer Tracking Program.

The survey asked potential ENs if the lack of computer software for tracking the status of ticket holders is a barrier to participation, and if having a software program available would encourage participation in the Ticket program. A plurality of providers

(43%) indicated that the lack of a computer tracking program is not the reason for not participating (Figure 4.36). However, 32% agreed that the lack of software is a barrier to participation, while 25% of the respondents don’t know.

107

Question 30 - Not Participated Because No Computer Tracking Program 50

0 Percent Disagree Agree Don't Know Level of Agreement Figure 4.36 Not Participated Because No Computer Tracking Program

As shown in Figure 4.37, 42% of the respondents agreed that they would participate if a software-tracking program were available. This level of agreement demonstrated that developing a tracking program for reporting ticket holder status might increase participation in the Ticket program. However, as seen in Figure 4.38, a plurality of respondents (41%) indicated that the lack of a software program is not the reason their organization is not participating in the Ticket program, while 34% of the respondents agreed that this is a barrier to their participation. Because one-third of the respondents agreed with this statement, the results may indicate that the development of a software program specific to tracking financial records for the Ticket program may encourage greater participation by providers by reducing administrative burdens. This may especially be true for non-profit providers and sole proprietors who have limited resources and support staff that can be dedicated solely to the Ticket program.

108

Question 31 - Would Participate if Could Use Software Program for Tracking Participants 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.37 Would Participate if Could Use Software Program for Tracking Participants

Question 33 - Not Participated Because No Program to Track Financial Records 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.38 Not Participated Because No Program to Track Financial Records

Ability to Serve Ticket Holders

The providers were asked if they were able to offer services to any ticket holder who requested assistance. As shown in Figure 4.39, a plurality agreed they are able to serve all ticket holders who would contact them. However, 33% disagreed, while 21% indicated they don’t know. The fact that one-third of the respondents indicated they can not provide services to all ticket holders who contact them may indicate that some 109

providers serve specific populations, and may not be the “best fit” for employment services for some SSI/SSDI beneficiaries. This may underscore the need of ENs to engage in comprehensive initial screening of ticket holders to ensure the best match between ENs and beneficiaries. However, comprehensive initial screening increases the upfront cost to participate in the program for prospective ENs.

Question 39 - Organization Can Provide Services to Any Ticket Holder Who Requests It 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.39 Organization Can Provide Service to Any Ticket Holder Who Requests It

Understand the Ticket Program and Requirements

Prospective ENs were asked if they have not participated because they do not understand the Ticket program or how it works. The responses were fairly evenly distributed between disagree, (46%) and agree (44%), as seen in Figure 4.40. Only 10% of the providers responded don’t know. The respondents who disagreed apparently understand the program, but may have chosen not to participate. For the providers who

110

agreed with this statement, further education regarding the Ticket program may be warranted in order to increase participation.

Question 35 - Not Participated Because Do Not Understand Program 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.40 Not Participated Because Do Not Understand Program

Lack of Defined Criteria for Evaluating ENs and the Dispute Resolution Process.

Providers have raised concerns regarding the criteria for the evaluation of ENs, and the dispute resolution process. The evaluation criteria have not been clarified, and both prospective and current ENs are unsure about how often they will be evaluated, and with what performance measures. Several questions explored this topic along with the dispute resolution process, which is related to operating as an EN and evaluating disputes over EN performance or sharing of payments.

Criteria for Evaluation of ENs.

The opinions of prospective ENs about what criteria should be used for evaluating the performance of ENs were solicited. As shown in Figure 4.41, an overwhelming

111

number of respondents (74%), indicated that the criteria for evaluating ENs should be based on the type of placements, level of work, and the population served.

Question 38 - Criteria for EN Evaluation - Placement, Level of Work, Population Served 80

0 Percent Disagree Agree Don't Know

Level of Agreement Figure 4.41 Criteria for EN Evaluation: Placement, Level of Work, Population Served

Prospective ENs were also asked whether the criteria for successful EN performance should be the number of placements. Although 43% of the providers agreed with this statement, 30% disagreed, while nearly the same amount responded don’t know, as seen in Figure 4.42. Those who disagreed most likely believe that a combination of factors, such as the type of placements, level of work, and the population served, would be the optimal criteria for evaluating the performance of EN, as evidenced by anecdotal comments on the survey.

112

Question 37 - Criteria for Evaluating ENs Should Be Number of Placements 50

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.42 Criteria for Evaluating ENs Should be Number of Placements

Dispute Resolution Process

The dispute resolution process has been perceived as a potential barrier to participation because the process was not initially defined clearly in the rules and regulations. Consequently, prospective providers assumed that the SSA would be the party to resolve any disputes between providers, or providers and beneficiaries.

Although, the dispute resolution process has become clearer as Maximus has begun implementing it, concerns are still evident. Providers were asked if the dispute resolution process of the Ticket program is unfair to ENs. They were also asked who should be responsible for resolving disputes between ENs, SVRAs and beneficiaries.

It is obvious from the responses that 64% of the providers don’t know if the process is unfair to ENs (Figure 4.43). The largest proportion of providers, 56%, agreed hat a third party, not Maximus or the SSA, should administer the dispute resolution process (Figure 4.44). The numbers of respondents who answered don’t know may

113

indicate their lack of knowledge of this aspect of the Ticket program and the current lack of information regarding the dispute resolution process.

Question 43 - Dispute Resolution Process Is Unfair to ENs 70

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.43 Dispute Resolution Process is Unfair to ENs

Question 46 - Third Party Should Be Responsible for Dispute Resolution 70

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.44 Third Party Should Be Responsible for Dispute Resolution

Interest in Being an EN and Upfront Funding/ Capitalization

The final research hypotheses addressed whether interest in becoming an EN was independent of the need for upfront funding or capitalization. Many providers have

114

expressed concern about the upfront costs for staffing the Ticket program. Current ENs

have testified that it is very labor intensive to screen the calls of ticket holders who are

looking for information (Expediter, 2002; Petkouskas, 2005; Webb, 2003; Zwirn, 2002).

Few non-profits or sole proprietors can afford to hire staff without the immediate promise of income to cover the costs of that additional staff. Consequently, the Ticket Advisory

Panel, Maximus and SSA have been exploring ways to provide upfront funds to enable more participation by providers. Prospective ENs were asked if up front funding would enable the provider to participate as an EN.

As shown in Figure 4.45, a majority, 67%, agreed that upfront funds would make it possible to become an EN. This information gives SSA the opportunity to examine how the payment options may be restructured to make upfront funding possible.

Although many providers agreed that upfront funding would encourage participation as an EN, it is clear from their response to the question shown in Figure 4.46, that the availability of low interest loans would not encourage greater participation as an EN.

Question 10 - Upfront Funds Enable

Participation as EN 80

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.45 Upfront Funds Enable Participation as EN

115

Become an EN If Low Interest Loans Were Available 100

0 Percent No Yes Agreement Figure 4.46 Become an EN if Low Interest Loans Were Available

A chi-squared test for independence was conducted for this hypothesis. The null

9 hypothesis (H 0) was the attitude toward the need for upfront funding was independent of

the interest in becoming an EN. Because this was a 2 x 3 contingency table, a Yates

correction was not required. In Table 4.21, X2(df=2, n=96). The calculated value of

X2=7.948 at the .05 α level. The critical value of X 2 = 5.99. Additionally, the p value of

.019 is less than the α value of .05. Therefore, at an alpha level of .05 with a one-tailed

test, the statistical hypothesis of independence was rejected. It was concluded that the

interest in becoming an EN was related to the need for upfront funding. However, the

Cramer’s V value of .288 indicated that the magnitude of the relationship was weak (See

Appendix G for Cramer’s V statistic). The high level of agreement for the need for

upfront funding impacted this result, as shown in Table 4.22.

116

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 7.948a 2 .019 Likelihood Ratio 11.102 2 .004 Linear-by-Linear .329 1 .566 Association N of Valid Cases 96 a. 1 cells (16.7%) have expected count less than 5. The minimum expected count is 3.44.

Table 4.20 Chi-Square Test for Interest in Becoming an EN and Need for Upfront Funding

Interest in Being an EN and Need for Upfront Funding

Level of Agreement Don't Disagree Agree Know Total Be an EN No Count 10 36 17 63 Expected Count 6.6 41.3 15.1 63.0 % of Total 10.4% 37.5% 17.7% 65.6% Yes Count 0 27 6 33 Expected Count 3.4 21.7 7.9 33.0 % of Total .0% 28.1% 6.3% 34.4% Total Count 10 63 23 96 Expected Count 10.0 63.0 23.0 96.0 % of Total 10.4% 65.6% 24.0% 100.0%

Table 4.21 Interest in Becoming an EN and Need for Upfront Funding

Additional Research Questions

Beyond the research hypotheses, additional important questions were asked based on the review of the literature and anecdotal information. The first was is there a relationship between the number of staff and the interest in becoming an EN? In order to facilitate data analysis, the continuous variable of number of staff was transformed into categorical level variables dispersed into quartiles. 117

Because this was a 2 x 4 contingency table, a Yates correction was not required.

In this chi-square test, X2(df=3, n=89). The calculated value of X2=2.682 at the .05 α

level. The critical value of X 2 = 7.81. Additionally, the p value of .443 is greater than

the α value of .05 (Table 4.22). Therefore, at an alpha level of .05 with a one-tailed test,

the statistical hypothesis of independence was accepted. It was concluded that the

interest in becoming an EN was independent of the total number of staff in an

organization. As shown in Table 4.23, organizations with four to twelve staff, or thirteen

to one hundred staff were more likely than others to express interest in participating in

the Ticket program. This information may assist Maximus and SSA in identifying a

potential target market in smaller providers.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 2.682a 3 .443 Likelihood Ratio 2.700 3 .440 Linear-by-Linear .003 1 .955 Association N of Valid Cases 89 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.31.

Table 4.22 Chi-Square Test for Interest in Being an EN and Total Number of Staff

118

Interest in Being an EN and Total Number of Staff

Total Staff 0 - 3 4 - 12 13 - 100 > 100 Total Be an No Count 13 17 16 12 58 EN Expected Count 15.0 15.0 14.3 13.7 58.0 % of Total 14.6% 19.1% 18.0% 13.5% 65.2% Yes Count 10 6 6 9 31 Expected Count 8.0 8.0 7.7 7.3 31.0 % of Total 11.2% 6.7% 6.7% 10.1% 34.8% Total Count 23 23 22 21 89 Expected Count 23.0 23.0 22.0 21.0 89.0 % of Total 25.8% 25.8% 24.7% 23.6% 100.0%

Note. Contingency table showing total number of staff in quartiles and interest in being an EN.

Table 4.23 Interest in Being an EN and Total Number of Staff

4.6.6 Demographics of the Respondents

Position Held by the Respondent

Based on the respondents’ self report, 50% were executive directors of their organization or agency, 21% were program managers responsible for vocational services, and 16% were vocational counselors who were given the responsibility to answer the survey on behalf of their executive director. Twenty-eight percent of the respondents did not answer this item. (See Appendix G for frequency table).

Type of Organization

Non-profit organizations comprised 50% of the respondents to the survey, while sole proprietors were the second largest grouping at 16%. The make up of other respondents included 11% “S” corporations, 10% State governments, 3% Limited

Liability Partnerships, 1%“C” corporations, 1% Tribal Governments, and no Limited

Partnerships, or Schools and Universities. (See Appendix G for frequency table).

119

Total Number of Clients Served

The distribution for number of clients served was positively skewed due to the statistical outlier of one organization that serves 30,000 people per year. The number of clients served ranged from none to 30,000 persons a year in the case of a major health care provider. The most frequently occurring number of clients served per year was 100.

One quarter of the respondents served up to 50 people annually, while 50% of the respondents served between 0-150. The majority of the organizations responding were relatively small entities that served 150 or fewer clients per year. (See Appendix G for frequency table).

Sources of Knowledge About the Ticket program

Respondents were asked to identify all sources of knowledge about the Ticket program. Table 4.24 shows their responses. The totals do not equal one hundred percent because some organizations learned about the Ticket program from multiple sources.

Based on this data, it was clear that Maximus’ presentations were providing information to approximately one-third of the respondents. However, the majority of organizations were getting their information regarding the Ticket program from professional conferences, journal articles and other rehabilitation providers. Maximus’ website appeared to be the least used source of information at the time of the study. This information may be able to assist SSA and Maximus in determining the direction of the future marketing efforts for the Ticket program.

120

Percent

Source of Information About the Ticket program Professional Conferences 52 Maximus Presentation 31 Maximus Web site 15 Maximus Mailing 22 Article about the Ticket 42 Other Rehabilitation Providers 39 Ticket holders 19

Table 4.24 Frequency Table for Sources of Information About the Ticket Program

4.6.7 General Knowledge of the Ticket program

The initial section of the survey, Section A, asked the respondents to rate their knowledge of the Ticket program, whether or not their organization was interested in becoming an EN, and the number of clients referred by the SVRA. The beginning of

Section A asked respondents to rate their knowledge on a four point scale: 1) Not at all knowledgeable; 2) Somewhat knowledgeable; 3) Very knowledgeable; and 4) Don’t know.

The five specific areas that were rated included the following: 1) the overall program; 2) payment options, 3) beneficiary status reporting requirements, 4) recordkeeping requirements, and 5) the requirements necessary to become an EN. The assumption was that the level of knowledge of the Ticket program would impact the decision to participate as an EN. In addition, an item in this section asked respondents to indicate if their organization was interested in becoming an EN.

Overall Knowledge of the Ticket program

As can be seen from Table 4.25, the majority, or 58% of the respondents, reported they were somewhat knowledgeable about the Ticket program, while those who reported

121

they had no knowledge or were very knowledgeable were almost evenly distributed at

21% and 20% respectively. The positive aspect of this finding is that 78% of the respondents are aware of the Ticket program. Conversely, since 21% of the participants have no knowledge of the program, Maximus and SSA need to evaluate their marketing and outreach in order to increase awareness and knowledge of the program among rehabilitation providers.

Level of Knowledge Frequency Percent Not at all 21 21.0 Somewhat 58 58.0 Very 20 20.0 Don’t know 1 1.0 Total 100 100.0

Table 4.25 Frequency Table for Overall Knowledge of the Ticket Program

A chi-square test of independence was performed to determine if the interest in being an EN was independent of the respondents’ overall knowledge of the Ticket program. Because this was a 2 x 4 contingency table, a Yates correction was not required. In this chi-square test, X2(df=3, n=96). The calculated value of X2=1.597 at the

.05 α level. The critical value of X 2 = 7.81. Additionally, the p value of .660 is greater than the α value of .05 (Table 4.26). Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of independence was accepted. It was concluded that the interest in becoming an EN was independent of the organization’s overall knowledge of the Ticket program.

122

Of the 96 respondents to this question, 16% reported they were very knowledgeable about the program, but were not interested in becoming an EN. Only five percent of the respondents were both very knowledgeable and interested in becoming an

EN. From Table 4.27, it appears that those organizations that had some knowledge of the program were more likely to not be interested in participating. This may be a clear indication that increased marketing and dissemination of success stories is necessary in order to improve perceptions regarding the Ticket program

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 1.597a 3 .660 Likelihood Ratio 1.954 3 .582 Linear-by-Linear Association .939 1 .332 N of Valid Cases 96 a. 2 cells (25.0%) have expected count less than 5. The minimum expected count is .34.

Table 4.26 Chi-Square Test for Interest in Being an EN and Overall Knowledge of the Ticket Program

Interest in Being an EN and Overall Knowledge of the Ticket Program

Overall Knowledge Don't Not at all Somewhat Very Know Total Be an EN No Count 12 35 15 1 63 Expected Count 12.5 36.8 13.1 .7 63.0 % within been 19.0% 55.6% 23.8% 1.6% 100.0% % within know1 63.2% 62.5% 75.0% 100.0% 65.6% % of Total 12.5% 36.5% 15.6% 1.0% 65.6% Yes Count 7 21 5 0 33 Expected Count 6.5 19.3 6.9 .3 33.0 % within been 21.2% 63.6% 15.2% .0% 100.0% % within know1 36.8% 37.5% 25.0% .0% 34.4% % of Total 7.3% 21.9% 5.2% .0% 34.4% Total Count 19 56 20 1 96 Expected Count 19.0 56.0 20.0 1.0 96.0 % within been 19.8% 58.3% 20.8% 1.0% 100.0% % within know1 100.0% 100.0% 100.0% 100.0% 100.0% % of Total 19.8% 58.3% 20.8% 1.0% 100.0%

Table 4.27 Interest in Being an EN and Overall Knowledge of the Ticket Program

123

Knowledge About Payment Options

A plurality of respondents answered this question as somewhat knowledgeable,

while the second largest response was not at all knowledgeable, as shown in Table 4.28.

The lack of knowledge regarding the Ticket program demonstrated that increased

education of potential ENs is critical for the future success of the programs.

Level of Knowledge Frequency Percent Not at all 39 39.0 Somewhat 47 47.0 Very 13 13.0 Don’t know 1 1.0 Total 100 100.0

Table 4.28 Frequency Table for Knowledge about Payment Options

A chi-square test of independence was performed to determine if the interest in

being an EN was independent of the respondents’ knowledge of the payment options.

Because this was a 2 x 4 contingency table, a Yates correction was not required. In Table

4.29, X2(df=3, n=96). The calculated value of X2=1.905 at the .05 α level. The critical

value of X 2 = 7.81. Additionally, the p value of .592 is greater than the α value of .05.

Therefore, at an alpha level of .05 with a one-tailed test, the statistical hypothesis of

independence was accepted. It was concluded that the interest in becoming an EN was

independent of the organization’s knowledge of the payment options. As shown in Table

4.30, sixty-seven percent of the respondents who indicated they had some knowledge of

payment options were not interested in becoming an EN. Fifty-six percent of the

124

respondents who reported they were either somewhat knowledgeable or very knowledgeable expressed interest in participating in the program.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 1.905a 3 .592 Likelihood Ratio 2.258 3 .521 Linear-by-Linear Association 1.764 1 .184 N of Valid Cases 96 a. 3 cells (37.5%) have expected count less than 5. The minimum expected count is .34.

Table 4.29 Chi-Square Test for Interest in Being an EN and Knowledge about Payment Options

Interest in Becoming an EN and Knowledge About Payment Options

Level of Knowledge about Payment Options Don't Not at all Somewhat Very Know Total Be an EN No Count 22 30 10 1 63 Expected Count 24.3 29.5 8.5 .7 63.0 % within been 34.9% 47.6% 15.9% 1.6% 100.0% % within know2 59.5% 66.7% 76.9% 100.0% 65.6% % of Total 22.9% 31.3% 10.4% 1.0% 65.6% Yes Count 15 15 3 0 33 Expected Count 12.7 15.5 4.5 .3 33.0 % within been 45.5% 45.5% 9.1% .0% 100.0% % within know2 40.5% 33.3% 23.1% .0% 34.4% % of Total 15.6% 15.6% 3.1% .0% 34.4% Total Count 37 45 13 1 96 Expected Count 37.0 45.0 13.0 1.0 96.0 % within been 38.5% 46.9% 13.5% 1.0% 100.0% % within know2 100.0% 100.0% 100.0% 100.0% 100.0% % of Total 38.5% 46.9% 13.5% 1.0% 100.0%

Table 4.30 Interest in Becoming an EN and Knowledge About Payment Options

Beneficiary Status Reporting Requirements

Prospective ENs have raised concerns regarding the paperwork and reporting requirements connected with the Ticket program. The former requirement to track pay stubs for a period of 60 months in order to receive reimbursement for services was a 125

significant issue. With a rules change, this requirement was changed within the past year

to reduce the burden of reporting requirements on the ENs. However, the burden remains

the same with the implementation of the employment verification form. The responses

related to knowledge in this area are summarized in Table 4.31. An interesting aspect of

these responses is the fact that none of the respondents answered “don’t know” to this

question. All respondents knew whether they had any level of knowledge of these

requirements. However, 52% of the respondents answered no knowledge at all. This

may be an indication that this group has not sought out information about reporting

requirements because they are not interested in the program. It may also confirm that

SSA needs to continue marketing and outreach to prospective ENs to increase interest.

Cumulative

Percent Frequency Percent Valid Not at all 52 52.0 52.0 Somewhat 37 37.0 89.0 Very 11 11.0 100.0 Total 100 100.0

Table 4.31 Frequency Table for Beneficiary Status Reporting Requirements

Record-keeping Requirements

Respondents were asked to rate their knowledge of record keeping requirements.

Ninety-one percent indicated they had either had no knowledge or only some knowledge of the specific record keeping requirements associated with the program, as shown in

Table 4.32. As with previous questions regarding level of knowledge about the Ticket

126

program, SSA has a clear opportunity to influence the attitudes and perceptions of providers since such a large majority of respondents are not aware of Ticket program record keeping requirements.

Level of Knowledge Frequency Percent Not at all 53 53.0 Somewhat 38 38.0 Very 6 6.0 Don’t know 1 1.0 Total 98 98.0 Total 100 100.0

Table 4.32 Frequency Table for Knowledge of Record Keeping Requirements.

Requirements for Providers to Become ENs

The study also explored the possibility that the application to become an EN, or specific aspects of the requirements for becoming an EN, were preventing prospective

ENs from participating. In order to identify the respondents’ attitudes toward these requirements, a general question about this area was initially asked and was followed by more specific questions later in the survey. Table 4.33 summarizes the organizations’ responses to this question. The majority of respondents (77%) indicated either no knowledge or some knowledge regarding the requirements to become an EN. These results indicated that the application process and staff qualifications or credentials may not be barriers to participation in the Ticket program.

Overall, in each area of knowledge regarding the Ticket program, a plurality or majority of respondents indicated either some knowledge or no knowledge at all. This 127

finding is significant in that these respondents indicated that they have some or little

knowledge, but have still made the decision not to participate in the program. Targeted

marketing that raises program awareness may impact providers’ knowledge and may

increase participation

Level of Knowledge Frequency Percent Not at all 42 42.0 Somewhat 35 35.0 Very 23 23.0 Total 100 100.0

Table 4.33 Frequency table showing overall knowledge about requirements to become an EN.

Number of Referrals from SVRA

Providers were asked what percentage of their clients were referred by the State

Vocational Rehabilitation Agency (SVRA) on an annual basis. As shown in Table 4.34,

the majority of providers, 46, reported that they receive no referrals from the State

agency. Thirty providers reported 1-25% of their annual referrals come from the State

agency, while only one provider received 100% of its referrals from the SVRA.

Percentage of Annual Referrals # of Providers 0 46 1-25 30 26-50 8 51-99 13 100 1

Table 4.34 Frequency table showing percentage of annual referrals from SVRAs and the number of providers in each group.

128

4.6.8 Other Disability Policy Issues Related to the Ticket to Work

In addition to questions specific to the Ticket program, this study also explored other SSA policy issues that might improve the program and result in greater participation by employment providers. Prospective ENs were surveyed regarding their trainings needs and several possible SSA policy changes.

SSA Policy Changes that Might be Beneficial to the Ticket Program

Referral Process

The prospective ENs were solicited to provide an opinion on overall policy changes that would facilitate the Ticket program. Initially, they were asked would they participate in the Ticket program if SSA implemented a referral process. As shown in

Figure 4.47, 40% of the respondents agreed that they would participate if SSA implemented a referral process. However, one-third of the providers disagreed, indicating that implementing a referral process alone would not be an incentive to participate. A large number of participants responded don’t know, which may indicate their lack of knowledge about the overall Ticket program., as shown in the responses to the level of knowledge questions previously discussed.

Question 41 - Would Participate if SSA Implemented Referral Process 50

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.47 Would Participate if SSA Implemented Referral Process

129

Providers were also asked to rank their preferred method of receiving referrals of ticket holders from the following options: 1) Maximus provides the list of ticket eligible beneficiaries on a password protected web site; 2) SSA representatives refer people directly to the provider; 3) files are reviewed to determine the feasibility and appropriateness for employment services with provider. Figure 4.48 displays the responses to each of these options. Many of the respondents, preferred receiving referrals from Maximus. The second most preferred method (18%) was for SSA representatives to refer directly to the organization, while the least preferred method was to personally review the files of the beneficiaries.

Most Preferred Method for Ticket Holder Referrals

File Review 3rd SSA 2nd 1st Maximus Method of ReferralMethod 0 10203040 Percentage of Respondents

Figure 4.48 Most Preferred Method for Ticket Holder Referrals

In addition to identifying the most preferred method for referrals, providers were also asked to select what types of information they would like to know at the time of referral and why. Figure 4.49 summarizes their responses. Most respondents selected medical history as their first choice for important information to receive at the time of referral. Work history was chosen by over 30% of the respondents as either the most 130

important, or second most important piece of information to receive at the time of referral. Following these two factors, education was selected as the third and fourth most important by a number of respondents. Gender was selected as the least important information to receive at time of referral by almost one-half of the respondents.

Important Information at Time of Referral

50 1st 40 2nd 3rd 30 4th 5th 20 6th

Number of RespondentsNumber 7th 10

0 Wk Hx MedHx Age Edu Gender Date of Last E Types of Information

Figure 4.49 Important Information at Time of Referral

The respondents were given the opportunity to explain why this information was important to their organization through open-ended comments on the survey. Providers reported that medical history and prior work history were the most critical information since they are integral to determining the employment potential of SSI/SSDI

131

beneficiaries. Additionally, they also indicated that collecting this type of information, as part of their employability assessment, is their standard practice, which is based on sound rehabilitation principles.

Time-limited Benefits

Prospective ENs were asked if SSA should provide time-limited benefits as an incentive and bridge for SSI/SSDI beneficiaries to return to work under the Ticket program. This model is similar to the benefit provisions for Long-Term Disability and

Workers’ Compensation recipients. As seen in Figure 4.50, 56% of the respondents agreed with this statement, while only 26% disagreed. A small number indicated they don’t know,which may indicate that those respondents were not familiar with the Social

Security disability process or have not served SSA beneficiaries in their programs.

Question 47 - SSA Should Provide Time Limited Benefits as Incentive/ Bridge to RTW 60

30 Percent 20

0 Disagree Agree Don't Know Level of Agreement

Figure 4.50 SSA Should Provide Time Limited Benefits as Incentive/Bridge to RTW

Provide a Ticket at Time of Application for Benefits if Presumed Eligible

As shown in Figure 4.51, 62% of the respondents agreed with the concept that a ticket to work should be provided to clients at the time of application for SSI/SSDI if the client is presumed to be eligible for benefits. This change in policy may increase the

132

success of the Ticket program by reducing the beneficiaries’ disincentives to work by creating an expectation of employment from the time of application. It may also increase provider participation since beneficiaries may be more motivated to work.

Question 18 - Provide Ticket at Time of Application for Benefits if Presumed Eligible 70

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.51 Provide Ticket at Time of Application for Benefits if Presumed Eligible

Providers were also asked whether SSA should refer potential beneficiaries to return to work services at the time of application for benefits. As shown in Figure 4.52,

51% of the respondents agreed, while 15%, disagreed, and 34% did not know. This change in policy could encourage participation by providers by increasing the motivation and availability of beneficiaries to return to work.

Question 45 - SSA Should Refer Beneficiary for RTW with Ticket at Application 60

0 Percent Disagree Agree Don't Know

Level of Agreement

Figure 4.52 SSA Should Refer Beneficiary for RTW with Ticket Application

133

Make the Ticket Mandatory for SSI/SSDI Beneficiaries.

The final question about changes in SSA policy addressed whether the Ticket program should be mandatory for all SSA beneficiaries with disabilities. A majority of respondents disagreed with this statement (56%). However, as seen in Figure 4.53, a quarter of the providers agreed that this program should be mandatory for disabled SSA beneficiaries.

Question 17 - Ticket Program Should Be Mandatory for All Disabled SSA Beneficiaries 60

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.53 Ticket Program Should Be Mandatory for All Disabled SSA Beneficiaries

In summary, there are policy changes that SSA could implement over the long term that may increase provider participation in the Ticket program by presuming the eligibility of beneficiaries for return to work services earlier in the disability determination process. However, the respondents’ agreement with these policy changes falls short of making the Ticket program mandatory for all SSA beneficiaries with disabilities. The numbers of “don’t know responses” most likely indicate that those providers are not currently working with SSI/SSDI beneficiaries and are not familiar with

SSA’s disability determination process.

134

Technical Assistance and Training Needs

Marketing

Prospective ENs were surveyed about their ability to identify and market to ticket holders. In general, a plurality of providers (45%) believed they did not have the ability to market to ticket holders, as shown in Figure 4.54. Additionally, as shown in Figure

4.55, forty-three percent of the respondents did not know if they needed help, while 36% agreed that they do need help in identifying and marketing directly to ticket holders. The number of providers who responded “don’t know” (28%), and the percent of agreement with this statement, may indicate a need to provide training to prospective ENs in how to effectively contact and market their services to ticket holders. Technical assistance could enable prospective ENs to determine how they might develop the capacity to market to ticket holders.

Question 42 - Do Not Have Ability to Market to Ticket Holders 50

0 Percent Disagree Agree Don't Know Level of Agreement Figure 4.54 Do Not Have Ability to Market to Ticket Holders

135

Question 44 - Need Help Identifying and Marketing to Ticket Holders 50

0 Percent Disagree Agree Don't Know Level of Agreement Figure 4.55 Need Help Identifying and Marketing to Ticket Holders

Developing an Individual Work Plan or Individual Plan for Employment (IPE)

As shown in Figure 4.56, the majority of respondents reported they did not need assistance developing individual work plans (61%). There was a small minority of respondents (15%) who indicated they do need help, and a larger number of providers who did not know. These responses demonstrated that technical assistance in this areamay be necessary for some providers in order to increase participation in the Ticket program.

Question 48 - Need Help Developing Individual Work Plan 70

0 Percent Disagree Agree Don't Know Level of Agreement

Figure 4.56 Need Help Developing Individual Work Plan

136

4.6.9 Demographic Information

In the final section of the survey, a variety of additional demographic information was collected, some of which was summarized earlier in this results section and will not be repeated here. The prospective ENs were asked what types of support services they offered, and the accessibility of transportation for their clients.

Types of support services available

Providers were surveyed regarding the types of support services they offered or had available through networking. This question was important because it helped to determine the range of services that may be available to ticket holders if these providers were recruited to become ENs. As shown in Figure 4.57, the majority of providers offered case management (CM) services. Many providers also offered physical rehabilitation (PH) and support groups (SUPGRP) for clients. One-third of the providers offered family member counseling and support (COUNSPT); interpreter services

(INTER); equipment for ambulation, hearing or vision (AMBHRVI); or transportation

(TRANS). A relatively small number of providers (19) offered supervised housing

(SUPERHOU), while 11 offered childcare (CHCARE),

Support Services Offered

80 60 40 20

Providers 0 Number of

E R T P R A A CM C H INTE C SUPGR

Types of Support

Figure 4.57 Support Services Offered 137

Types of transportation available to clients

Prospective ENs were also surveyed regarding the types of transportation options available to their clients. This question was necessary in order to determine if transportation for beneficiaries was readily accessible so they could access employment services and eventually work. As shown in Figure 4.58, the majority of providers (84) indicated that they go to their clients (OTRAN), rather than providing transportation services. Taxis and the public bus line were the next most frequently occurring responses. Only 23 respondents reported they actually provide transportation for their clients. Twenty providers used staff personal vehicles (STAFFVEH), while 27 providers indicated their clients use para-transit services (PARTRANS). The least frequently used types of transportation were subway, commuter train, and ferry.

Transportation Services Available

100 80 60 40 20 0

Number of Providers Number N S Y N H S I Y R A N BU A TAX RR HE VTRA UBW FE OT O S PR COMTRAISTAFFVEPARTR Types of Transportation

Figure 4.58 Transportation Services Available

4.7 Responses to Open-Ended Questions and Text Analysis

The text analysis provided by NCSA found that the dominant themes of the open- ended questions related to the payment structure, payment issues, and the recordkeeping requirements. The responses indicated that prospective ENs may be interested in the 138

Ticket program, but the adequacy of the payment options is preventing participation. In addition, it was also clear there was a lack of understanding of the payment options/issues, and an overriding belief that providing services under the Ticket program would not be cost effective. The following unedited comments also revealed that SSA has an opportunity to encourage more program participation through better information dissemination and regulatory changes. They are reported as written in the open-ended sections of the survey. These responses are representative of the replies that were given in response to the question, “Please tell us under what circumstances your organization would participate in the Ticket program.”

• If the payment structure were set up to pay for services provided rather than an

outcome or milestone/outcome method.

• Better financial and reporting arrangements

• As may be clear from my responses, the payment structure needs to be

reconfigured to improve payments and make payments more timely

• If some of the financial reimbursements and rules were changed. There is too

much upfront time in screening and assessment that is not reimbursable – If SSA

is looking for only 1% to be successful – think about how much screening and

non-reimbursable services are expended to reach this goal with all of the other

SSI/SSDI beneficiaries to get down to that one percent that is capable of utilizing

this program!

• Our organization would be interested in participating in the Ticket program if the

administrative, operational, and fiscal benchmarks, outcomes, and payment plan

was similar to this state’s VR contractual arrangement. 139

• Need information re payment, eligibility for my organization, record keeping, etc

before making determination. Would definitely like to know more so can make

informed decision.

• I need to review the program, particularly pay rates, administrative rules,

tracking and statistic keeping.

• I understand this program is not set up where Tribal programs are able to get

reimbursed. It would take a lot of paperwork to satisfy the requirements, which

we are not set up to do.

• The majority of individuals we work with are through Supported Employment-

reaching SGA is not a realistic goal for them.

140

CHAPTER 5

5. CONCLUSIONS AND IMPLICATIONS

5.1 Overall Implications

The Ticket to Work-Work Incentives Improvement Act has produced several consequences that were not intended by its creators. The Ticket program was not intended to supplant SVRAs, but rather to complement their work. It was intended to attract a wider diversity of providers, especially from the private sector, to increase beneficiary choice in rehabilitation programs, not further deter providers from serving

SSA beneficiaries. It was intended to increase the number of beneficiaries who find employment, not continue the same rehabilitation system, which has been shown to be so ineffectual in its employment initiatives. Lastly, it was not intended to solve all of the problems of the disability program of the SSA; yet, the success of the program appears to be closely tied to SSA’s current disability policies. Reactions of various stakeholders to the problems with the Ticket program have diverted attention from the law’s primary objectives: increased employment of individuals with disabilities, and improved fiscal integrity of the Disability Trust Fund.

Overall, with the majority of respondents not interested in becoming ENs, identification of those issues/factors that are barriers to participation was critical.

141 Although statistical significance was not found in most of the analyses due to the number of don’t know responses, there was significance in the responses of the participants from a practical standpoint. The results of this survey provided critical information to SSA regarding barriers to participation in the Ticket program. It may also have identified potential changes that would facilitate greater participation on the part of prospective

ENs. Time is of the essence for SSA to make regulatory changes to the Ticket program to improve the performance of the program and increase employment service options for beneficiaries. The results of the survey highlighted several key findings. It should be noted that these results are only generalizable to the population of prospective ENs surveyed in the six states and not to the entire population of prospective ENs across the country.

The purpose of this study was to explore and describe the relationship between the interest in becoming an EN and demographic characteristics and attitudes of prospective EN’s regarding the Ticket program. At the beginning of the study, it was assumed that a relationship might be found between these variables. The results of the statistical analysis showed that the interest in becoming an EN was dependent upon the respondents’ attitude toward the need for upfront funding. The high level of agreement with the statement upfront funds would enable participation as an EN was a major factor in finding this relationship. However, the Cramer’s V statistic for the measure of the relationship was weak.

Surprisingly, it was determined that the interest in becoming an EN was not dependent upon the other variables, which were assumed to be of primary importance prior to the study. There are several potential explanations for these findings. The first is

142

related to the sample size. It may be that the sample size was not large enough to find

significance where it existed due to the number of variables that were analyzed, thus

causing a Type II error of accepting the null hypothesis when it was actually false.

A second explanation may be the number of don’t know responses. These

responses had a practical significance because they revealed a great deal about the need

for increased marketing and knowledge of the Ticket program. However, they may have

negatively affected the ability to discern a relationship when there was one, especially

when combined with the effect of the sample size. Consequently, this may have created

an opportunity for a Type II error (Beta) because there was not enough power in the

analysis to find significance.

A third potential reason for these findings may be that the lack of interest in

becoming an EN was not related to any one variable of interest, but was a function of the

Ticket program design, as a whole. The overall effect of one issue may be a barrier to participation when combined with the many other issues and areas of concern. This explanation may be the most cogent, since frequency counts and the distributions on the various program components revealed a strong level of agreement on certain components being viewed as barriers to participation in the Ticket program, yet statistical significance was not found. The practical significance of these findings is explored in the following sections.

5.2 Interest in Becoming an EN

The respondents to this survey overwhelming indicated they have no interest in participating in the Ticket program. Multiple possible explanations for non-participation were explored. The results of the survey showed that neither geographic location, nor the

143

work site locations of the prospective ENs were variables that influenced the providers’ decision not to participate in the Ticket program. In addition, the numbers of services the providers offered, the relationship with SVRAs, and the EN reporting requirements also were not factors in the providers’ decision not to participate in the program. Although the need for upfront funding was shown to influence the interest in becoming an EN, from a practical standpoint, the most influential factor may be the combination of the attitudes towards the payment system, payment options and issues, delays in reimbursement, and lack of upfront funding.

5.3 Primary Factors for Non-Participation for Prospective ENs

5.3.1 The Payment System

The current payment system for ENs is inadequate to attract private-for-profit and non-traditional providers, such as staffing agencies, to participate in the Ticket program.

A restructuring of the payment system is not only indicated, but also critical to the success of the program, as stated in the Ticket to Work Advisory Panel’s Annual Report to the President and Congress (Ticket to Work Advisory Panel, 2003).

To date, a small number of providers are currently enrolled as ENs and some of the same problems/barriers identified in earlier programs continue to exist. The provider payment schedule is an on-going concern and a barrier to participation, as evidenced by the results of this survey. The Social Security Advisory Board echoed these concerns regarding the adequacy of incentives for the ENs. In their advisory report on design issues, the Board encouraged the study of the financial incentives to ENs, especially as it relates to the length of time in receiving reimbursement and the impact for both large and small providers. The historic inadequacies of the SVRA’s cost reimbursement payment

144

schedule continue to plague lawmakers, who do not seem to know how to balance the need for provider accountability against profitability in an employment program.

Discussion regarding these issues took place at the EN summit in May 2003. More substantive activities have taken place since that time.

The AOI Group (2004) most recently addressed this issue in its July 2004 report.

In their proposal of regulatory changes, they recommended revisions to the payment structure by rewarding ENs for assisting beneficiaries in a gradual movement off the disability rolls. These proposed changes are supported by the results of the survey. The

AOI Group wisely suggested a plan by which three stages of self-sufficiency could be attained. These stages ranged from initial efforts, defined as $600/month to partial self- sufficiency ($830/month) to full self-sufficiency (the point at which the beneficiary ceases to receive cash benefits). The payments for milestone/outcome and outcome only reimbursements would be tied to this graduated plan, rather than the all or nothing approach of the current payment systems. This would allow providers to be reimbursed for services much earlier in the rehabilitation process, which may help to defray the initial upfront costs of screening and marketing to eligible beneficiaries.

In this same report, the AOI Group also recommended changing the reimbursement rate from 40% to 60% of the savings to the Disability Trust Fund. This change would provide more money for the provider in the Ticket program, and would still save the Disability Trust Fund from bankruptcy. Additionally, by eliminating the issue of the use of SGA as criteria for reimbursement, one potential barrier to participation in the Ticket program (as identified by this survey) may be diminished. The results of this survey support these regulatory changes. The SSA recently submitted

145

proposed changes to the payment system to the Office of Management and Budget

(OMB). At this time, the proposed changes are under further review since the OMB has

indicated they would be too expensive to implement (Gerry, 2005).

Providers will not recuperate their expenses if the payment system’s structure

continues. One of the focuses of the Ticket program was to increase consumer choice in

rehabilitation providers beyond the state-federal system. As long as the payment

structure remains status quo, the rehabilitation “choice” of consumers will continue to be

primarily the SVRAs. This is presently evidenced by the fact that over 90% of the

Tickets (over 87,000) have been assigned to the SVRAs. In contrast, approximately 100

rehabilitation providers are actively accepting tickets and have less than 2,000 tickets

assigned to date (F. Stielkowicz, personal communication, April 16, 2005).

5.3.2 Payment Options and Issues

The results of this research project clearly demonstrated that the payment options for providers is simply not sufficient to attract most prospective ENs. As in any flourishing business, rehabilitation providers have multiple sources of referrals and revenue. It is a business decision, in part, for each provider to decide which referred

clients they will serve. Most private-for-profit rehabilitation providers serve many

clients, and must decide on a daily basis whom they will serve based on eligibility criteria

and expectation of a successful outcome, a return to work.

When the Ticket program was first envisioned, SSA had been so frustrated with

its return on investment for rehabilitation services within the SVRAs, it concluded that an

outcome-only model of payment would be its safest method of payment to providers.

Yet, the problems of capitalization and time between services-rendered and payment are

146

so prohibitive that over the first three years of the Ticket program, less than a thousand providers across the country had been willing to participate in this new program

(Maximus, 2003; OESP, 2003). Although this number has increased to over 1,000 providers, only approximately 100 providers are actively taking tickets (Maximus, 2005).

Inequities between SSI and SSDI reimbursements

The respondents to the survey clearly indicated that the disparity in reimbursement between these two programs created a financial and service delivery dilemma. The inequities in reimbursement may be a major barrier to participation since some organizations exist specifically to serve the harder to serve populations such as those with chronic mental illness or development disabilities. At this time, the reimbursements for providers serving these populations is not an adequate income source.

Many of these programs are non-profits who especially cannot afford to loose money on the Ticket program when their budgets are already substantially constrained.

Making the SSI/SSDI reimbursement schedules more equitable may encourage greater participation by providers. It is an area of SSA policy, which needs to be considered over the long term. Creating parity between these two groups of beneficiaries will increase program participation on the part of ENs and expand the range of provider options for beneficiaries. Even though the savings to the Disability Trust Fund for SSI beneficiaries is not as high as the SSDI population, there are substantial reasons to increase the parity for reimbursement between the two programs. The most salient of the reasons is to reduce “creaming” in the provision of services.

The Social Security Advisory Board also recommended that SSA consider random sampling of ENs to collect data regarding the perception that the people in the

147

four target groups will ultimately cost more to serve than is available in the outcome/milestone payment program (Board, 2002). Based on the results from this survey, collecting data regarding barriers to serving the four target groups warrants exploration. Respondents clearly expressed concern regarding serving clients under the

Ticket program who require more extensive services and are more costly to serve.

Further research with prospective ENs regarding the perceived need for these changes is needed.

5.3.3 Delays in Reimbursement

The respondents to the survey indicated that the potential for the delay in reimbursements was a major barrier to participation in the program, even though it did not rise to the level of statistical significance. They also responded that waiting more than 60 days for reimbursement would be a financial hardship for their organization. At this time, providers need reassurance that payments will be made in a timely fashion, thereby reducing their financial burden and risk. Since the majority of respondents to this survey were either non-profit organizations or sole proprietors, timely reimbursement is critical for their fiscal viability.

Over the past two years current ENs have testified before the Ticket Advisory

Panel quarterly meetings (Forand, 2004: Webb, 2003) to emphasize the plight of the ENs.

One EN, which is a medium size independent living center in Arizona reported $30,000 in receivables from the Ticket program as of May 2003. Another EN (a nationally-based, large organization) reported it has been given a year by the company to make this program profitable or they will have to discontinue participation since they were owed over $20,000 as of May 2004. These programs cannot continue with outstanding

148

receivables of this nature. The Ticket program will not survive, let alone grow, if the

timely processing of payments is not improved and made a priority by SSA. SSA has

recognized this difficulty, but has been plagued by a declining work force and

understaffing in the area responsible for processing payments (F. Stielkowicz, personal

communication, April 16, 2005.)

5.3.4 The Relationship between SVRAs and ENs

One of the most salient reasons for the results of the Ticket program so far has been the resistance of the SVRAs to embrace change. As with the deregulation of any industry or consumer service in our economy, change within this system has been slow.

The SSA and Congress envisioned ENs as a pathway to employment for those beneficiaries who did not need extensive services, but short-term, goal-directed assistance in securing employment.

As the legislation currently stipulates, SVRAs are both providers and regulators of services. As a regulator of the law, the SVRA in each state prescribes the nature of the relationship between the state agency and the ENs for SSA beneficiaries requiring their mutual assistance. Thus, an inherent conflict of interest exists when a public entity is both a competitor and a regulator. The relationship between SVRAs and ENs needs to be better differentiated, and some assurances need to be given to both parties that their funding would not be adversely affected by these changes, if possible.

The poor fiscal situation for many states makes it unlikely that SVRAs will ever have sufficient personnel to serve the number of ticket holders who will request services. Of the 89,000 tickets deposited by SSA beneficiaries, over 87,000 are held by

SVRAs (Maximus, 2005). Testimony before the Ticket Advisory Panel revealed that the

149

backlog is substantial with waiting lists of six months or more for services (Bates-Harris,

2003). Although it is two years since this testimony, this backlog continues to exist (C.

Bates-Harris, personal communication, March 17, 2005). Denying ticket holders access to services in a timely fashion has the potential to become a major deterrent for SSA beneficiaries who want to return to work and the overall success of the Ticket program.

In some States, like Ohio, over 40% of the individuals being served by SVRA are

SSA beneficiaries (ORSC, 2002). The SVRA program cannot help all of these individuals, but it has been reluctant to encourage the development of alternate providers under the Ticket program. Under an operating budget in which appropriated funding is based upon the number and types of individuals served, the state administrators of SVRA have considered the Ticket program as new, and unwanted, competition. Naturally, there is a tremendous reluctance to encourage referrals to ENs if the belief is that it will jeopardize their dominance of SSA beneficiaries and associated funding.

The SVRAs have been receiving guidance from SSA regarding how to treat non- ticket holders, new cases and cases that were currently being processed or evaluated before the implementation of the Ticket program. As previously discussed in the literature review, Transmittal 17 has raised particular concerns because it allows the

SVRAs to make a ticket assignment on behalf of a beneficiary once an IPE has been completed. This procedure for a ticket assignment precludes the client from seeking out other providers unless a reassignment for the ticket is requested from Maximus.

Additionally, if services have been provided and reimbursement is due the SVRA, the potential value of the newly assigned ticket is reduced. Guidelines such as this are having the undesired effect of creating a monopoly for the state system, and increased

150

confusion regarding the roles of ENs and SVRAs, and their relationships regarding

reimbursements.

The results of this survey demonstrated that many respondents were not aware of

issues related to depositing tickets with the SVRAs, or did not see working with the

SVRAs as a barrier to participation. This may be an indication that they are not currently

working with SVRAs to serve clients. It may also be an indication that these

organizations prefer to accept the reimbursement rate of SVRAs, rather than assuming

the risk of serving that beneficiary under the Ticket program. Since the Ticket program

was created to expand the range of services and to create networks of providers to serve

clients, if the state cooperative agreements are perceived as harmful by prospective ENs,

the range of services will be reduced and the number of ENs will not increase. Although

this survey did not document this finding, testimony before the Ticket Advisory Panel has

addressed this issue over the past several years (Harles, 2003; Zwirn, 2001).

Fundamental, legislative change needs to occur for SVRAs so that the Ticket

program becomes successful. Not surprisingly, the Ticket Advisory Panel made this

recommendation in its 3rd Annual Report to the President and Congress. In this report, the Panel recommended that Congress review the Ticket program and compare it with other relevant programs administered by SSA and RSA to ensure that these programs

“operate in a positive and complementary fashion” (Ticket Advisory Panel, 2003). If cooperative agreements are developed without EN input and require the deposit of the ticket before services are provided, or repayment for services before reimbursement is received, they undermine the spirit of the Ticket program, which is creating networks to provide options for employment services for ticket holders. However, the current

151

reauthorization of the Rehabilitation Act, which only occurs every five years, did not include systemic changes conducive to the overall implementation of the Ticket program.

The results of this survey did document that some providers are concerned about competing with SVRAs for clients and fear the loss of their funding from their SVRA for their clients who need services that are more comprehensive. They also fear losing their ticket holder to the SVRA if they refer that client for employment-related services.

The issue of the state relationship with ENs is one that SSA is actively investigating (Gerry, 2005). Although the respondents to this survey indicated they were not aware of problems with SVRAs and agreements for services, the relationship between the SVRAs and current ENs still merits further investigation based on anecdotal information received by the researcher from prospective ENs from around the country.

Since the results of this survey are not generalizable to the entire population of prospective ENs from across the country, it would be beneficial to SSA to determine if difficulties exist on a national level that SSA could address and rectify through contact with the SVRAs. Specifically, the cooperative agreements between the SVRAs and ENs in each state need to be evaluated to identify policies and procedures that may impede competition and consumer choice in rehabilitation providers.

5.3.5 EN Credentials/Requirements

Based on the results of the survey, prospective ENs believe they have professional staff with the proper credentials to serve ticket holders. Therefore, the credentials of their staff are not perceived as a barrier to participation in the Ticket program. However, most of the respondents indicated that the application to become an EN is a barrier to participation because it is too cumbersome (over 100 pages) and time consuming to

152

complete. Based on this limited sample, SSA may need to examine the application process to reduce the paperwork and burden on providers in order to increase program participation.

5.3.6 EN Reporting Requirements

A majority of the survey respondents reported that collecting pay stubs for 60 months is difficult and a barrier to participation. This barrier to participation has been partially alleviated through a recent change in regulations that change the income reporting requirements. However, the current procedure of employment verification is seen as more burdensome (Harles, 2005). Because the largest number of respondents was non-profit organizations and sole proprietors, the infrastructure may not exist for these programs/companies to meet the reporting and tracking requirements of the Ticket program. Since respondents also reported that they would participate in the Ticket program if software were available to assist with tracking beneficiary status, SSA may want to explore the development of a software program specific to the Ticket program.

The responses of the participants may indicate that the development and availability of a software program specific to tracking the progress and payments for ticket holders may increase provider participation.

5.3.7 Criteria for the Evaluation of ENs

Nearly three-fourths of the respondents indicated that the criteria for measuring the success of ENs should be based on the types of placements, level of work activity of the beneficiaries, and the population the organization serves. Although the majority of respondents supported evaluating ENs based on the number of placements, one-third of the respondents disagreed. Prospective ENs have expressed concern anecdotally about

153

how their performance will be evaluated under the Ticket program. Participation by

providers may increase once SSA makes public the evaluation process and criteria. More

information regarding evaluation criteria may enable prospective ENs to feel comfortable

that they can meet the standards required of the program.

5.3.8 Upfront Funding/Capitalization

Not surprisingly, a significant finding of the study was the relationship between the agreement for the need of upfront funding and the interest in becoming an EN. The provision of upfront financial support is critical to the success of the Ticket program.

The providers who have chosen to participate need to remain involved. Current ENs have noted a high-volume of responses on the part of beneficiaries to the rollout of the tickets. This high-volume of response has resulted in the need for administrative support to conduct pre-screening activities (Expediter, 2002; Forand, 2004; Rehabilitation

Partnership, 2002; Webb, 2003). This initial upfront cost is not unique to the private and non-profit providers. The Wisconsin VR agency also reported the high-volume of calls, cost of responding to the rollout, staff training, and increased reporting requirements, as challenges related to the Ticket program. However, the Wisconsin VR agency is able to recoup these costs through the cost reimbursement option for payments.

In testimony provided during the Ticket Advisory Panel’s quarterly meeting in

Washington D.C. in August 2002, ENs testified regarding a variety of concerns and barriers to continuing with the program (Expediter, 2002; Rehabilitation Partnership,

2002; Zwirn, 2002). ENs reported that it is difficult to recoup initial costs. They also indicated that it is a longtime before a break-even point is reached concerning serving ticket holders. They noted that beneficiaries usually are eager and enthusiastic about

154

returning to work. However, they reported that many are discouraged. The lack of

available ENs to meet the demand is critical. One organization reported beneficiary

statements such as, “all the training organizations are full” and “ there is no other EN in

my area” (Expediter, 2002: Rehabilitation Partnership, 2002). Recommendations from

this testimony included additional support to ENs to provide needed services to ticket

holders and careful consideration of marketing to employers (Wisconsin VR, 2002).

Results of this study demonstrated that providing up front funding to cover the cost of

initial screenings would increase provider participation.

5.4 Unanticipated Findings of This Study

Two unanticipated findings occurred through the course of this study that may have an impact on provider participation, but were not measured formally in this research. The first relates to the actual number of providers that are theoretically available to provide services to SSA beneficiaries. Based on the knowledge of the researcher, the assumption around the country among the rehabilitation community is that only a small proportion of providers have elected to become ENs out of a large number of potential providers. This study found that approximately 30% of the businesses that were contacted were no longer in existence. If this number holds true on a national level, the actual population of providers may be smaller than anticipated and so the participation rate of the rehabilitation providers is at a higher level than assumed. This could have a positive impact on the public image of the Ticket program and the current measure of its success. Additionally, this may also result in less access to services and consumer choice.

155

A second finding was the number of non-profit organizations who did not provide employment services in the community, but maintained “in-house” programs, such as sheltered workshops. In terms of integration and equal access of people with disabilities to employment in the community, finding 25-30 organizations that did not provide any community-based employment was a concern. Further research on the actual number of providers who offer employment services in the community may be warranted, both as a factor of marketing efforts to prospective ENs, and as a matter disability public policy, as well as increased access to employment for people with disabilities.

5.5 Other Disability Policy Issues Related to the Ticket to Work

5.5.1 SSA Policy Changes

The results of this study demonstrated that major regulatory changes might be needed, as well as further research into the implications of the implementation of the

Ticket program. Changes may need to occur in the overall design of SSA’s disability and employment programs, beginning with the disability determination process and extending to a presumption of a return to work and the offer of rehabilitation services upon application for benefits.

Definition of Disability /Early Use of the Ticket

Within the federal system, multiple definitions of disability exist, creating differing standards for eligibility for programs. SSA’s definition of disability, an inability to perform any work, differs from the Rehabilitation Act of 1973 that entitles individuals to vocational rehabilitation and presumes they can work with the right vocational rehabilitation interventions. These definitions are also different from the ADA, which defines disability in terms of disruption of functioning in major life areas, and

156

encourages the use of reasonable accommodations to enable an individual with a

disability to work. In order for SSA employment programs to be successful in returning

individuals to work, SSA’s definition of disability must change from the standards of 25

years ago to meet today’s standards that reflect the current knowledge about disability,

technology and accommodations, and employment issues. It must reflect levels of

employability, and accept the premise that not all individuals with disabilities can

become employed, but those who can should, by moving towards a definition of

disability that recognizes levels of partial and total disability that can be either temporary

or permanent, much like the private sector insurer’s systems.

Moreover, the troubles of SSA associated with the Ticket program are only

symptomatic of some of its overall concerns. As the Social Security Advisory Board

(2001) proclaimed in Charting the Future of Social Security’s Disability Programs: The

Need for Fundamental Change, SSA should address early intervention in a systematic fashion. The Ticket program cannot be successful if services are rendered only after a lengthy application process and subsequent disability award. Early intervention, as a fundamental change, means refining SSA’s current disability determination process. The

SSA needs to modify the process from an application by a prospective beneficiary and disability determination by adjudicators, to a viable employability assessment and offer of rehabilitation services by ENs through the Ticket program. The SSA has begun to address this issue through the development of the Early Intervention Program, a model program funded through the Disability Research Institute, to determine how early intervention could be structured and who the appropriate candidates would be for such a program (Berkowitz, 2003). However, this program does not include a thorough

157

employability assessment at the beginning of the disability determination process. Since a majority of respondents to the survey agreed with providing potential SSI/SSDI beneficiaries with a ticket at the time of application for benefits, this procedural change warrants further investigation by SSA.

The problems of the Ticket legislation, in part, are what it tries to accomplish in face of these tremendous barriers to reform. Overall, SSA needs to adequately address the following concerns: an effective mechanism for referral to rehabilitation at the time of application for disability benefits; the amount of time it takes to secure a decision about disability determination; the permanent (rather than temporary) nature of that decision; and finally, the inherent disincentives for beneficiaries to obtain work. The Ticket legislation was not intended to solve all of the problems of the SSA disability program, yet its successful implementation appears to be so closely linked to these issues that if they are not addressed through regulatory changes, the Ticket program may fail. The following additional recommendations arise out of the results of the prospective EN survey.

5.6 Additional Recommendations for Ticket Program Changes

From the standpoint of implementation, the following additional recommendations are presented so that the primary objective of the Ticket program (ie, higher employment of SSA beneficiaries, and reduced costs for the SSA Disability Trust

Fund) can be realized.

158

5.6.1 Allow ENs Access to SSA Data so that They Can Select Beneficiaries who are

Feasible Candidates for Rehabilitation

The responses to the survey highlighted the most preferred method to contact ticket holders, and important information that is needed to determine feasibility for services and employment potential. The majority of providers preferred to have

Maximus provide a list of eligible ticket holders in their area. The respondents preferred to know the medical and employment history of the beneficiary to determine if a ticket holder is an appropriate candidate for services with their organization. In other indemnity programs in which rehabilitation professionals are paid to return beneficiaries to work, the insurance company provides access to critical background information about prospective clients in order to make informed decisions regarding employment potential.

Providers can determine who they will assist, what services might be needed, whether they are the most appropriate provider of those services, and most importantly for most providers, what their return-on-investment might be.

As demonstrated by testimony from current ENs before the Ticket Advisory Panel

(Expediter, 2002; Forand, 2004; Petkaukos, 2005; Rehabilitation Partnership, 2002;

Webb, 2004; Zwirn, 2002), without access to such information in the referral process, rehabilitation providers spend an inordinate amount of time and/or money screening clients on the phone, and collecting data to determine employment potential when that data is readily available. Having access to referral data early in the process can expedite the delivery of services by matching the services of the provider to the needs of the beneficiary early in the process. One straightforward way of implementing this change is to allow all of the Benefit Planning and Assistance Organizations (BPAOs), created

159

under the Ticket program to be both a BPAO and an EN and to provide counseling to beneficiaries on the impact of working on their benefits. This model is especially helpful for beneficiaries with severe mental illness who need to establish trust and rapport with their providers. In fact, in some states, the BPAO is the SVRA; therefore, this type of arrangement already exists as a service model.

5.6.2 Utilize Vocational Experts who Participate in the SSA Disability Determination

Process to Evaluate Employment Potential before Benefits are Awarded.

Consistent with the GAO recommendation that the SSA should develop a systematic procedure for returning beneficiaries to work, the SSA needs to accept responsibility that

SSA applicants for disability are not using other avenues of rehabilitation services for returning to work. A major flaw in the current system is that too many applicants for disability under SSA are not receiving rehabilitation services before they are awarded disability benefits. Not coincidentally, many vocational experts for SSA, who assist in disability determination decisions at the appeals level, are also rehabilitation providers.

The SSA’s disability determination process needs to incorporate their services so that applicants with re-employment potential are identified and become involved in rehabilitation before SSA awards disability benefits (Darling, Growick, et al., 2002;

Drew & Growick, 2003). Respondents to the survey substantially supported changes to

SSA policies to allow beneficiaries access to return to work services at the time of application for benefits, if the beneficiary can be presumed eligible.

5.6.3 Technical Assistance and Training

Create a Research and Training Center

Congress provided funding in TWWIIA for BPAOs so that the needs of

160

the beneficiaries are met. This emphasis on the demand side of the Ticket program was necessary, in part, to help beneficiaries overcome their fear of losing benefits. The needs of the supply side, the providers, have been largely ignored in this regard. The establishment of a Research and Training Center to assist potential ENs with training related to business plan development, employability determinations, IPEs, and marketing to beneficiaries and employers is sorely needed. A work group on developing the agenda for a Research and Training Center highlighted and confirmed these trainings needs during the Employment Network Summit held in Washington, D.C. on May 22-23, 2003.

Presently, a national ad hoc group of current ENs and Ticket program advocates has formed the National Employment Network Association to address these issues. The SSA may want to look to this group as a strategic partner in the success of this program.

Like the beneficiaries, ENs, too, are experiencing fear. They fear providing services without adequate or timely reimbursement, and that they will be overwhelmed by governmental regulations and paperwork. They are adverse to shouldering much of the financial risk in this program. Currently, the SSA is trying to grow a program in the face of three years of negative word-of –mouth. An R&T Center devoted to the Ticket program with special emphasis on supply-side issues can disseminate information regarding best practices for success, and provide needed support to struggling ENs. At the same time the overall viability of the Ticket program can be researched.

Marketing and Training Needs and Recruitment of Beneficiaries

Creative ways of accessing and marketing to beneficiaries is a critical to enable

ENs to continue to be active in the program. One EN noted the success of establishing a

“Hot Jobs” newsletter as a way to market to beneficiaries and to advise beneficiaries

161

regarding the local labor market (Zwirn, 2002). Other programs, such as the Wisconsin

VR system, have built on vendor partnerships and established new partnerships to assist

beneficiaries in returning to work (Wisconsin VR, 2002). More needs to be known about

the ENs’ ability to form viable partnerships with business and industry to facilitate

employment for ticket holders.

According to recent testimony before the Ticket Advisory Panel, “interest and

skepticism about the ‘Ticket’ program is quite high” (among ENs). (Emphasis added by

author). This testimony went on to note that ENs need assistance in developing business

plans, collaboration and partnerships, marketing, and benefits training (Harles, 2002).

The capacity of ENs to adjust their placement strategy from the place a person in a job to

a marketing approach to employment is critical to their success under the Ticket program.

Research indicates that employers consider agencies that place people with disabilities as reliable sources of information related to the skills, abilities and need for accommodations for their clients (Gilbride, 2000). However, many employers may be confused and overwhelmed by the number of resources available for job accommodation and tax incentives (O’Leary & Dean, 1998). The expertise of ENs with marketing perspectives geared toward both beneficiaries and employers is integral to the success of the Ticket program.

The respondents to the survey indicated they need help identifying and marketing to ticket holders. Many non-profit agencies receive referrals directly from outside sources and have never had the need to develop a marketing strategy. Most have never had to gear up for the number of calls and screenings necessary to assess the feasibility and appropriateness of working with a ticket holder. Developing a viable business plan

162

that can anticipate the start up cost and staffing patterns needed to serve ticket holders are critical skill sets for ENs. To date, no training or education is provided to current or prospective ENs to address these issues. A technical assistance and training center is could be a vital partner in the success of the Ticket program. It is a critical missing component.

The training needs of ENs regarding marketing to employers and beneficiaries, plan development, and infrastructure development need to be fully understood.

Employment Networks need to be assisted in moving from the traditional placement model to a marketing and disability management approach to placement. In this regard, it is critical that ENs be surveyed to determine the following factors: a) their capability for marketing to beneficiaries and employers; b) what resource issues they are facing in terms of serving beneficiaries (e.g. transportation, housing or lack of access to Medicaid

Buy-Ins); and c) identifying creative ways in which they are making the system work, as potential “best practice” models for other ENs.

Marketing Efforts and the Recruitment of ENs

With the passage of the Ticket program, recruitment of ENs has been ongoing.

One barrier to EN participation is the ability of the Project Manager to identify and market to the providers. The difficulty with establishing a viable population to survey for this project is indicative of the current lack of a comprehensive list of providers across the country. How can the marketing efforts of Maximus be successful without accurate contact lists; without knowing who the market is and what their needs are; and without the financial resources to wage an all out campaign to reach prospective ENs? A project to identify rehabilitation providers from across the country and to merge them into a

163

comprehensive database is currently underway at the University of Wisconsin-Stout.

This database will be helpful for future efforts. However, it must be maintained so that current information is available for marketing purposes by the Program Manager, and for providing beneficiaries with viable vocational options. Although this will be costly and time-consuming, new marketing efforts need to be geared to an audience that can be identified and is accessible.

5.7 Summary

Overall, the Ticket program is experiencing some successes as evidenced by the thousands of dollars that have been paid to ENs for returning beneficiaries to work.

However, as long as the payment system continues to be perceived as complex, and the payment options are not sufficient to encourage providers to participate, SSA will fail to achieve its goal of savings to the Disability Trust Fund, and more beneficiaries working.

This research project has identified critical changes that need to occur in the

Ticket program in order to improve its chances to succeed. This information, in concert with other ongoing research projects regarding the Ticket program, has the potential to enable SSA to target specific program components for improvements, and thereby to increase provider participation and beneficiary levels of employment.

164

REFERENCES

Adequacy of Incentives Advisory Group. (2003). Recommendations for improving implementation of the Ticket to work and self-sufficiency program (regulatory and administrative changes). Champaign, IL: Author

Ary, D., Jacobs, L. C., & Razavieh, A. (2002). Introduction to research in education (6th ed.).Belmont, CA: Wadsworth/Thomson Learning.

Bates-Harris, C. (2003, May). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, DC.

Berkowitz, M. (2003a). The Ticket to work program: The complicated evolution of a simple idea. In S. H. Bell & K. Rupp (Eds.), Paying for results in vocational rehabilitation: Will provider incentives work for Ticket to work? (1-29). Washington, DC: The Urban Institute.

Berkowitz, M. (2003b). Early intervention process demonstration. Paper presented at the ticket to work advisory panel quarterly meeting, Washington, D.C.

Bovbjerg, B. D. (2001). SSA disability: Other programs may provide lessons for improving return-to-work efforts.: Washington, DC: Government Accountability Office.

Bosnjak, M. (2003). Prepaid and promised incentives in web surveys. Social Science Computer Review, 21(2), 208-217.

Campbell, D.T., & Stanley, J.C. (1963). Experimental and quasi-experimental designs for research. Chicago: Rand McNally & Co.

Cebula, R.A. (2003, February). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, D.C.

Couper, M.P. (2000). Web surveys: A review of issues and approaches. Public Opinion Quarterly, 64(4) 464-495.

Darling, W. T., Growick, B. S., & Kontosh, L. G. (2002). Transferable skills analysis in rehabilitation: Issues in definition and application. Journal of Vocational Rehabilitation, 17, 1

165 Dillman, D. (1991). The design and administration of mail surveys. Annual Review of Sociology, 17, 225-249.

Dillman, D. (2000). Mail and Internet surveys: The tailored design method (Second ed.). NewYork: John Wiley & Sons, Inc.

Drew, J. (2001). Rhodes to independence. Unpublished manuscript, Cranston, RI.

Drew, J. L., & Growick, B. S. (2003). The role and functions of the vocational expert in social security: Strategies for the 21st century. Journal of Forensic Vocational Analysis, 8, .43-53.

Forand, T. (2004, May) Testimony presented to the ticket to work and work incentive advisory panel quarterly meeting, Washington, D.C.

Employment Payment Network Systems. (2002). Retrieved 02/25/2002, 2002, from http://www.ssa.gov/work/resourcestoolkit/paymentsststem_factsheet.html

Einschlag, H. (2003, May). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, D.C.

Expediter Corporation. (2002, August). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, D.C.

Fallavolita, B. S., & Bordelon, B. H. (1997). Social Security disability programs lag in promoting return to work. NARPPSS Journal, 12(4), 169-180.

Fehner, C. (2003, February). Testimony presented to the ticket to work and work incentives advisory panel, Albuquerque, NM Forgiel, K., & Growick, B. (1997). Inspiring a partnership between private sector rehabilitation and the social security administration. NARPPSS Journal, 12(4), 153-158.

Fowler, F. (1995). Improving survey questions. Sage Publications: Thousand Oaks, CA

Gerry, Martin. (2005, April). Presentation at the Disability Research Institute Symposium, Washington, D.C.

Gilbride, D. (2000). Employers’ attitudes toward hiring persons with disabilities and vocational rehabilitation services. Journal of Rehabilitation, 66(4), 17-23.

Growick, B. (2000a). The political implications of the ticket to work-work incentives improvement act. Rehab Pro, 8(5), 31-34.

166

Growick, B. (2000b). The ticket to work and work incentives improvement act: A review. RehabPro, 8(1), 17-19.

Growick, B. (2001). The ticket to work law to be implemented. Rehab Pro, 9(3), 18.

Growick, B., & Drew, J. (2003). The ticket to work: The unintended consequences of an imperfect law. Journal of Forensic Vocational Analysis, 6, 49-54.

Growick, B., Drew, J., & Johnston, C. (2002). The evolution of RCES: Meeting the demands of private sector rehabilitation. Unpublished manuscript.

Harles, A. L. (2002, November). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, D.C.

Harles, A.L. (2005, May). Testimony presented to the ticket to work and work incentives advisory panel, quarterly meeting, Arlington, VA.

Hays, W. L. (1994) Statistics. (5th Edition). New York: Harcourt Brace, pp.862-863.

Hopkins, K.D., Hopkins, B.R. & Glass, G.V. (1996). Basic statistics for the behavioral sciences. (3rd edition). Boston: Allyn and Bacon. p. 242.

Heerwegh, D. & Loosveldt, G. (2002). An evaluation of the effect of response formats on data quality in web surveys. Social Science Computer Review, 20(4), 471-484.

Kindersley, D.(1998). Oxford Illustrated American Dictionary. New York: Dorling Kindersley Ltd.

Kornfield, R., & Rupp, K. (2000). The net effects of the project network return-to-work case management experience on participant earnings, benefit receipt and other outcomes. Social Security Bulletin, 63, 12-33.

Lewin Group. (2003). Evaluation design for the Ticket to work program: preliminary process evaluation. Washington, D.C., author.

Marini, I., & Reid, C. R. (2001). A survey of rehabilitation professionals as alternative provider contractors with social security: Problems and solutions. Journal of Rehabilitation,67(2), 36-41.

Marini, I., & Stebnicki, M. (1999). Social security administration’s alternative provider program: What can rehabilitation administrators expect? Journal of Rehabilitation, 23(1), 31-41.

167

Maximus. (2003, May). Program manager update for the quarterly ticket to work advisory panel meeting. Testimony presented to the ticket to work advisory panel quarterly meeting. Baltimore, MD.

Maximus. (2003, May). Ticket assignments report. Testimony presented to the ticket to work advisory panel quarterly meeting. Baltimore, MD.

Mertler, C. (2002). Demonstrating the potential for web-based survey methodology with a case study. American Secondary Education, 30(2), 49-61.

Office of Employment Support Programs. (2003). SSA quarterly update on the Ticket to work and work incentives improvement act implementation, third fiscal quarter of 2003. Report presented at the ticket to work advisory panel quarterly meeting. Washington, DC: Social Security Administration.

Ohio Rehabilitation Services Commission. (2002). Annual Report. Colu6mbus, OH: author.

O’Leary, P., & Dean, D. (1998). International research project on job retention and return to work strategies for disabled workers: Study report – USA. United Kingdom: International Labour Organization, University of York.

Pallant, J. (2001). SPSS survival manual. New York: Open University Press.

Petkauskos, K. (2005, May). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting. Arlington, VA.

Rehabilitation Partnership (2002, August). Paper presented at the ticket to work and work incentives advisory panel quarterly meeting, Washington, DC.

Schillewaert, N., Lanagerak, F., & Duhamel, T. (1998). Non-probability sampling for www surveys: A comparison of methods. Journal of Market Research Society, 40(4), 307-322.

Schmidt, W.C. (1997). World-wide web survey research: Benefits, potential problems, and solutions. Behavior Research Methods, Instruments & Computers. 29(2), 274-279.

Sills, S.,& Song, Chunyan. (2001, April). Innovations in survey research: An application of web-based surveys, Pacific Sociology Association Session 176 Methods Roundtable. San Francisco, CA.

Smith, C. B. (1997). Casting the net: Surveying an internet population [Electronic version]. Journal of Computer Mediated Communication, 3(1).

168

Social Security Advisory Board. (2001a). Charting the future of social security’s disability programs: The need for fundamental change. Washington, DC: author.

Social Security Advisory Board. (2001b). Preliminary advice report on the adequacy of incentives study design: Washington, DC: author.

Social Security Administration. (2003a). 2002 SSI Annual Report. Washington, DC: author.

Social Security Administration. (2003b). Ticket to work and work incentives advisory panel. Washington, DC: author.

Social Security Administration. (2002). Employment network payment systems. Retrieved 02/25, 2002, from Social Security Online. (2002). The Ticket to Work and Self-Sufficiency Program: Final Rule. Retrieved May 13, 2002.

Social Security Administration. (2001). 2000 SSI annual report. Washington, DC: author.

Stevens, J.P. (2002). Applied multivariate statistics for the social sciences, 4th edition. Mahwah, N.J.: Lawrence Erlbaum Associates.

Taylor, H. (2000). Employment trends of people with disabilities 1998-2000. Harris Poll.

Tenney, F. E., & McCray, D. (1997). Project network: Successful linkages between social security and private sector rehabilitation. NARPPS Journal, 12(4), 137-152.

Ticket to Work and Work Incentives Advisory Panel. (2003a). Ticket to work and work incentives advisory panel 3rd annual report to the President and the Congress. Washington, DC: author.

Ticket to Work and Work Incentives Improvement Act, Pub. L. No. 106-170 (1999).

U.S. General Accountability Office. (2002a). SSA disability: Enhanced procedures and guidance could improve service and reduce overpayments to concurrent beneficiaries. GAO-02-802. Washington, D.C., author.

U.S. Government Accountability Office. (1996). SSA disability: Return-to-work strategies from other systems may improve federal programs. Washington, DC.: author.

169

Webb, S . (2003, November). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, D.C.

Wisconsin Vocational Rehabilitation. (2002, August). The Wisconsin Experience. Paper presented at the ticket to work and work incentives advisory panel quarterly meeting, Washington, DC.

Yun, G. W. (2000). Comparative response to a survey executed by post, e-mail and web form [Electronic version]. Journal of Computer Mediated Communication, 6(1), 1-30.

Zhang, Yin. (1999). Using the internet for survey research: A case study. Journal of the American Society for Information Science, 51(1), 57-68

Zwirn, S. (2000, August). Testimony presented to the ticket to work and work incentives advisory panel quarterly meeting, Washington, DC.

170

APPENDICES

171

APPENDIX A LIST OF PANEL OF EXPERTS

Name Title Organization Location Steven Start President & CEO SL Start & Associates Spokane, WA Susan Webb Director ABIL Phoenix, AZ Mary Satterfield Program Manager Maximus Alexandria, VA Dr. Fred Menz Director Univ. Wisconsin-Stout Menomonie, R & T Center WI Bryon Director World Institute on Oakland, MacDonald Disability CA Margaret Glenn Director National Council of Morgantown Rehabilitation ,WV Educators Thomas Dunleavy Owner Disability Innovations Oak Park, IL Robert Coons Executive Director Curative Care Milwaukee, Network WI

172

APPENDIX B COVER LETTER TO PANEL OF EXPERTS

Dear:

We are writing to ask for your assistance in the development of an important survey regarding the Ticket to Work. You have been chosen as a member of our panel of experts because of your commitment to and knowledge of the field of private rehabilitation.

This survey proposes to measure the attitude of private providers toward various Ticket program components, and the decision to participate as an EN. Additionally, we will be analyzing the relationship between demographic characteristics and participation in the Ticket program as an EN.

This review by the panel of experts will help establish the validity of this instrument. We are requesting that you review the survey to determine the appropriateness and clarity of the questions and their applicability to the Ticket components.

If you choose to participate, you will be asked to review the survey items and provide written comments. A self-addressed stamped envelope will be provided for your convenience. A full explanation of the process will be enclosed with your packet. If you have any comments or questions about this study or process, please feel free to call or email.

We hope you will be able to participate in this timely project.

Sincerely,

Judith L. Drew, ABD, CRC Doctoral Candidate Rehabilitation Services The Ohio State University 614-247-6670 [email protected]

173

APPENDIX C SURVEY NOTIFICATION LETTER

«First» «Last» «Title_2» «Company_1» «Company_2» «Address_1» «Address_2» «PO_Box» «City», «State» «Zip»

Dear «Title_1» «Last»:

We are writing to ask for your help with our survey project regarding the Ticket to Work and provider participation. Because your participation in this project is voluntary, we are even more appreciative of your willingness to take the time to respond. Your completion and submission of the survey indicates consent for your responses to be used for this study.

You will be completing the survey on-line (or by mailed questionnaire, if appropriate). It will take approximately 20 minutes to complete. Specific instructions for completing the survey, and the survey can be found at the survey website:

«URL»

Your survey identification number is «ID_». Your unique ID number must be entered to submit your survey and will ensure the confidentiality of your responses.

Please complete the survey within the next week to ten days. If you are unable to do so, please advise us as soon as possible by phone, or email at [email protected].

Thank you for participating in our project. We have enclosed a small token of our thanks for your time. We look forward to receiving your response.

Sincerely,

174

Judith L. Drew, ABD, CRC Doctoral Candidate Rehabilitation Services The Ohio State University 614-247-6670 [email protected]

175

APPENDIX D COVER LETTER FOR MAILED OR FAXED SURVEYS

«First» «Last» «Company_1» «Company_2» «Address_1» «Address_2» «PO_Box» «City», «State» «Zip»

Dear Ms. «Last»:

Thank you for your interest in the Ticket to Work survey. Enclosed please find the hard copy of the survey, as you requested. Please remember to include your ID number on page one of the survey.

«ID_»

We really appreciate your input and response to this questionnaire. Because your participation in this project is voluntary, we are even more appreciative of your willingness to take the time to respond. Your completion and submission of the survey indicates consent for your responses to be used for this study.

Once you have completed the survey, and return the survey by fax (614-292-4255) or mail, we will enter you into the drawing for $50 of free gas. Twenty respondents waschosen on September 22nd . We look forward to receiving your response.

Sincerely,

Judith L. Drew, ABD, CRC Doctoral Candidate Rehabilitation Services The Ohio State University 614-247-6670 [email protected]

176

APPENDIX E DRAWING POSTCARD

Complete the survey and return this postcard by August15th to be entered in a drawing for $50.00 of free gas. The drawing washeld on August 22nd . Twenty cards will be randomly chosen.

«First» «Last» «Company_1»

177

APPENDIX F REMINDER POSTCARD

Time is running out! A letter about our online survey asking your opinion regarding the Social Security Administration’s Ticket to Work program was mailed to you. Our server at OSU was down for several weeks, but now we are back on line. We want to hear from you!

We especially appreciate your help, because it is voluntary. We feel your organization can benefit from the Ticket program, but we need your opinions so that we can improve it for you. Please, take a minute to respond today. Here is your ID # «ID_» and web survey address: «URL»

Please remember to use your ID number. We look forward to hearing from you!

Judith Drew, ABD, CRC Doctoral Candidate The Ohio State University

178

APPENDIX G TABLES AND FIGURES

Worksite Location

Cumulative Frequency Percent Valid Percent Percent Valid Inner City 4 4.0 4.3 4.3 Urban 17 17.0 18.1 22.3 Suburban 6 6.0 6.4 28.7 Small Town 11 11.0 11.7 40.4 Rural 7 7.0 7.4 47.9 Urban/Suburban 9 9.0 9.6 57.4 SmallTown/Rural 8 8.0 8.5 66.0 Urban/Rural 5 5.0 5.3 71.3 Suburban/Rural 1 1.0 1.1 72.3 Urban/Surburban/Rural 2 2.0 2.1 74.5 Multisite- all 5 worksites 6 6.0 6.4 80.9 Urban/Suburban/Small 6 6.0 6.4 87.2 Town/Rural InnerCity/Urban/ 4 4.0 4.3 91.5 Suburban InnerCity/Urban/Small 1 1.0 1.1 92.6 Town/Rural Inner City/Suburban 1 1.0 1.1 93.6 InnerCity/SmallTown/ 1 1.0 1.1 94.7 Rural InnerCity/Suburban 1 1.0 1.1 95.7 Urban/Suburban/Small 1 1.0 1.1 96.8 Town InnerCity/Urban/ 1 1.0 1.1 97.9 Suburban/SmallTown InnerCity/Urban/ 1 1.0 1.1 98.9 Suburban/Rural Suburban/SmallTown/ 1 1.0 1.1 100.0 Rural Total 94 94.0 100.0 Missing System 6 6.0 Total 100 100.0

Table G.1 Frequency Table Worksite Locations

179

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 19.756a 20 .473 Likelihood Ratio 24.738 20 .212 Linear-by-Linear .424 1 .515 Association N of Valid Cases 90 a. 38 cells (90.5%) have expected count less than 5. The minimum expected count is .36.

Table G.2 Chi-Square Test of Independence for Worksite Location

Number of Services

N Valid 88 Missing 12

Mean 5.85 Median 6.00 Std. Deviation 2.098 Range 9 Percentiles 25 5.00 50 6.00 75 7.00

Table G.3 Measures of central tendency for number of services with interquartile ranges.

180

Number of Services

Cumulative Frequency Percent Valid Percent Percent Valid 1 5 5.0 5.7 5.7 2 5 5.0 5.7 11.4 4 7 7.0 8.0 19.3 5 18 18.0 20.5 39.8 6 16 16.0 18.2 58.0 7 17 17.0 19.3 77.3 8 14 14.0 15.9 93.2 9 5 5.0 5.7 98.9 10 1 1.0 1.1 100.0 Total 88 88.0 100.0 Missing System 12 12.0 Total 100 100.0

Table G.4 Frequency table for number of services offered.

Interest in Being an EN and Number of Services Offered

Number of Services 1 2 4 5 6 7 8 9 10 Total Be an No Count 4 5 4 10 8 11 7 3 0 52 EN Expected Count 3.1 3.1 3.7 11.1 9.3 9.9 8.0 3.1 .6 52.0 % within been 7.7% 9.6% 7.7% 19.2% 15.4% 21.2% 13.5% 5.8% .0% 100.0% Yes Count 1 0 2 8 7 5 6 2 1 32 Expected Count 1.9 1.9 2.3 6.9 5.7 6.1 5.0 1.9 .4 32.0 % within been 3.1% .0% 6.3% 25.0% 21.9% 15.6% 18.8% 6.3% 3.1% 100.0% Total Count 5 5 6 18 15 16 13 5 1 84 Expected Count 5.0 5.0 6.0 18.0 15.0 16.0 13.0 5.0 1.0 84.0 % within been 6.0% 6.0% 7.1% 21.4% 17.9% 19.0% 15.5% 6.0% 1.2% 100.0%

Table G.5 Contingency table for chi-square for interest in being an EN and number of services offered.

Chi-Square Tests

Asymp. Sig. Value df (2-sided) Pearson Chi-Square 6.912a 8 .546 Likelihood Ratio 8.991 8 .343 Linear-by-Linear 2.636 1 .104 Association N of Valid Cases 84 a. 11 cells (61.1%) have expected count less than 5. The minimum expected count is .38.

Table G.6 Chi square test of independence for number of services and interest in becoming an EN.

Factor Analysis - Payment System and Issues 181

Correlation Matrix

comp6 comp5 comp13 comp3 comp34 comp14 comp4 comp12 comp28 comp23 comp1 comp15 comp16 comp19 comp36 comp11 comp7 comp8 comp9 Correlation comp6 1.000 .621 .347 .098 .197 .264 .519 .307 .155 .361 .344 .170 .114 .280 .231 .141 .550 .704 .335 comp5 .621 1.000 .476 .281 .318 .217 .490 .391 .094 .362 .359 .146 .358 .397 .268 .087 .307 .397 .149 comp13 .347 .476 1.000 .431 .409 .232 .299 .531 .267 .427 .081 .373 .516 .507 .295 .029 .165 .196 .299 comp3 .098 .281 .431 1.000 .230 .233 .242 .264 .156 .379 .385 .317 .345 .251 .184 .062 .067 .076 .206 comp34 .197 .318 .409 .230 1.000 .232 .150 .343 .123 .262 .226 .370 .328 .267 .190 -.001 .156 .080 .185 comp14 .264 .217 .232 .233 .232 1.000 .385 .279 .299 .207 .267 .265 .104 .220 .240 .345 .149 .113 .314 comp4 .519 .490 .299 .242 .150 .385 1.000 .212 .174 .214 .388 .240 .190 .300 .228 .113 .386 .335 .367 comp12 .307 .391 .531 .264 .343 .279 .212 1.000 .309 .392 .192 .246 .280 .366 .331 .196 .079 .186 .262 comp28 .155 .094 .267 .156 .123 .299 .174 .309 1.000 .247 .159 .233 .164 .228 .051 .189 .077 .124 .257 comp23 .361 .362 .427 .379 .262 .207 .214 .392 .247 1.000 .197 .122 .211 .291 .197 .098 .169 .320 .335 comp1 .344 .359 .081 .385 .226 .267 .388 .192 .159 .197 1.000 .110 .194 .154 .133 .304 .267 .260 .127 comp15 .170 .146 .373 .317 .370 .265 .240 .246 .233 .122 .110 1.000 .442 .219 .237 -.071 -.042 .082 .279 comp16 .114 .358 .516 .345 .328 .104 .190 .280 .164 .211 .194 .442 1.000 .356 .178 .160 .034 .135 .189 comp19 .280 .397 .507 .251 .267 .220 .300 .366 .228 .291 .154 .219 .356 1.000 .165 .100 .111 .170 .241 comp36 .231 .268 .295 .184 .190 .240 .228 .331 .051 .197 .133 .237 .178 .165 1.000 .033 .026 .040 .301 comp11 .141 .087 .029 .062 -.001 .345 .113 .196 .189 .098 .304 -.071 .160 .100 .033 1.000 .134 .209 .227 comp7 .550 .307 .165 .067 .156 .149 .386 .079 .077 .169 .267 -.042 .034 .111 .026 .134 1.000 .639 .197 comp8 .704 .397 .196 .076 .080 .113 .335 .186 .124 .320 .260 .082 .135 .170 .040 .209 .639 1.000 .352 comp9 .335 .149 .299 .206 .185 .314 .367 .262 .257 .335 .127 .279 .189 .241 .301 .227 .197 .352 1.000 Table G.7 Correlation matrix for payment system and issues.

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .790

Bartlett's Test of Approx. Chi-Square 612.292 Sphericity df 171 Sig. .000

Note: KMO of >.6 and Bartlett’s test of <.05 permit factor analysis for payment system

Table G.8 Payment system and issues KMO and Bartlett’s Test of Sphericity.

Scree Plot

3 Eigenvalue 2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Component Number

Figure G.1 Scree plot for payment systems and issues

182

Component Matrixa

Component 1 2 3 4 5 comp6 .689 -.515 comp5 .699 -.338 comp13 .705 .361 comp3 .509 .316 .408 comp34 .499 .302 comp14 .505 .551 comp4 .628 .355 comp12 .617 comp28 .396 .429 comp23 .584 -.315 comp1 .489 .639 comp15 .458 .452 .400 comp16 .519 .407 comp19 .569 comp36 .418 .485 comp11 .662 -.314 comp7 .440 -.650 comp8 .542 -.626 comp9 .543 -.421 Extraction Method: Principal Component Analysis. a. 5 components extracted.

Table G.9 Component matrix for payment system and issues factor analysis.

183

Rotated Component Matrixa

Component 1 2 3 4 5 comp6 .841 comp5 .511 .544 comp13 .810 comp3 .490 .511 comp34 .507 comp14 .478 .536 comp4 .527 .447 .343 comp12 .620 .328 comp28 .631 comp23 .565 comp1 .795 comp15 .368 .638 comp16 .646 comp19 .630 comp36 .665 comp11 .724 .330 comp7 .792 comp8 .839 comp9 .521 .444 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 12 iterations.

Table G.10 Varimax rotation of principal components for payment system and issues

Factor Analysis: Payment System – Options and Structure

Correlation Matrix

comp6 comp5 comp4 comp7 comp8 Correlation comp6 1.000 .621 .519 .550 .704 comp5 .621 1.000 .490 .307 .397 comp4 .519 .490 1.000 .386 .335 comp7 .550 .307 .386 1.000 .639 comp8 .704 .397 .335 .639 1.000

Table G.11 Correlation Matrix for payment system factor analysis

184

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .762

Bartlett's Test of Approx. Chi-Square 199.139 Sphericity df 10 Sig. .000

KMO of >.6 and Bartlett’s test of <.05 permit factor analysis for payment system

Table G.12 Payment system and issues KMO and Bartlett’s Test of Sphericity.

Total Variance Explained

Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 3.001 60.024 60.024 3.001 60.024 60.024 2 .849 16.970 76.994 3 .568 11.357 88.351 4 .358 7.164 95.515 5 .224 4.485 100.000 Extraction Method: Principal Component Analysis.

Eigenvalues > 1 are retained.

Table G.13 Total variance explained for payment system factor analysis.

185

Scree Plot

2 Eigenvalue

1 2 3 4 5 Component Number

Figure G.2 Scree plot for payment system PCA.

Component Matrixa

Component 1 comp6 .892 comp5 .720 comp4 .687 comp7 .747 comp8 .810 Extraction Method: Principal Component Analysis. a. 1 components extracted.

Table G.14 Unrotated component matrix for payment system

Rotated Component Matrixa

a. Only one component was extracted. The solution cannot be rotated.

186

Statistics

REGR factor score 1 for analysis 1 N Valid 95 Missing 5 Mean .0187 Std. Error of Mean .10166 Median .0623 Std. Deviation .99090 Range 3.33 Percentiles 33.33333333 -.5072 66.66666667 .7038

Table G.15 Regression factor score for payment system

Factor Analysis – Payment Issues and Options

Correlation Matrix

comp13 comp34 comp14 comp12 comp28 comp15 comp16 comp19 comp36 comp11 comp7 comp8 comp9 Correlation comp13 1.000 .409 .232 .531 .267 .373 .516 .507 .295 .029 .165 .196 .299 comp34 .409 1.000 .232 .343 .123 .370 .328 .267 .190 -.001 .156 .080 .185 comp14 .232 .232 1.000 .279 .299 .265 .104 .220 .240 .345 .149 .113 .314 comp12 .531 .343 .279 1.000 .309 .246 .280 .366 .331 .196 .079 .186 .262 comp28 .267 .123 .299 .309 1.000 .233 .164 .228 .051 .189 .077 .124 .257 comp15 .373 .370 .265 .246 .233 1.000 .442 .219 .237 -.071 -.042 .082 .279 comp16 .516 .328 .104 .280 .164 .442 1.000 .356 .178 .160 .034 .135 .189 comp19 .507 .267 .220 .366 .228 .219 .356 1.000 .165 .100 .111 .170 .241 comp36 .295 .190 .240 .331 .051 .237 .178 .165 1.000 .033 .026 .040 .301 comp11 .029 -.001 .345 .196 .189 -.071 .160 .100 .033 1.000 .134 .209 .227 comp7 .165 .156 .149 .079 .077 -.042 .034 .111 .026 .134 1.000 .639 .197 comp8 .196 .080 .113 .186 .124 .082 .135 .170 .040 .209 .639 1.000 .352 comp9 .299 .185 .314 .262 .257 .279 .189 .241 .301 .227 .197 .352 1.000

Table G.16 Correlation matrix for payment issues factor analysis.

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .729

Bartlett's Test of Approx. Chi-Square 299.171 Sphericity df 78 Sig. .000

KMO of >.6 and Bartlett’s test of <.05 permit factor analysis for payment issues

Table G.17 KMO and Bartlett’s Test of Sphericity.

187

Total Variance Explained

Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 3.784 29.110 29.110 3.784 29.110 29.110 2.805 21.574 21.574 2 1.692 13.013 42.122 1.692 13.013 42.122 1.754 13.493 35.068 3 1.245 9.574 51.696 1.245 9.574 51.696 1.666 12.813 47.880 4 1.002 7.710 59.406 1.002 7.710 59.406 1.498 11.526 59.406 5 .895 6.884 66.290 6 .827 6.364 72.654 7 .801 6.161 78.815 8 .658 5.063 83.878 9 .582 4.476 88.354 10 .499 3.838 92.192 11 .431 3.312 95.504 12 .309 2.375 97.879 13 .276 2.121 100.000 Extraction Method: Principal Component Analysis.

Eigenvalues > 1 are retained.

Table G.18 Total variance explained for payment issues/options factor analysis.

Scree Plot

2 Eigenvalue

1 2 3 4 5 6 7 8 9 10 11 12 13 Component Number

Figure G.3 Scree plot for payment issues/options PCA

188

Component Matrixa

Component 1 2 3 4 comp13 .758 comp34 .562 comp14 .526 .523 comp12 .674 comp28 .470 .375 -.348 comp15 .565 -.385 comp16 .598 comp19 .607 -.318 comp36 .450 .678 comp11 .442 .543 comp7 .314 .717 -.422 comp8 .413 .706 -.364 comp9 .582 .328 Extraction Method: Principal Component Analysis. a. 4 components extracted.

Table G.19 Component Matrix for payment issues

Rotated Component Matrixa

Component 1 2 3 4 comp13 .799 comp34 .597 comp14 .611 .457 comp12 .540 .327 comp28 .315 .622 comp15 .584 .371 comp16 .745 comp19 .654 comp36 .827 comp11 .779 comp7 .890 comp8 .878 comp9 .328 .322 .540 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 5 iterations.

Table G.20 Rotated component matrix.

189

Statistics

REGR factor score 1 for analysis 2 N Valid 87 Missing 13 Mean .0151817 Std. Error of Mean .10744119 Median .1311691 Std. Deviation 1.002145 Range 4.95377 Percentiles 33.33333333 -.4362000 66.66666667 .5001438

Table G.21 Regression factor score for payment issues/options including mean and median values

Factor Analysis – EN Reporting Requirements/Credentialing and Evaluation

Correlation Matrix

comp30 comp29 comp31 comp25 comp35 comp27 comp39 comp33 comp42 comp44 comp49 comp50 Correlation comp30 1.000 .374 .544 .339 .321 .127 .299 .600 .264 .124 .439 .323 comp29 .374 1.000 .386 .247 .190 .243 .274 .500 .215 .248 .376 .392 comp31 .544 .386 1.000 .446 .355 .162 .334 .585 .165 .133 .355 .189 comp25 .339 .247 .446 1.000 .380 .259 .319 .508 .204 .207 .368 .262 comp35 .321 .190 .355 .380 1.000 .079 .375 .377 .311 .233 .295 .093 comp27 .127 .243 .162 .259 .079 1.000 .310 .194 .251 .411 .146 .319 comp39 .299 .274 .334 .319 .375 .310 1.000 .362 .261 .206 .444 .235 comp33 .600 .500 .585 .508 .377 .194 .362 1.000 .372 .233 .422 .250 comp42 .264 .215 .165 .204 .311 .251 .261 .372 1.000 .407 .367 .188 comp44 .124 .248 .133 .207 .233 .411 .206 .233 .407 1.000 .240 .292 comp49 .439 .376 .355 .368 .295 .146 .444 .422 .367 .240 1.000 .400 comp50 .323 .392 .189 .262 .093 .319 .235 .250 .188 .292 .400 1.000 Table G.22 Correlation matrix for EN reporting requirements factor analysis

190

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .830

Bartlett's Test of Approx. Chi-Square 316.485 Sphericity df 66 Sig. .000

KMO of >.6 and Bartlett’s test of <.05 permit factor analysis

Table G.23 KMO and Bartlett’s Test of Sphericity

Extraction Sums of Squared Rotation Sums of Squared Initial Eigenvalues Loadings Loadings Compo % of Cumulati % of Cumulati % of Cumulati nent Total Variance ve % Total Variance ve % Total Variance ve % 1 4.437 36.978 36.978 4.437 36.978 36.978 3.208 26.734 26.734 2 1.402 11.686 48.664 1.402 11.686 48.664 1.970 16.420 43.153 3 1.070 8.916 57.580 1.070 8.916 57.580 1.731 14.427 57.580 4 .865 7.207 64.787 5 .811 6.760 71.547 6 .679 5.660 77.207 7 .621 5.178 82.385 8 .555 4.624 87.009 9 .497 4.139 91.148 10 .399 3.322 94.470 11 .383 3.189 97.659 12 .281 2.341 100.000 Extraction Method: Principal Component Analysis. Eigenvalues > 1 are retained

Table G.24 Total variance explained for EN reporting requirements factor analysis.

191

Scree Plot

2 Eigenvalue

1 2 3 4 5 6 7 8 9 10 11 12 Component Number

Figure G.4 Scree plot for EN reporting requirements PCA

Component Matrix(a)

Component 1 2 3 comp30 .688 -.346 -.178 comp29 .621 .023 -.431 comp31 .674 -.414 -.065 comp25 .638 -.148 .116 comp35 .554 -.191 .563 comp27 .433 .607 -.072 comp39 .608 .044 .194 comp33 .780 -.274 -.022 comp42 .531 .321 .396 comp44 .468 .626 .196 comp49 .684 -.016 -.072 comp50 .523 .335 -.523 Extraction Method: Principal Component Analysis. A 3 components extracted.

Table G.25 Component matrix for EN reporting requirements

192

Rotated Component Matrixa

Component 1 2 3 comp30 .774 .024 .162 comp29 .654 .363 -.112 comp31 .747 -.062 .261 comp25 .521 .128 .393 comp35 .290 -.022 .759 comp27 .087 .738 .097 comp39 .376 .272 .441 comp33 .747 .102 .339 comp42 .104 .449 .574 comp44 -.005 .725 .352 comp49 .574 .296 .240 comp50 .471 .611 -.253 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 9 iterations.

Table G.26 Rotated component matrix.

Factor Analysis SVRAs and Competition or Loss of Funding and ENs

Correlation Matrix

comp21 comp20 comp26 comp49 Correlation comp21 1.000 .076 .306 .459 comp20 .076 1.000 .168 .143 comp26 .306 .168 1.000 .388 comp49 .459 .143 .388 1.000

Table G.27 Correlation matrix for relationship between SVRAs and ENs

193

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .650

Bartlett's Test of Approx. Chi-Square 40.468 Sphericity df 6 Sig. .000

KMO of >.6 and Bartlett’s test of <.05 permit factor analysis

Table G.28 KMO and Bartlett’s Test of Sphericity

Total Variance Explained

Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 1.832 45.798 45.798 1.832 45.798 45.798 2 .958 23.960 69.758 3 .686 17.142 86.900 4 .524 13.100 100.000 Extraction Method: Principal Component Analysis.

Eigenvalues > 1 are retained.

Table G.29 Total variance explained for SVRA and competition factor analysis.

194

Scree Plot

2.00

1.75

1.50

1.25 Eigenvalue 1.00

0.75

0.50

1 2 3 4 Component Number

Figure G.5 Scree plot for EN Competition and Loss of Funding PCA

Component Matrixa

Component 1 comp21 .740 comp20 .351 comp26 .718 comp49 .804 Extraction Method: Principal Component Analysis. a. 1 components extracted.

Table G.30 Unrotated component matrix for Competition and Loss of Funding

195

Statistics

REGR factor score 1 for analysis 1 N Valid 89 Missing 11 Mean -.0771 Std. Error of Mean .10304 Median -.0508 Std. Deviation .97212 Range 4.07 Percentiles 33.33333333 -.5173 66.66666667 .1882

Table G.31 Regression factor score for SVRA-EN factor analysis, separated into equal appearing thirds.

EN and Need for Upfront Funding

Symmetric Measures

Value Approx. Sig. Nominal by Phi .288 .019 Nominal Cramer's V .288 .019 N of Valid Cases 96 a. Not assuming the null hypothesis. b. Using the asymptotic standard error assuming the null hypothesis.

P<.05 shows dependency. Cramer’s V .288 shows weak relationship.

Table G.32 Cramer’s V statistic for interest in being an EN and need for upfront funding

196

Position in Organization

Cumulative Frequency Percent Valid Percent Percent Valid Exec Director 30 30.0 39.0 39.0 Program Mgr 21 21.0 27.3 66.2 Voc Counselor 16 16.0 20.8 87.0 Other 10 10.0 13.0 100.0 Total 77 77.0 100.0 Missing System 23 23.0 Total 100 100.0

Table G.33 Frequency table for position held by respondents

Organization Types

Type of Organization

Cumulative Frequency Percent Valid Percent Percent Valid Non-profits 50 50.0 53.2 53.2 Sole proprietors 16 16.0 17.0 70.2 State Gov 10 10.0 10.6 80.9 S Corp 11 11.0 11.7 92.6 C Corp 1 1.0 1.1 93.6 Limited Liability 3 3.0 3.2 96.8 Partnerships Tribal 1 1.0 1.1 97.9 Government Other, not 2 2.0 2.1 100.0 specified Total 94 94.0 100.0 Missing System 6 6.0 Total 100 100.0

Table G.34 Frequency table for type of organization

197

Number of Services

Table G.35 Frequency table for number of services provided.

198

APPENDIX H WELCOME TO THE PROSPECTIVE EMPLOYMENT NETWORK

SURVEY

The Ohio State University appreciates your willingness to participate in this survey because it will provide the Social Security Administration with valuable information about your opinions, perceptions and attitudes toward the Ticket program. By completing this survey, you are agreeing to be a participant. Your responses are important to us and are kept strictly confidential. No individual responses will be reported to the Social Security Administration. We begin by asking you about your organization's general knowledge of the Ticket to Work program, and then continues with specific aspects about its implementation. Each section has a heading for your convenience. If you make a mistake or change your mind, simply change your answer by clicking on another response option. Your final response for each question will not be recorded until you click on the SUBMIT button. If you are unhappy overall with your responses, you can click on the RESET button and start over again.

Your responses will not be recorded until you click on the submit button. Once you have clicked on the SUBMIT button, you cannot change your answers. You also cannot submit a new survey. Your ID and password allow you to submit your responses only once and cannot be reused. 0

If you have any problems completing this survey or need this survey in an alternate format, please contact Judi Drew at 614-247-6670 or via e- mail at [email protected]. Your personal ID (for example, 123450)

199

Section A. General Knowledge of the Ticket to Work Program Please answer the following questions about your organization's general knowledge of the Ticket program by clicking on the button below your choice. Answer each question by completing this statement: The staff of our organization is very knowledgeable about... Not at all Somewhat Very Don't

Knowledgeable Knowledgeable Knowledgeable Know 0 01 02 03 ...the Ticket program overall

0 01 02 03 ...the payment options

...the beneficiary status reporting 0 01 02 03 requirements ...the record keeping 0 01 02 03 requirements ...the requirements for providers 0 01 02 03 to become Employment Networks (EN).

Of the total number of people you serve what % percentage are referred by the state vocational rehabilitation program? Please enter a number for the percent referred by VR.

My organization is interested in becoming an EN. 0 1 No Yes Please tell us the sources of your knowledge about the Ticket to Work. Check all that apply. 1 A conference presentation 1 A MAXIMUS presentation 1 From the MAXIMUS web site 1 a MAXIMUS brochure or mailing 1 An article in a rehabilitation publication 1 Other rehabilitation providers 1 Ticket holders 1 other - please specify Please indicate your position with your organization by checking the appropriate box below. 1 executive director

200

1 program manager 1 financial officer 1 informative technology manager 1 counselor 1 other, please specify

Section B: Ticket to Work Program Components Please indicate to what extent your organization agrees or disagrees with each statement. Remember, your responses are kept strictly confidential.

Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know Having more funds available 0 01 02 03 04 05 06 up front would enable my organization to participate as an EN. Q 10 The criteria for evaluating the 0 01 02 03 04 05 06 effectiveness of an EN should be the types of placements, level of work (competitive employment, transitional employment, etc) and the population the EN is serving. Q 38 The payment structure 0 01 02 03 04 05 06 adequately compensates providers for the services delivered. Q 6 The payment structure for 0 01 02 03 04 05 06 returning beneficiaries to work will encourage my organization to become an EN. Q5 Serving Ticket holders who 0 01 02 03 04 05 06 will require long term services and support services in order to be employed is a barrier to our participation. Q24 Strongly Somewhat Disagree Agree Somewhat Strongly Don't Disagree Disagree Agree Agree Know

My organization has not 0 01 02 03 04 05 06 participated in the Ticket program because the way the payment options are calculated is unclear. Q 13 We have not participated in the 0 01 02 03 04 05 06 Ticket because we do not have a computer-based client tracking program that would enable us to keep records of the 201

case status of Ticket participants. Q 30 Social Security should refer a 0 01 02 03 04 05 06 potential beneficiary for return to work services along with a ticket to work at the time of application for benefits. Q45 The difference in the amount of 0 01 02 03 04 05 06 payments for returning SSI versus SSDI beneficiaries to work prevents us from serving SSI beneficiaries. Q2 We understand how payments 0 01 02 03 04 05 06 to ENs are calculated by SSA under the milestone/outcome payment options. Q3 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know We have not participated in the 0 01 02 03 04 05 06 Ticket because it would be difficult for us to track pay stubs, as required by the Ticket program. Q29 We have not participated in the 0 01 02 03 04 05 06 Ticket because we are concerned that it will take longer than 60 days to get paid for services after returning someone to work above SGA. Q34 The Ticket program should 0 01 02 03 04 05 06 reimburse providers for returning people to work who never reach SGA, as long as they are working at least 10 hours/week and paying taxes on their income. Q 14 We would participate in the 0 01 02 03 04 05 06 Ticket program if we could find/use a software tracking program for Ticket participants. Q31 We would be more interested 0 01 02 03 04 05 06 in becoming an EN if SSA implemented a referral process. Q41 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know The payment structure is clear 0 01 02 03 04 05 06 and easy to understand. Q4 The Ticket program should be 0 01 02 03 04 05 06 mandatory for all SSA beneficiaries with disabilities. Q17 202

Many of our clients never reach 0 01 02 03 04 05 06 SGA when they work, so participating in the Ticket program is not financially feasible for us. Q12 ENs should be allowed 0 01 02 03 04 05 06 reimbursement for services provided on a cost reimbursement basis like the state VR agency. Q28 We have not participated in the 0 01 02 03 04 05 06 Ticket because the application to become an EN is too cumbersome or time consuming to complete. Q25 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know We have not participated in the 0 01 02 03 04 05 06 Ticket program because we do not understand what it is and how it works. Q35 We agree that we should share 0 01 02 03 04 05 06 payment for services if we have referred our client to another agency, or state VR, for support services that we can not provide. Q23 Our organization is concerned 0 01 02 03 04 05 06 about competing for clients with the state vocational rehabilitation (VR) agency. Q21 Our organization employs staff 0 01 02 03 04 05 06 who are qualified to serve Ticket holders Q27 We understand the difference 0 01 02 03 04 05 06 between outcome and milestone/outcome payments. Q1 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know In the state(s) where we do 0 01 02 03 04 05 06 business, the public VR system has required that we sign a cooperative agreement indicating the Ticket must be deposited with the state agency. Q20 My organization can provide 0 01 02 03 04 05 06 services to any Ticket recipient who asks for assistance, regardless of disability. Q39 The dispute resolution process 0 01 02 03 04 05 06 is unfair to ENs. Q43 203

Payment 0 01 02 03 04 05 06 schedules/reimbursements should be the same regardless of SSI/SSDI status. Q32 An impartial third party should 0 01 02 03 04 05 06 be responsible for administering the dispute resolution process. Q46 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know The criteria for evaluating the 0 01 02 03 04 05 06 effectiveness of an EN should be the number of SSA recipients an EN returns to work. Q37 Waiting 45 days or more for 0 01 02 03 04 05 06 reimbursement for services after someone has begun working would be financial hardship for my organization. Q15 We have not participated in the 0 01 02 03 04 05 06 Ticket because we do not have a computer program that we can use to help us maintain and track the financial records for the Ticket program. Q33 We have not participated in this 0 01 02 03 04 05 06 program because we are concerned about the length of time it may take to be reimbursed for our work. Q16 The possibility of 0 01 02 03 04 05 06 overpayments to the beneficiary, which then affects EN payments, is an area of concern to my organization. Q19 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know If we compete for Ticket 0 01 02 03 04 05 06 clients with the state agency, we believe we will lose our state agency funding for other clients we serve. Q26 Payments to ENs should be 0 01 02 03 04 05 06 based on a sliding scale of earnings by the Ticket holder, rather than SGA or $0 cash benefits Q36 We have not participated 0 01 02 03 04 05 06 because we do not believe our staff has the credentials to work with Ticket holders. Q22 204

Social Security should provide 0 01 02 03 04 05 06 people with time-limited benefits as an incentive and bridge to return to work under the Ticket program. Q47 Many of our clients engage in 0 01 02 03 04 05 06 supported employment, so the Ticket program is not useful to us. Q11 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know People who apply for disability 0 01 02 03 04 05 06 benefits who could be presumed eligible for benefits should be provided with a Ticket at the time of application for benefits. Q18 We do not have the ability to 0 01 02 03 04 05 06 market to ticket holders. Q42 The difference in payment 0 01 02 03 04 05 06 schedules/reimbursements for SSI and SSDI recipients creates a financial and service delivery dilemma for ENs in terms of who we serve. Q40 ENs are adequately reimbursed 0 01 02 03 04 05 06 for returning SSI beneficiaries to work. Q7 ENs are adequately reimbursed 0 01 02 03 04 05 06 for returning SSDI beneficiaries to work. Q8 Strongly Somewhat Disagree Agree Somewhat Strongly Don't

Disagree Disagree Agree Agree Know ENs are not adequately 0 01 02 03 04 05 06 reimbursed for clients who receive SSI because they need more services, so our costs are higher to serve them. Q9 We need help identifying and 0 01 02 03 04 05 06 marketing to ticket holders. Q44 We need help developing an 0 01 02 03 04 05 06 individual work plan (IWP). Q48 My organization is not an EN 0 01 02 03 04 05 06 because we are concerned we will lose other funding (Medicaid, VR or DOL) if we participate in the Ticket program. Q49 Collecting pay stubs from 0 01 02 03 04 05 06 beneficiaries for 60 months is a barrier to participation. Q50

205

Would you become an EN if low interest start up loans were available for initial start up costs? Please check only one. 1 yes 0 no

What type of payment option would encourage your organization to participate as an EN? Please check all that apply. 1 outcome only 1 milestone/outcome 1 hourly reimbursement for time spent 1 lump sum payments for both employability evaluation and return to work plans 1 combination of hourly reimbursement and lump sum payments for employability evaluation and return to work plans 1 the payment structure is fine as it is

Please rank order your responses to the following question. Use a 1-4 scale. 1 is most effective. 4 is least effective. Remember to rank order.

The most effective way for ENs to access information for contacting beneficiaries would be the following: SSA (MAXIMUS) provides the list of ticket beneficiaries for our area(s) on a password protected web site. SSA representatives refer people directly to me. We personally review files to determine VR feasibility and appropriateness for services at our agency/organization.

Other - please describe -

Please rank order your responses to the following question. Use a 1-7 scale. 1 is most important. 7 is least important.

206

What information would your organization want to know in order to determine the feasibility of Ticket holders for employment services? Remember to rank order. work history medical history/functional capacities age education gender date last employed other - please describe -

Why is this information important to your organization?

Section C: Demographic Information Please give us some details about your organization by responding to the following questions.

1. The site(s) where your organization provides services is (are) located in the following geographic area(s). Please check all that apply. 1 inner-city 1 urban 1 suburban 1 small town 1 rural 1 other - please specify

2. How many staff (in total) are employed by your organization? Please enter number for total staff employed.

207

3. How is your organization classified? Please check only one. 1 non-profit entity 1 sole proprietor 1 state or local government 1 S-corp 1 C-corp 1 Limited partnership 1 Limited liability partnership 1 school/university 1 tribal government program 1 federal government program 1 other, please specify

4. Including Ticket and non-Ticket, approximately how many people per year does your organization serve? Please enter number for total people served.

5. Does your organization provide any of the following services either by directly providing them or partnering with other agencies? Please check all that apply. 1 Evaluation and assessment 1 Employment search/job development 1 Rehabilitation plan development 1 Job readiness skills 1 Employment support and maintenance 1 Career counseling 1 Job Training 1 Supported employment 1 Benefits planning and outreach 1 Occupational certificates and licenses 1 Other - please specify

208

6. Does your organization provide any of the following support services either by directly providing them or partnering with other agencies? Please check all that apply. 1 childcare 1 family member counseling and support 1 interpreter, reader, and note-taker 1 physical rehabilitation 1 assistive technology 1 equipment for ambulation, hearing or vision 1 support groups for clients 1 transportation 1 case management 1 supervised housing 1 other - please specify

7. On average, how many miles do your clients travel to get to your main location? Please enter number for total miles traveled.

8. What types of transportation do your clients have access to in order to get to your main location? Please check all that apply. 1 We provide transportation 1 Public bus line 1 Subway 1 Commuter train 1 Staff personal vehicles 1 Para-transit 1 Taxi 1 Ferry 1 Other - please specify

209

Please tell us under what circumstances your organization would want to participate in the Ticket program.

Section D: Other Comments

We have tried to be comprehensive in our survey. However, there may be other issues or ideas that were not covered. This is your chance to tell the Social Security Administration what you want them to know about the Ticket to Work program. Remember your comments are strictly confidential and will not be individually reported to SSA.

Submit Form Res et Form

210