Client satisfaction and quality vocational rehabilitation.

Introduction

Total Quality Management (TQM) is one of the latest concepts to receive attention in the management literature. The Federal Total Quality Management Handbook defines the concept as "a way of managing at all levels, top management to front-line, to achieve customer satisfaction by involving all employees in continuously improving the work processes of the organization" (United States Office of Personnel Management 1992, p. 1). The concept has been recognized in the vocational rehabilitation literature by authors such as Lubeck and Davis (1991) who described the relationship of W. E. Deming's 14 points for "quality in management" to the field of rehabilitation and by Schwab and Fenoglio (1992) who described the Texas Rehabilitation Commission's system for measuring quality rehabilitation.

Reports of long-term and widespread dissatisfaction with the way in which the quality of vocational rehabilitation counselors' work has been evaluated are no secret (see, for example, Downes, McFarland & Alston, 1974; Emener & Placido, 1982a; Lorenz, 1979; Moriarity, 1977: Vernon, Bussey & Day, 1979). Improvements have been made, such as taking into consideration the severity of clients' disabilities, but most systems continue to focus on case closure rates. The Rehabilitation Act of 1973 included a program evaluation standard specific to client satisfaction (Moriarity, 1977: Patterson & Leach, 1987). Although response to it has been uneven, the consumer-oriented Americans with Disabilities Act of 1990 and the Rehabilitation Amendments of 1992 (HR 5482) will likely encourage greater attention to the concept.

Consumer satisfaction is a central component of quality management. According to Johansson and McArthur (1988), "quality is first and foremost a perception in the consumer's eyes", and according to Hutchins (1986), "in searching for a usable and quantifiable basis for an organization's quality strategy, the focus changes from internal to external--to the customer". Much of the client satisfaction literature comes from the fields of health (see Winter & Keith, 1988) and mental health (see Parloff, 1983). In 1974, reports were that more than half of state rehabilitation agencies were using client satisfaction measures (Bassett, 1974), but in a 1982 study, Emener and Placido concluded "that current rehabilitation counselor evaluation practice...does not utilize systematic client feedback in a manner designed to facilitate the professional development of rehabilitation counselors" (1982a, p.75). The way in which Emener and Placido (1982a) arrived at this finding has been criticized (Bordieri, 1982; Bedient, 1982), but their interest in client satisfaction remains important.

Howe, Minch, and Fay (1980) noted the lack of research and documentation on consumer involvement in rehabilitation and wrote "that consumer involvement in rehabilitation is still in its infancy of development". In a more recent article, Patterson and Leach (1987) reviewed the vocational rehabilitation literature on client satisfaction and concluded that "although there was a great deal of activity in the 1970s in the area of client satisfaction and most states have continued to collect client satisfaction information, there has been a paucity of rehabilitation research on client satisfaction in recent rehabilitation articles".

Client satisfaction is clearly an essential element of quality rehabilitation, but there are other elements of quality as well. In 1987 the Texas Rehabilitation Commission (TRC) introduced a system for assessing counselor performance based on both the quality and quantity of services they render (Schwab & Fenoglio, 1992). The four components of the system are (1) client satisfaction, (2) client needs assessment fulfillment, (3) case folder review, and (4) client advocacy and community relations. The study reported in this article updates the discussion of client satisfaction in vocational rehabilitation by describing the TRC's efforts to measure client satisfaction as one component in assessing the productivity of vocational rehabilitation counselors and the performance of the vocational rehabilitation agency.

Methodological Approaches in VR Client Satisfaction Studies

Previous approaches to assessing client satisfaction have included mail surveys (Hills & Hills, 1975), telephone interviews (Winter & Keith, 1988), and face-to-face interviews Roessler & Bolton, 1985; Greenley & Schoenherr, 1981), as well as combinations of methods (JWK International Corporation, 1980; Reagles, Wright & Butler, 1970). In one of the larger satisfaction studies reported in the literature, the JWK Corporation attempted to conduct a sample of 13,482 former vocational rehabilitation clients from across the United States. Contacts were first made by mail to determine former clients' willingness to participate and to obtain their informed consent. This was followed by telephone calls to schedule a face-to-face interview or visits to the home if the former client could not be reached by phone. Twenty-one percent of the sample completed interviews. Most interviews were done in the client's home. Greenley and Schoenherr (1981) also used face-to-face interviews to conduct a satisfaction study of clients of 11 social service agencies (including two rehabilitation agencies) in a single county. They used a smaller sample of recent clients and had a much higher response rate. Initial contacts to ask clients to participate were made at the agencies. Ultimately, the researchers were able to conduct face-to-face interviews with 441 clients, 65% of the original sample. Interviews were also conducted in the clients' homes.

Bitter, Kunce, Lawver, Miller, and Ray (1972) decided to compare the effectiveness of different follow-up methods with former clients of a state employment security office. They used telephone contacts first. If they got no response they used mail (letters), and if they still got no response, they attempted a personal visit. Thirty-one percent of the sample of 216 was contacted, 17.1% by telephone, an additional 8.8% by letter, and 5.1% more by a personal visit. However, only 19 of 179 individuals answered letters, and of these, only 11 agreed to participate, while 11 oF 12 individuals contacted by personal visit agreed to participate. Although Bitter et al. (1972) acknowledge that their results are dependent on the order of the contact methods, they conclude that telephone surveys are particularly useful, followed by face-to-face interviews, and that mail contacts are ineffective. Cook (1977) reviewed the literature on client satisfaction studies and concluded that "telephone surveys are the most effective and efficient method of post-service client contact". A study by Winter and Keith (1988) also supports the usefulness of telephone surveys. They had an 88% response rate to their telephone satisfaction study of 171 current and former rehabilitation patients. However, their 151 respondents included 113 former patients as well as 38 other household members.

These studies demonstrate various survey methods and the different types of samples that have been used in satisfaction studies. Regardless of the methods used, contacting former clients is especially difficult. Both Bitter et al. (1972) and the JWK Corporation (1980) found that approximately half of the former clients in their respective studies could not be contacted, mostly because they were no longer at their last known address.

Methods to insure respondents' confidentiality and to eliminate response bias in vocational rehabilitation satisfaction studies have not been as well reported in the literature. In most cases, organizations or researchers independent of the rehabilitation agency have apparently collected and analyzed the data (Greenley & Schoenherr, 1981; JWK International Corporation 1980). The JWK Corporation (1980) did discuss the informed consent procedures it used. Former clients were assured in writing of confidentiality and that their individual responses would not be linked to them. Even with these assurances, response bias can be a concern. Respondents may be those who were particularly pleased or displeased with services, and there is concern about acquiescent or socially desirable responses (Cook, 1977). Anonymous mail surveys may produce responses that more closely reflect the respondents' true feelings, but their typically low response rates (Kerlinger, 1986) make their representativeness suspect. Telephone surveys and face-to-face interviews may increase response rates but are more subject to interaction effects since the client is not totally anonymous and the interviewer may affect the clients' responses.

Likert-type scales have been used most frequently to collect data about client satisfaction (see Greenley & Schoenherr, 1981; Patterson & Leach, 1987), but some studies have used combinations of types of questions such as Likert-type and "yes or no" or open ended questions (Reagles et al. 1970; Roessler & Bolton, 1985; Winter & Keith, 1988; also see Cook, 1977 on questionnaire formats).

Findings of VR Client Satisfaction Studies

Most studies indicate high levels of satisfaction with VR services (Greenley & Schoenherr, 1981; JWK International Corporation, 1980; Moriarity, 1977; Winter & Keith, 1988), but clients who are successfully rehabilitated have reported greater satisfaction (JWK International Corporation, 1980).

A few studies indicate less overall satisfaction with services. Hills and Hills (1975) found that in Vermont only 34% of clients credited vocational rehabilitation services with placing them in a better job than they could have found without this assistance, and Roessler and Bolton (1985) reported that only 54% of their respondents perceived vocational rehabilitation services positively. JWK International Corporation (1980) found that one-third of clients wanted services that had not been provided to them, and Roessler and Bolton' s (1985) study indicated that 79% of clients "voiced concerns about the quality of vocational rehabilitation training services",

Additional information on the relationship between client characteristics and client satisfaction comes from Winter and Keith's (1988) study. Unlike several of the studies noted herein which involve state vocational rehabilitation counselors and programs, Winter and Keith (1988) studied client satisfaction with outpatient services of a private, non-profit rehabilitation hospital. Findings were that older clients were more satisfied with specific aspects of services, while more educated clients and clients with an earlier onset of disability were less satisfied. However, these variables, as well as gender, were not related to level of overall client satisfaction. Respondents generally felt their appointments were scheduled conveniently. Those who thought they needed additional therapy were less satisfied with services.

Findings of Reagles et at. (1970) were that the only relationship of "pre-rehabilitation" variables to client satisfaction was a positive correlation with motivation for employment. The "rehabilitation process" variables they identified as positively related to satisfaction were total cost of a client's rehabilitation, total number of contacts with or on behalf of the client and total time spent in making these contacts. Many of the individual contact and time variables they studied were also positively related to satisfaction. At the end of services, earnings at closure were positively related to satisfaction, and clients placed in competitive employment were more satisfied than those in non-competitive employment. "Follow-up" variables positively related to satisfaction were being employed, hours worked, ability to support one's family, family's support for seeking services, and perceiving one's disability as not precluding employment.

Greenley and Schoenherr's (1981) research indicates that some personal characteristics of clients (adult household roles and expectations of services, but not sociodemographic characteristics) are the most important predictors of client satisfaction (defined in their study as "humaneness of services"). They also found that organizational variables, such as lack of bureaucratic rigidity, also contribute positively to satisfaction.

The study reported in the following pages pertains to the Texas Rehabilitation Commission's efforts to determine clients' satisfaction with the vocational rehabilitation services they received; the relationship of service delivery variables, work status variables, closure status variables, and client characteristics to client satisfaction; and factors that explain the variance in client satisfaction.

The Current Study

A Quality Task Force was appointed by the Texas Rehabilitation Commission to devise a system for measuring quality rehabilitation, including an instrument to measure client satisfaction. The task force began by sending a request to all state VR agencies to identify current systems used to measure quality rehabilitation. Fifteen states responded. All 15 responses contained an assessment of client satisfaction, indicating the need to include this component to insure credibility of any system purporting to measure quality rehabilitation. However, the methods these states used to obtain the information varied, and no agency indicated that its approach had been utilized on a regular basis or that it had a particularly unique or superior system. The TRC decided, therefore, to develop its own satisfaction instrument.

Method

The instrument initially developed by the task force was revised three times. The final version contained 11 questions. The response set for the first five questions was a 3-point Likert-type scale: seldom, some of the time, and most of the time. Questions 6 through 8 were answered yes or no. The question 9 response set contained three categories: 1-4 months, 5-8 months, or 9 or more months. Question 10 had a 5-point Likert-type scale ranging from very satisfied to very unsatisfied. Question 11 was open ended. This analysis focuses on the five service delivery questions, items 1 to 5, and the satisfaction question, item 10.

The first instruments were pre-tested using mail surveys due to the economy of this approach, but response rates of approximately 9% to 11% were unsatisfactory. In order to increase the return rate, a pilot telephone survey was attempted. A random sample of 139 closed cases provided a 43% return rate. This much improved response rate led to a decision to adopt the telephone method for future use. The telephone survey was more costly and time consuming, but the agency's desire to conduct the study led to a decision to allocate the necessary funds and to address the inconveniences in the hope of increasing client participation. In addition, the quality of information obtained from the clients may have been improved with the telephone method, because the interviewers were able to clarify questions and to insure that questions were not skipped and that answers were recorded correctly. Telephone surveys are also more personal than mail surveys, and clients may find it more difficult to turn down a caller than to throw away a mail survey.

The desire to match clients' responses to the satisfaction survey with their records in the TRC computerized data base in order to determine the correlates of client satisfaction precluded the use of an anonymous survey. While the anonymity provided by a mail survey might have resulted in more valid responses, the return rate would likely have been poor. An anonymous survey would also have required asking clients about information that is already in the agency' s data base. The approach used greatly minimized the time the client was asked to contribute to the study. Client confidentiality was respected through use of external interviewers and external consultants who analyzed the data. Staff of the TRC were not provided with information that could be linked to any individual client.

In 1991, the TRC contracted with a rehabilitation facility in San Antonio to conduct the client satisfaction interviews via telephone. Interviews were conducted by clients of the facility. The interview schedule and procedures were approved by the Commissioner of the TRC. The interviewer began the call by saying:

I am _____ calling from the Texas Rehabilitation Commission.

May I speak with _____?

I am following up on some clients to find out how TRC can better provide services and help its clients return to work.

Have you ever received vocational rehabilitation services from the Texas Rehabilitation Commission? Would you be willing to help us by taking a few minutes to answer some questions about your experience with TRC?

If the former client agreed, the questions found in Table 1 were asked. While all responses were treated as confidential, confidentiality procedures were not discussed with clients unless they requested this information. At the end of the calls the interviewers thanked respondents and gave them the TRC toll free number in case they had any questions about service or wanted anything else from the agency.

In 1992, the rehabilitation facility conducting the interviews gave the data and the respondents' client numbers to the staff of the University of Texas (UT) at Austin, the contract organization responsible for tabulation and data analysis. The UT staff matched the clients' responses to the satisfaction survey with the clients' records in the TRC computerized data base. The School of Social Work received approval through the University of Texas at Austin, Internal Review Board procedures to participate in the study.

Sample

Clients were initially selected using simple random sampling of closed cases in status 26 (successfully rehabilitated) or status 28 (not successfully rehabilitated) from the previous fiscal year. The cases in status 26 were closed between October 1, 1990 and April 30, 1991; those in 28 status were closed between October 1, 1990 and May 31, 1991. In order to increase the number of respondents, attempts were eventually made to telephone all 15,684 clients whose cases were closed during these time periods. A total of 2,400 clients participated, providing an effective response rate of 15.3% of this population. The number of respondents per counselor caseload varied from zero to 22. The analysis reported in this study can only, therefore, be used to describe client satisfaction in the aggregate. It is not an analysis of client satisfaction and individual counselor performance.

The problems encountered in obtaining respondents led to questions about whether the sample was representative of the population from which it was drawn. To evaluate this, chi-square goodness-of-fit tests were used to examine the pattern of similarity between the sample and the population on variables of gender, race, Hispanic ethnicity, age, marital status, highest grade completed, monthly family income, type of primary disability, severity of primary disability, and 26 or 28 closure status. The chi-square tests were conducted on the proportion of clients for each value of the variables rather than actual client counts, because of the sensitivity of chi-square to large frequencies which are present in this analysis.

The only variable tested for which the sample proportion was not representative of the population proportion was case closure status. In the sample, 72.6% of the clients were in status 26 (successfully rehabilitated) and 27.4% were in status 28 (not successfully rehabilitated) compared to the client data base in which 63.1% were in status 26 and 36.9% were in status 28 (X2=3.88, 1 DF, p=.0489). This difference in the pattern of responses for cases closed in status 26 and those closed in status 28 is taken into account in the analyses described below.

Results

Survey responses. Responses showed that clients were generally satisfied with the services of the TRC. Our discussion concerns responses to the first five survey items and item 10 on overall satisfaction, all Likert-type questions. We begin by discussing questions one through five, starting with the question that received the most positive responses and ending with the question that received the fewest positive responses. (1) Of all clients, 92.8% responded that they had been treated with courtesy and respect "most all of the time" with slightly more of the successfully rehabilitated clients (94.3%) feeling this way than clients who were not successfully rehabilitated (88.6%). (2) Regarding the scheduling of appointments, most clients (85.2%) felt their appointment had been scheduled promptly "most of the time" with slightly more successfully rehabilitated clients (87.1%) than unsuccessfully rehabilitated clients (80. 1%) giving this response. (3) To the question, "Do you feel your counselor took time to listen to your needs?," the majority of clients (84.8%) felt this had happened "most of the time" with 87.6% of the successfully and 76.9% of the unsuccessfully rehabilitated clients responding in this manner. (4) Of all clients, 80.9% felt that their calls had been returned "most of the time," with 83.3% of the successfully rehabilitated and 74.1% of the unsuccessfully rehabilitated clients responding this way. (5) Concerning client involvement in decision making about services, 79.5% of all respondents felt they had this involvement "most of the time" with 82.3% of successfully rehabilitated clients and 71.6% of the unsuccessfully rehabilitated clients giving this response.

To question 10 on clients' overall satisfaction or rating of their experience with the TRC, 78.2% said they were satisfied or very satisfied; however, 82.5% of the successfully rehabilitated clients were satisfied or very satisfied, compared with 66.2% of those who were not successfully rehabilitated."

Interitem Correlations. Correlation analysis was performed to examine the pattern of clients' responses to the questions. All correlations were positive (i.e., higher scores on one variable are associated with higher scores on the other variable). The largest correlation was between "needs listened to" and "involvement in decision making" (r=0.610). All other correlations were less than 0.60, suggesting a relatively weak relationship among the variables.

This pattern of correlations indicates that overall client satisfaction is a complex concept not captured by any single variable used in this study. Each question is apparently tapping a different aspect of client satisfaction, and clients did not simply give socially desirable responses or have an acquiescent response set in which all questions were answered with the same response.

Multivariate Analyses. A major question addressed in this analysis is whether or not client satisfaction is a function of factors associated with the service delivery process or factors related to clients that are beyond the control of the counselor. To the extent to which client satisfaction is a response to services provided by the counselor, it is reasonable to use client satisfaction as an element in evaluating counselor performance. The technique employed to evaluate this question is stepwise regression analysis. This approach is typically used to identify a subset of independent variables that is useful in predicting a dependent variable and to eliminate those independent variables which do not provide additional predictive utility (Tabachnick & Fidell, 1989, p. 147).

In the current analysis, the dependent variable was clients' overall satisfaction with the agency (question 10, "How would you rate your experience with TRC?," from the client satisfaction survey instrument). Independent variables included responses to the five service delivery questions from the client satisfaction survey as well as client characteristics from the TRC data base, including demographic characteristics such as gender, age, race,marital status, primary disability, and severity of the primary disability. Also included were variables related to services received and variables related to earnings and support at application and closure. Most variables were converted into dichotomous categories and coded as 0 or 1. The exceptions were age (in years), number of dependents, number in family, number of services provided by TRC, number of similar benefits provided, number of agencies providing services, total number of services or agencies providing services, vouchered amount on a client case, monthly family income in dollars, earnings at application, and weekly hours worked at application. These factors were measured as continuous variables. Closure status was forced to enter the regression analysis on the first step, because as noted above, the response pattern differed for status 26 and status 28 closures.

It was possible to match 2,136 of the 2,400 clients who responded to the satisfaction survey with the 15,684 records in the retrospective data base of closed cases compiled for use in this study in April 1992. Not all clients could be matched with their case record because the TRC computerized data base contains only the current status of clients, and some clients' statuses had changed since the study was conducted.

The question of whether or not client satisfaction is a function of factors associated with the service delivery process or factors beyond the control of the counselor can be answered by examining the order in which variables enter the regression equation. If the service delivery variables from the client satisfaction survey are statistically significant and enter before the variables from the client data base, then we may conclude that they are more important in explaining variations in the client satisfaction scores than are the variables from the client data base.

The results confirmed that the service delivery variables are the best predictors of overall client satisfaction. All five of the service delivery variables entered the model before any of the client data base variables, cumulatively explaining about 38% of the variance in client satisfaction ratings. An additional ten variables from the client data base passed the statistical test for inclusion in the model (vouchered amount on a client's case, gender, marital status, disability, and so forth), but they explained only an additional 2% of the variance in client satisfaction ratings.

In sum, a substantial proportion of the variance in client satisfaction ratings remains unexplained by the variables contained in the study, but there is clear evidence that the variance which can be explained is affected by service delivery variables. These service delivery variables may be influenced by the counselor, while many of the other variables studied, such as clients' demographic characteristics or the severity of their disabilities, cannot.

Discussion

Like some studies described in the preceding literature review, the instrument used in this study contained a combination of types of questions to assess client satisfaction. Most of the questions were Likert-type and are the focus of this discussion. As suggested by several authors, the telephone survey approach helped to increase the total number of responses over the mail survey; however, the response rate fell short of expectations. It was lower than other telephone surveys and face-to-face interviews that have been reported in the literature. This was probably due to several factors. First, clients were contacted one year after their cases were closed and they may have moved. Our experience was similar to that of other researchers who also found that many clients cannot be located for follow-up. In order to improve the chances of locating clients, the TRC is now contacting clients 90 days after their cases are closed to assess satisfaction. Second, counselors may not have been conscientious about updating home phone numbers at the time of case closure. In the future, more care might be taken to see if updating phone numbers improves response rates. Third, a decision was made not to contact clients at their current place of employment, and this may have also resulted in a lower than expected participation rate. The sample of respondents contained more successfully than unsuccessfully rehabilitated clients, but otherwise, the sample was similar to the general TRC client population. Like the majority of other studies, clients in our study generally reported high levels of satisfaction. Successfully rehabilitated clients were more pleased with the various aspects of service delivery and reported higher overall satisfaction than clients who were not successfully rehabilitated.

Like Greenley and Schoenherr (1981), Reagles et al. (1970), and Winter and Keith (1988), our study also indicates that sociodemographic variables are not strong predictors of overall client satisfaction. Some service intensity variables related to satisfaction that have been reported in other studies, such as amount spent on a client's case (Reagles et al., 1970), were also statistically significant in our multivariate analysis, but they explained very little of the variance in client satisfaction after the service delivery variables in our study were taken into account.

Also using regression analysis, Greenley and Schoenherr (1981) found that some client characteristics as well as some organizational variables explained 40% of the variance in client satisfaction. This is similar to the amount of variance explained by our model using service delivery and client data base variables. However, they found that certain client characteristics explained more of the variance in client satisfaction than organizational variables, while we found that service delivery variables explained more of this variance than client characteristics. The studies are, of course, not directly comparable because of the substantially different sets of variables used. The 60% of the variance not explained by their regression models or ours, however, does indicate that some of the major factors that explain client satisfaction have yet to be identified.

Reagles et al. (1970) found that five of the 14 items on their client satisfaction instrument, "the counselor was thoughtful and considerate," "the counselor listened," "the counselor understood the client's problems and feelings," "appointment times and places were convenient," and "the counselor was accessible at times other than regular appointments" did not discriminate well between clients with high and low overall satisfaction scores. They attributed this to the type of questions used which offered only "yes" or "no" response categories and to the tendency of clients to answer in a socially desirable manner. The content of our questions was similar, but questions were Likert-type which gave the client a greater range of responses, and clients in our study did not appear to answer in an acquiescent or socially desirable manner. Unlike Reagles et al., we found that involvement in decisions, needs "listened to, prompt scheduling, and courtesy and respect best explained the variance in client satisfaction.

Ju and Thomas (1987) concluded the report of their client satisfaction study by stating that "it is absolutely critical that counselors develop their listening and empathy skills and take time to get to know their clients". Our study also supports this conclusion. The practical implications of these findings are that VR professionals may be wise to recruit individuals into the field who place a high value on interpersonal relationships and that colleges and universities which prepare rehabilitation counselors may do well to emphasize interactions with clients. Although these ideas are not novel, the same suggestions may be made for continuing education for VR staff. Not only should factors that promote clients' satisfaction be mentioned, but VR education should provide ample opportunity for counselors to learn communication styles and methods that facilitate the involvement of clients in planning for their rehabilitation. Active listening skills in particular should be modeled and practiced. Skills such as these may be readily mastered by those in the field, but active listening involves more than its name implies. The same may be said for factors of courtesy and respect. Clients' expectations of what is courteous and respectful may vary depending on their gender, age, ethnicity, and so forth. The best way to determine this is to get to know the client. Our results also suggest that those VR counselors who will find their work particularly rewarding are those who take satisfaction in their relationships with clients.

These suggestions are consistent with the spirit of the Rehabilitation Amendments of 1992 (HR 5482) which require that the individualized written rehabilitation program (IWRP) be "jointly developed, agreed upon, and signed by" the client or the client's authorized representative and the VR counselor. Many VR agencies have already utilized these process to some extent, and the 1992 amendments further emphasize consumer involvement. States will eventually be required to report the extent of their compliance with new standards and indicators based on the 1992 amendments. Although the new standards are not due to be available for review and comment until September 30, 1994, they will be developed with input from state VR agencies, rehabilitation professionals, and consumers.

Many issues about client satisfaction ratings remain to be addressed, such as the extent to which clients' opinions are useful in counselor evaluations (Borderi, 1982; Downes et al., 1974; Ju & Thomas, 1987; McConnell, 1982; Parloff, 1983; Reagles et al., 1970), and exactly how client satisfaction should be used in assessing quality rehabilitation (Bordieri, 1982; Emener & Placido, 1982b). But it seems appropriate that the standards for the 1992 amendments reflect the broad definitions of quality described in the management literature and the operationalization of these definitions by agencies such as the Texas Rehabilitation Commission. The federal government can further demonstrate its commitment to consumer involvement by including measures of satisfaction as part of the standards and indicators it will promulgate for evaluating vocational rehabilitation agencies. Service delivery variables are apparently important factors explaining client satisfaction that may be under the control of vocational rehabilitation counselors. If quality management is to be adopted as a guiding principle of vocational rehabilitation, inclusion of client satisfaction as one part of regular counselor evaluation as well as for purposes of agency improvement will be needed.

References

Bassett, P.T. (1974). Measurement of outcomes. First institute on rehabilitation issues. West Virginia: West Virginia Rehabilitation Research and Training Center. Cited in D.W. Cook (1977). Guidelines for conducting client satisfaction studies. Journal of Applied Rehabilitation Counseling, 8, 107-114.

Bedient, M. (1982). Comments on the Emener, Placido research article. Journal of Rehabilitation Administration, 6, 76-77.

Bitter, J.A., Kunce, J.T., Lawyer, D.L., Miller, D.E., & Ray, C.C. (1972). Follow-up: A methodological note. Rehabilitation Counseling Bulletin, 16(2), 64-70.

Bordieri, J.E. (1982). Comments on the Emener, Placido research article. Journal of Rehabilitation Administration. 6, 77-78.

Cook, D.W. (1977). Guidelines for conducting client satisfaction studies. Journal of Applied Rehabilitation Counseling, 8, 107-114.

Downes, S.C., McFarland, F.S., & Alston, P.P. (1974). Survey of the NRCA membership regarding the basis for evaluating counselor performance. Journal of Applied Rehabilitation Counseling, 5(4), 196-200.

Emener, W.G., & Placido, D. (1982a). Rehabilitation counselor evaluation: An analysis and critique. Journal of Rehabilitation Administration. 6, 72-78, 90.

Emener, W., & Placido, D. (1982b). Client feedback: A valuable source of counselor development. Journal of Applied Rehabilitation Counseling. 13(1), 18-23.

Greenley, J.R., & Schoenherr, R.A. (1981). Organization effects on client satisfaction with humaneness of service. Journal of Health and Social Behavior. 22, 2-18.

Hills, W.G., & Hills, K.W. (1975). Consumer's measurement of vocational rehabilitation in Vermont fiscal year 1974. Norman, OK: University of Oklahoma Regional Rehabilitation Research Institute. Cited in Patterson, J.B., & Leach, R. (1987). Client satisfaction in vocational rehabilitation. Journal of Rehabilitation, 53, 40-43.

Howe, S.S., Minch, J.A., & Fay, F.A. (1980). Consumer involvement in rehabilitation: A national perspective. Journal of Rehabilitation. 46(3), 52-56.

Hutchins, D. (1986). Quality is everybody's business. Management Decisions. 24, 3-6.

Johansson, H., & McArthur, D. (1988). Rediscovering the fundamentals of quality. Management Review. 77, 34-37.

Ju, J.J., & Thomas, K. R. (1987). The accuracy of counselor perceptions of client work values and client satisfaction. Rehabilitation Counseling Bulletin. 30(3), 157-166.

JWK International Corporation. (1980, September). Follow-Up of Former Vocational Rehabilitation Clients. Annandale, VA: Author.

Kerlinger, F.N. (1986). Foundations of Behavioral Research. 3rd ed. New York: Holt, Reinhart and Winston.

Lorenz, J.R. (1979). Setting performance objectives and evaluating individual performance in rehabilitation settings. Journal of Rehabilitation Administration, 3(1), 5-13.

Lubeck, R.C., & Davis, P.K. (1991). W.E. Deming's 14 points for quality: Can they be applied to rehabilitation? Journal of Rehabilitation Administration, 15, 216-221.

McConnell, L.R. (1982). Comments on the Emener, Placido research article. Journal of Rehabilitation Administration. 6, 78, 90.

Moriarity, J.B. (1977). Issues in the evaluation of vocational rehabilitation. Professional Psychology, 8, 641-649.

Parker, R.M., & Thomas, K.R. (1980). Fads, flaws, fallacies, and foolishness in evaluation of rehabilitation programs. Journal of Rehabilitation. 46(1), 32-34.

Parloff, M.B. (1983). Who will be satisfied by "consumer satisfaction" evidence? Behavior Therapy. 14, 242-246.

Patterson, J.B., & Leach, R. (1987). Client satisfaction in vocational rehabilitation. Journal of Rehabilitation. 53, 40-43.

Reagles, K.W., Wright, G.N., & Butler, A.J. (1970). Correlates of client satisfaction in an expanded vocational rehabilitation program. Wisconsin Studies in Vocational Rehabilitation. Monograph XII, Series 2. Madison, WI: The University of Wisconsin Regional Rehabilitation Research Institute.

Roessler, R., & Bolton, B. (1985). Employment patterns of former vocational rehabilitation clients and implications for rehabilitation practice. Rehabilitation Counseling Bulletin. 28(3), 179-187.

RSA Region In Final Report. (1975). In S.E. Rubin (ed.). Studies on the evaluation of state vocational rehabilitation programs. Arkansas Rehabilitation Research and Training Center, pp. 297-307. Cited in Cook, D.W. (1977). Guidelines for conducting client satisfaction studies. Journal of Applied Rehabilitation Counseling. 8, 107-114.

Schwab, A.J., & Fenoglio, J.A. (1992). The Texas Rehabilitation Commission system for measuring quality rehabilitation. Journal of Applied Rehabilitation Counseling. 23(2), 21-25.

Tabachnick, B.G., & Fidell, L.S. Using Multivariate Statistics, 2nd ed., New York: Harper Collins Publishers, 1989.

United States Office of Personnel Management. (1992, May). Federal total quality management handbook, educating and training for total quality management in the federal government. 3. Washington, DC: U.S. Government Printing Office.

Vernon, M., Bussey, P., & Day, D.S. (1979). The "closure system" and accountability in vocational rehabilitation. Journal of Rehabilitation. 45(4), 45-47.

Winter, P.L., & Keith, R.A. (1988). A model of outpatient satisfaction in rehabilitation. Rehabilitation Psychology. 33(3), 131-142.

Table 1

Questions comprising the client satisfaction survey.

1. Were you treated with courtesy and respect by the office staff?

2. Did the people at the Texas Rehabilitation Commission promptly schedule appointments?

3. Do you feel your counselor took time to listen to your needs?

4. Did they return your phone calls?

5. Did you feel involved in making decisions about services that helped you get or keep a job?

6. Did the counselor provide you services that helped you get or keep a job?

7. Are you working now?

8. If so, is this job the same type of work you had when you were with TRC?

9. How long have you been working at this job?

10. How would you rate your experience with TRC?

11. What are one or two things that caused your satisfaction or dissatisfaction?

A. James Schwab, Ph.D., Center for Social Work Research, School of Social Work, The University of Texas at Austin, 2609 University Avenue, Austin, TX 78712.