Relationship of rehabilitation client outcome to level of rehabilitation counselor education.

The relationship between rehabilitation client outcome and level of rehabilitation counselor education was examined in a state VR agency. Outcomes were examined separately for clients with severe disabilities and those whose disabilities were not severe. For clients with severe disabilities, master's degree rehabilitation counselors (MRCS) achieved significantly better outcomes than their counterparts with bachelor's and unrelated master's degrees (B-UM). No significant group differences were found between MRCs and counselors with related master's degrees (RMs) although outcomes for the RM group were consistently between those of the MRC and B-UM groups. For clients with non-severe disabilities, there were no significant outcome differences among the counselor education levels.

Section 101(a)(7)(B) of the Rehabilitation Act was amended to require delivery of rehabilitation services by qualified personnel" Rehabilitation Act Amendments, 1986), however, debate continues regarding the definition of the term qualified" (Graves, Coffee, Habeck, & Stude, 1987; Walker & Myers, 1988). Presently, state vocational rehabilitation agencies typically hire individuals with varied types and levels of college degrees and work experience as rehabilitation counselors (Hershenson, 1988; Kuehn, Crystal, & Ursprung, 1988).

The variation in rehabilitation counselor hiring criteria may reflect the fact that although the rehabilitation counselor has been recognized as the major agent of the state-federal vocational rehabilitation (VR) program (Bolton, 1987), little research has shown a positive relationship between rehabilitation counselor competence and service outcomes for rehabilitation clients (Rubin & Beardsley, 1987). Some early studies did demonstrate connections between counselor characteristics or behaviors and client perceptions or outcomes (e.g., Ayer, Wright, & Butler, 1968; Jenkins, West, & Anderson, 1975; Rubin, Bolton, Krauft, Bozarth, & Richardson, 1974), however, more recent studies have been unable to establish relationships between counselor education and client outcome in state vocational rehabilitation (VR) agencies (e.g., Danek, 1978; Emener, 1980; Giesen & McBroom, 1986). Although these studies provided the foundation for the current study, the relative paucity of empirical evidence relating rehabilitation counselor education to rehabilitation client outcome may be more from methodological limitations rather than the actual absence of such a relationship. Potential limitations may have arisen from: (a) inadequacy of outcome measures, (b) different outcome patterns for clients with severe disabilities as opposed to those whose disabilities are classified as non-severe, and (c) failure to account for the potential interactive relationship of counselor education with counselor work experience in relation to client outcome.

The traditional 26 closure" criterion, although often used in previous research, has been criticized as a limited outcome measure (Bolton, 1987; Walls & Tseng, 1987), which does not include the following available VR information: (a) type of employment at closure (i.e., competitive employment, sheltered employment, homemaking) (Cook & Cooper, 1979), b) relative probability of rehabilitation (i.e., rehabilitation rate), and (c) cost of service delivery. Consideration of such additional information reveals different outcome patterns for persons with severe disabilities as contrasted with their peers whose disabilities are classified as non-severe. In 1985, only 73.3% of rehabilitated persons meeting the Rehabilitation Services Administration (RSA) definition for severe disability entered competitive employment as compared with 89.7% of rehabilitants with non-severe disabilities. In addition persons with severe disabilities were less likely to be rehabilitated than their counterparts with non-severe disabilities (overall rehabilitation rates of 62.2% vs. 67.4%) and more costly to serve (RSA, 1988).

An additional confounding factor has been the interactive relationship of years of counselor work experience with rehabilitation counselor education in relation to client outcome. Wright, Leahy, & Reidesel (1987) suggested a relationship between years of experience and counselor perceived competency. Jenkins, West, and Anderson (1975) and Graves, Bagley, and Chen (1985) suggested a relationship between years of counselor work experience and client outcome. Additionally, Kunce, Thoreson, and Parker (1975) suggested that the relationship of counselor experience to client outcome might be complicated by selection of male counselors with master's degrees in rehabilitation counseling who had high general ability test scores out of direct service positions to administration earlier in their careers than females or males with lower scores.

A preliminary study by Szymanski and Parker (1989) addressed the above sources of potential limitations and demonstrated a significant relationship between level of counselor education and client outcome for clients with severe disabilities in one state vocational rehabilitation agency. Although the current study focused on the same state VR agency, it was much broader in scope. The preliminary study included only clients with severe disabilities and only one dependent variable. The current study, although it involved the same state VR agency and the same type of research design (aptitude-treatment-interaction), differed in the following ways:

1. All closures were considered, not just those of clients with severe disabilities.

2. Severity of client disability was used to partition the sample so that outcomes could be compared for the two client groups (clients with severe disabilities and those with nonsevere disabilities).

3. Three dependent variables were considered. Thus the current study afforded a much more detailed investigation of the relationship between rehabilitation client outcome and rehabilitation counselor education than the previous study which was preliminary in nature. The nature of the current findings, which are described in the following sections, further support the potency of the methodological limitations discussed in this section.

Method Participants

Participants in this study were counselors employed by the New York State Office of Vocational Rehabilitation (NYSOVR) and their clients whose cases were closed during the period from April 1, 1986 to March 31, 1987. During the Fall of 1987, a routine questionnaire was sent by the NYSOVR staff development and training office to professional field staff to determine job title, educational attainment, area of concentration, certification status, years of agency work experience, and specialty caseload experience. Agency assigned counselor identification numbers, rather than names, were used to identify respondents and match counselor data with client data. Usable questionnaires were obtained from 273 of 360 caseload carrying counselors (a 76% response rate), however, the further match of counselor data with client closure data resulted in only 238 complete counselor profiles (66%). These 238 counselors closed a total of 19,114 clients during the reporting period, including 8,808 who were classified as severely disabled.

Counselors were categorized into one of three groups according to level of education: (a) master's degrees in rehabilitation counseling (MRCs), (b) related master's degrees (RMs), or (c) bachelor's and unrelated master's degrees (B-UMs). Master's degrees considered related included guidance and counseling, agency counseling, and any counseling or special education related disciplines. Level of rehabilitation counselor education was distributed as follows: 122 MRCS, 52 RMs, and 64 B-UMS. Counselor years of agency work experience ranged from less than one year to 30 years with the following group means for each counselor educational category. MRCs 11.6, RMs 10.9, and B-UMs 12.0 years. Variables

The independent variable was level of counselor education. Length of counselor agency work experience was considered a mediating variable. Severity of client disability, a dichotomous variable, was used to partition the client sample into two mutually exclusive groups (clients with severe disabilities). and clients with non-severe or non-classified disabilities). The following dependent variables (DVs) were chosen for analysis: (a) competitive closure rate (CCR), which was computed from the ratio of competitive employment status 26 closures to afl closures including status 26 (rehabilitated) closures, and closure statuses 08 (closed from applicant status), 30 (closed before individual written rehabilitation plan [WRP] initiated), and 28 (closed not rehabilitated after IMP initiated); (b) number of non-competitive closures (NCC), which was computed as the sum of all closures in statuses 08,28, and 30 added to the sum of status 26 homemaker and sheltered employment closures; and (c) net case service encumbrances (ENCUM), which was the per counselor, total of case service dollar encumbrances for cases which were closed non-competitive during the reporting period. Data Analysis

The following null hypotheses were tested using an aptitudetreatment-interection (ATI) design (Borich, 1986; Pedhazur, 1982) which controlled for years of counselor work experience by inclusion in the design: There are no differences between MRCs and RMs or MRCs and B-UMs on the following measures:

1. CCR for clients with severe disabilities.

2. CCR for clients with non-severe disabilities.

3. NCC of clients with severe disabilities.

4. NCC of clients with non-severe disabilities.

5. ENCUM for non-competitive closures of clients with severe disabilities.

The ATI design employs the Johnson-Neyman technique to determine regions of statistically significant difference between the regression lines of two groups for a dependent variable while controlling a mediating variable (Pedhazur, 1982). Borich, Godbout, and Wunderlich's (1976) ATILINIP computer program was used to perform the ATI analyses. Where significant differences between the groups were found, the results were reported as regions of significance, that is, the range of years of work experience within which the difference between the two educational groups was statistically significant.

Results

The unstandardized regression coefficients and Y intercepts for the regression equations indicating the relationship of the dependent measures to counselor years of agency work experience for each level of rehabilitation counselor education are reported in Table 1. These equations were used in the pairwise ATI comparisons which are reported in Table 2.

The three analyses which addressed service delivery outcomes for clients with severe disabilities afl yielded significant results. No significant results were obtained in the two analyses which measured service delivery outcomes for persons with non-severe disabilities. The MRC counselors were shown to have significantly higher CCRs for clients with severe disabilities than B-UM counselors from the beginning of their agency service through 10.48 years of tenure. The difference between the MRC and RM counselors on CCR for clients with severe disabilities failed to reach statistical significance through the range of counselor years of agency experience; although the performance of the RMs remained lower than that of the MRCs through 11 years of agency tenure. In both of the CCR comparisons the regression lines for the two groups intersected well within the range of data, indicating disordinal interactions.

MRCs were found to have significantly fewer NCCs of clients with severe disabilities than their B-UM counterparts from 4.54 through 14.89 years of agency experience. The lower boundary of 4.54 years may be the result of counselors being assigned smaller caseloads in the beginning of agency service. The group difference between MRCs and RMs failed to reach statistical significance in the same comparison, although MRCs had fewer NCCs than RMs through 14 years of agency experience.

MRC counselors were found to have significantly lower ENCUMs for non-competitive closures of clients with severe disabilities than their B-UM counterparts from the beginning of agency service through 14.84 years. As in the previous comparisons, the group difference between MRCs and RMs failed to reach statistical significance on the same dependent measure, although the MRCs had lower ENCUMs than the RMs throughout the range of counselor work experience. Comparisons of the MRC group with both the B-UM and RM groups failed to reach statistical significance on either of the comparisons addressing service delivery to clients with nonsevere disabilities. Thus, no group differences were found in relation to CCR and NCC for clients with non-severe disabilities.

Discussion The results suggest that counselors with master's degrees in rehabilitation counseling are not only more effective (i.e., achieve higher competitive closure rates and lower numbers of non-competitive closures) but also more cost-efficient (i.e., encumber fewer case service dollars on non-competitive closures) in serving clients with severe disabilities than their B-UM counterparts for a substantial portion of their agency tenure. These results suggest a need for a VR agency policy that B-UMs be required to pursue a master's degree in rehabilitation counseling. For example, the Connecticut VR agency either hires master's degree counselors or requires those hired with only bachelor's degrees to matriculate and earn master's degrees in rehabilitation counseling or a related discipline within three years. In the interim, such individuals are given reduced caseloads and assigned to experienced supervisors (M. Campbell, personal communication, August 1, 1988).

The results of the tests of group differences on net case service dollar encumbrances for non-competitive closures of clients with severe disabilities suggest that such policies might result in decreasing case service dollars spent on non-competitive closures of clients with severe disabilities. The difference in the mean net case service dollar encumbrances for non-competitive closures of clients with severe disabilities (ENCUM) of the MRC and B-UM groups amounted to $16,337 per counselor per year.

The results also suggest that agency supervisors should focus on monitoring closures and cases in process and intervene when inappropriate services are planned rather than simply evaluating total numbers of clients closed in each status. This is particularly important for B-UM counselors serving clients with severe disabilities.

The results of this study also have implications for rehabilitation research. The complex nature of the relationships of level of rehabilitation counselor education, years of counselor work experience, and client outcome resulted in disordinal interactions of regression lines for the dependent variable (DV) CCR. Although interactions were more pronounced with CCR, interactions also were found with other DVs for clients with severe disabilities as indicated by the Table 1 regression equations. It is suggested that researchers investigating the relationship of counselor education with client outcome consider ATI designs which account for such interactions.

The limitations of traditional VR outcome measures are highlighted by the results of this study coupled with nationwide closure data. As noted earlier, only 73.3% of clients with severe disabilities who are rehabilitated enter competitive employment as contrasted with 89.7% of those with nonsevere disabilities (RSA, 1988). Thus, failure to differentiate clients by severity of disability and work status at closure can obscure existing relationships between counselor education and client outcome. Limitations The results of the current study are limited by threats to both external and internal validity. External validity is limited in that the study examined only New York State data and thus is not generalizable beyond NYSOVR without replication. Internal validity was limited by the nature of the experimental design, rival hypotheses, and distributional deviations.

The quasi-experimental research design statistically controlled her than manipulated the independent and mediating variables, therefore, only relationship and not causality can be inferred from the results (Bolton & Parker, 1987). Education and years of experience may serve as markers for other variables such as counselor attitude or orientation which were not considered in the study.

Rival hypotheses involving caseload size, caseload percentage of clients with severe disabilities, and group differences on counselor years of experience were tested as potential threats to internal validity and revealed no significant group differences (see Szymanski, 1988). Deviations of variable distributions from normality were also considered, however, multiple linear regression and related techniques used in this research are known to be robust in the presence of such deviations for large samples (Harris, 1975; Pedhazur, 1982).

Summary

For clients with severe disabilities, counselors with master's degrees in rehabilitation counseling achieved significantly better outcomes than their counterparts with bachelor's and unrelated master's degrees throughout the following intervals of counselor agency service: (a) zero through 10.48 years for competitive closure rate, b) 4.54 through 14.89 years for non-competitive closures, and c) zero through 14.84 years for net case service dollar encumbrances for non-competitive closures. No significant group differences were found between MRCs and counselors with related master's degrees although outcomes for the RM group were consistently between those of the MRC and B-UM groups.

The relationship between rehabilitation counselor education and rehabilitation client outcome needs further investigation. An immediate priority might be a multiple state replication of this study. In the interim, however, serious consideration should be given to eliminating the practice of hiring persons with bachelor's and unrelated master's degrees to serve as rehabilitation counselors unless such individuals are required to pursue formal education in rehabilitation counseling and afforded experienced supervision.