Craig Ravesloot, Ph.D. and Tom Seekins, Ph.D., send email, Research and Training Center on Disability in Rural Communities, The University of Montana Rural Institute, Missoula, Montana. A version of this article was accepted for publication by the Rehabilitation Counseling Bulletin.

---

Quality of life is intimately related to an individual's major life activities (Antonovsky, 1987). For many individuals, productive employment is of central importance. Unfortunately, people with disabilities confront many difficulties in finding or keeping work; with overall unemployment rates for this population as high as 50%, and as high as 60% for people with disabilities living in rural areas (Rural Institute on Disabilities, 1993).

The national system of vocational rehabilitation (VR) services represents one of our nation's commitments to assisting people with disabilities achieve their goals of productive employment (Rehabilitation Act Amendments, 1992). Historically, the focus of vocational rehabilitation has been preparing employees to work at jobs traditionally available in the private and public sectors. In 1988, the system of state-administered VR programs, on a budget of $1.5 billion, helped approximately 650,000 individuals become employed. This represents only about 5-7% of the eligible population, however (General Accounting Office, 1993).

During this same period, the structure of the U.S. labor market began undergoing significant changes in a more competitive global economy. Among these changes is a shift toward more contingent employment arrangements (Belous, 1989). Contingent employment includes temporary, part-time, subcontracted, and self-employment.

Indeed, self-employment is one of the fastest growing employment options (Becker, 1984; Silvestri, 1991). Interestingly, Seekins (1992) reports that people who report a work disability report being self-employed at nearly twice the rate of the general population (14% vs. 8%). Unfortunately, the trend in the growth of self-employment is not reflected in vocational rehabilitation practice nor can vocational rehabilitation services account for the surprising rate of self-employment among those who report a work disability. The Rehabilitation Services Administration (1988) reports that only about 2-3% of its annual case closures are to self-employment, ranging from .53% of cases in California to 7.34% of cases in Vermont. This disparity between national employment trends and rehabilitation practice may be most detrimental to individuals living in rural areas, where self-employment is nearly twice as prevalent as in urban areas and may represent the most viable employment option in some cases.

The VR system is federally sponsored, but administered by the various states (Rehabilitation Act Amendments, 1992). Federal policies set broad guidelines and regulations that are then interpreted and applied by states. VR counselors operating in a state system have a great deal of discretion within a broad framework that calls for the development of individualized written rehabilitation plans (IWRP). Theoretically, counselors and consumers are free to pursue a wide range of employment options (e.g., competitive employment, supported employment, sheltered employment, and self-employment). From a service perspective, counselors make many judgments and recommendations that will influence the course of a consumer's rehabilitation choices, the quality of the person's work life, and ultimately, the quality of life in general. Ideally, counselors focus on the needs and goals of consumers in the development of IWRPs. Yet, some of their recommendations may vary systematically (General Accounting Office, 1993). For example, Seekins (1992) has shown that the use of self-employment as a VR case closure is more likely to be used by counselors in rural states.

One cause of variations in recommendations and services may involve attitudes that develop among the staff of a program (e.g., Schein, 1979). Employee attitudes are often shaped by other employees, supervisors, and the general work climate. These attitudes can directly influence the interpretation and application of organizational policy (e.g., Griffin & Moorhead, 1986 p. 208). Even if inaccurate, staff assessments of supervisor's attitudes about particular issues can significantly affect performance (e.g., Schein, 1979). In vocational rehabilitation, for example, even though self-employment is a legitimate option, supervisors may convey or staff may develop the attitude that competitive employment in existing jobs is preferred or that the use of self-employment as an option is discouraged.

Research into the factors that affect use and non-use of self-employment rehabilitation strategies can focus on a number of different areas. For instance, state policies governing the use of self-employment may facilitate or restrict its use by counselors (Arnold & Seekins, 1994). Likewise, availability of funding and other resources for self-employment strategies certainly affect its use. Alternatively, the content of training materials for VR counselors may be investigated to determine whether they are adequate in the area of self-employment strategies.

While these factors undoubtedly contribute to the use of self-employment strategies, the focus of this study is the attitudes of VR counselors. The purpose of the present study was to investigate the development and effects of VR counselors' attitudes toward self-employment as a vocational rehabilitation strategy. It is one part of a larger study on the use of self-employment as a VR closure in rural rehabilitation.

Methods

Subjects

One hundred VR counselors were selected from states that demonstrated unusually high or low self-employment closure rates according to statistics collected by the Rehabilitation Service Administration (RSA) in 1988. Of the VR counselors in each of these states, the 5 with the highest and 5 with the lowest self-employment closure rate during the 1990 calendar year were randomly selected. Seventy-eight of the 95 counselors contacted returned usable surveys for an 82% response rate.

Instruments

The self-employment rehabilitation survey was constructed by initially interviewing nine VR counselors serving rural areas. These individuals were interviewed to profile the use of self-employment rehabilitation strategies. Interviews were semi-structured with open ended questions and lasted approximately 45 minutes. The interview data were used to write items concerning areas that were contradictory or otherwise unclear following the interviews. Items were structured with a nine-point Likert-type scale anchored at both ends. Various service providers and other VR professionals reviewed these items during the survey construction process.

Counselor attitudes toward the self-employment rehabilitation strategy were assessed using the semantic-differential technique (Nunally, 1978). This technique uses bipolar adjective pairs such as "Successful-Unsuccessful" on a Likert-type scale to measure a concept directly. For this survey, counselors were asked to "describe how you usually feel about self-employment as a vocational rehabilitation strategy" using adjective pairs. The final survey contained 42 items including the 11 semantic-differential items listed in Table 1.

Table 1 Description

Semantic Differential Adjective Pairs Used to Measure Counselor Attitudes

	1	2	3	4	5	6	7	8	9
Inefficient										Efficient
Unsuccessful										Successful
Inflexible										Flexible
Risky										Promising
Unrealistic										Realistic
Difficult										Easy
Unfamiliar										Familiar
Confusing										Clear
Complex										Simple
Slow										Fast
Expensive										Inexpensive

Results

All data were entered and checked for accuracy by two research assistants working independently. Discrepancies in data entry were checked and corrected. Both univariate and multiple regression analyses were performed to test for the predicted relationships. For each regression analysis, scatterplots of the regression residuals, predicted values, and independent variables were inspected.

A priori predictions were made regarding the content of attitude scales and their relationship to other variables in the self-employment survey. The items selected to measure counselor attitudes toward self-employment rehabilitation strategies were selected to measure two components: evaluation and familiarity (Nunally, 1978). Both the evaluation and familiarity scales were predicted to covary with items judged to be precursors to attitudes and effects of attitudes. Table 2 contains these items.

The items predicted to measure the evaluation and familiarity components were correlated to assess the covariation of items before summing them into scales. Table 3 contains inter-correlations of items for both attitude scales.

---

Table 2 Description

Items Hypothesized to be Predictive of, and to Show Effects from, Counselor Attitudes

Items hypothesized to predict attitude development

1. My past experiences with or observations of self-employment as a rehabilitation strategy have been positive. (Agree-Disagree)
2. My department's policies and procedures facilitate the use of self-employment rehabilitation strategies. (Agree-Disagree)
3. The atmosphere in my office is generally pessimistic toward self-employment rehabilitation strategies. (Agree-Disagree)

Items hypothesized to show effects from counselor attitudes

1. Self-employment should be as available to a client as any other rehabilitation strategy. (Agree-Disagree)
2. The state's investment in self-employment (e.g. equipment and supplies purchased) should be carefully protected. (Agree-Disagree)
3. In general, how would you rate your personal reaction (not what you say to the client) to a client's inquiry about self-employment.   (Positive-Negative)
4. In general, how do you think most clients find the processes they must go through in pursuing self-employment (i.e. paperwork, completing a business plan, etc.)? (Discouraging-Encouraging)

¹Items were presented with a nine-point likert-type scale anchored by the words in parenthesis (e.g. agree-disagree).

---

Table 3 Evaluation and Familiarity Scales Item Intercorrelations

	Efficient	Successful	Flexible	Promising	Realistic	Easy	Quick	Inexpensive	Familiar	Clear
Evaluation
Efficient	1.000
Successful	0.792	1.000
Flexible	0.386	0.346	1.000
Promising	0.451	0.654	0.209	1.000
Realistic	0.474	0.630	0.283	0.577	1.000
Easy	0.455	0.570	0.343	0.644	0.458	1.000
Quick	0.420	0.513	0.395	0.622	0.278	0.568	1.000
Inexpensive	0.373	0.496	0.338	0.341	0.302	0.536	0.397	1.000
Familiarity
Familiar	0.472	0.548	0.415	0.392	0.481	0.588	0.341	0.429	1.000
Clear	0.348	0.320	0.492	0.304	0.365	0.402	0.412	0.491	0.642	1.000
Simple	0.427	0.397	0.291	0.464	0.251	0.674	0.529	0.445	0.585	0.562

Table 4 contains the descriptive statistics and reliability of each attitude scale. Inspection of the boxplots indicated that each scale approximates multivariate normality. Overall, counselors evaluated self-employment neutrally (mean rating = 4.6) but were slightly unfamiliar with it (mean rating = 4.0). Coefficient alpha for each scale indicated that both scales possessed sufficient reliability. However, following inspection of the correlation between the evaluation familiarity scales (.71) and the correlation of each scale with all items, the familiarity scales was dropped from further analysis.   Based on these indices, the familiarity scale did not appear to add to the usefulness of the evaluation scale.

Table 4  Descriptive Statistics of Evaluation and Familiarity Scales

	N	Mean	Median	SD	Min	Max	Alpha
Evaluation	76	36.82	37.00	11.51	9.00	60.00	87.0
Familiarity	75	12.49	12.00	4.90	3.00	27.00	82.0

Initially, the Self-Employment Evaluation Scales (SEES) was used to predict the number of self-employment closures counselors estimated they had annually.  These results indicated that counselors' attitudes accounted for 11% of the variance in counselors' self-reported closure rate to self-employment.  This relationship was further demonstrated by comparing counselors in the 1st and 4th quartile on the evaluation scale for the number of self-employment closures they estimated having annually.  The mean number of self-employment closures for these two groups (.67 and 2.33) differed significantly (t ₁₈=-2.45, p<.025) suggesting a correspondence between the counselors' attitudes toward self-employment strategies and their use of them.

Next, each of the variables conceptualized as precursors to attitude was used to predict counselors' attitudes.  These results, presented in Table 5, indicate that each of the variables has a significant impact on counselors' attitudes.

Table 5  Univariate Regression of Self-Employment Evaluation Scale Scores on Precursor Variables

1. Past Experiences
SEES = 21.4 + 2.90 past experience
SOURCE	DF	SS	MS	F	p
Regression	1	3605.2	3605.2	46.17	0.000
Error	73	5700.2	78.1
Total	74	9305.4
R-sq = 38.7% 2. Policy Facilitates SE
SEES = 27.1 + 1.64 policy facilitates
SOURCE	DF	SS	MS	F	p
Regression	1	1313.1	1313.1	11.29	0.0001
Error	73	8489.6	116.3
Total	74	9802.7
R-sq = 13.4% 3. Office Atmosphere
SEES = 48.9 - 2.483 office atmosphere
SOURCE	DF	SS	MS	F	p
Regression	1	3180.41	3180.41	38.13	0.000
Error	72	6005.59	83.41
Total	73	9186.00
Rsq = 34.6%

To assess the collective impact of these variables on counselor attitude, the variables were entered into a stepwise multiple regression procedures.  These results are listed in Table 6.  The counselors' past experience variable entered the equation first, accounting for 38% of the variance and office atmosphere entered next, accounting for an additional 11% of the variance.  Collectively, these variables accounted for 49% of the variance in counselors' attitudes.

Table 6  Multiple Regression of Precursor Variables on SEES Scores

SEES = 33.69 - 1.601 office atmosphere + 2.039 past experience

SOURCE	DF	SS	MS	F	p
Regression	2	4548.48	2274.24	34.82	0.000
Residual	71	4637.52	65.32
Total	73	9186.00
R-sq = 49.5%

This stepwise procedure did not result in significant direct effects for counselors' ratings of their states' policies on attitude. However, indirect effects were also investigated. Investigation of indirect effects for state policy was accomplished by looking at the impact of policy on office atmosphere. These results are presented in Table 7. Counselors' opinion regarding whether their states' policies facilitate the use of self-employment strategies accounted for 30% of the variance in their rating of their offices' atmosphere regarding self-employment strategies. These results suggest that state policy has a significant impact on the office atmosphere toward self-employment. Thus, a state's policy has an indirect effect on a counselor's attitudes toward self-employment strategies.

Table 7 Univariate Regression of Office Atmosphere on State Policy

office atmosphere = 8.24 - 0.581 policy

Source	DF	SS	MS	F	p
Regression	1	157.39	157.4	31.60	0.000
Error	72	358.57	74.98
Total	73	515.96
R-sq = 30.5%

Table 8  Univariate Regression of Evaluation Scale Scores on Effect Variables

1. SE availability
availability = 4.303 + .0733 SEES
SOURCE	DF	SS	MS	F	p
Regression	1	49.306	49.306	10.586	0.000
Error	72	335.680	4.662
Total	73	384.986
R-sq = 12.8% 2. State's Investment
state's investment = 9.02 - 0.0523 SEES
SOURCE	DF	SS	MS	F	p
Regression	1	27.202	27.202	9.44	0.003
Error	74	213.153	2.880
Total	75	240.355
R-sq = 11.3% 3. Personal Reaction
personal reaction = 0.749 + 0.126 SEES
SOURCE	DF	SS	MS	F	p
Regression	1	157.20	157.20	47.53	0.000
Error	74	244.74	3.31
Total	75	401.93
R-sq = 39.1% 4. Process
process = 28.7%
SOURCE	DF	SS	MS	F	p
Regression	1	88.505	88.505	29.72	0.000
Error	74	220.376	2.978
Total	75	308.882
R-sq = 28.7%

Results for the variables expected to show effects from the counselors' attitudes are presented in Table 8. The variables showing the largest effects were the counselors' personal reactions to a consumer's inquiry regarding self-employment and their expectations of the consumer's experience in pursuing self-employment closure. The more positively counselors evaluated self-employment as a strategy, the more positive was their reaction to inquiries about self-employment. Additionally, the more positive their evaluation, the easier they rated the process of pursuing a self-employment closure for consumers.

Finally, the variables predicted to show effects from counselors' attitudes were summed into a scale. It is argued that the sum of these items represents the counselors' estimation of how facilitative the VR process is for consumers pursuing self-employment. Inspection of the boxplot produced for this variable indicated that the scale approximates mulivariate normality. Next, the SEES was used to predict this self-employment facilitation variable. The results, presented in Table 9, indicated that over 38% of the variance in self-employment facilitation was accounted for by VR counselors' attitudes.

Table 9 Univariate regression of counselor facilitation of self-employment on counselor attitudes

facil = 14.06 + 0.238 evalsum

SOURCE	DF	SS	MS	F	p
Regression	1	565.67	565.67	46.94	0.000
Error	74	891.74	12.05
Total	75	1457.41
R-sq = 38.8%

Discussion

The frequency with which self-employment is used as a rehabilitation strategy is affected by counselor attitudes toward such strategies. Attitudes accounted for more than 10% of the variance in the number of self-employment closures counselors reported during the previous year. These results suggest that the rates of self-employment in the VR system could become equitable with rates of self-employment in the general population by interventions that make self-employment a more attractive closure to the VR counselor.

Three variables were predicted to be instrumental in the development of counselor attitudes toward self-employment vocational rehabilitation strategies. Additionally, four variables were predicted to show effects of the counselors' attitudes. The results of this study suggest plausible cause and effect relationships between VR counselors' attitudes toward self-employment as a rehabilitation strategy and the ease with which VR consumers are able to pursue self-employment. Additionally, these results suggest the state policies governing self-employment may have an indirect, but important impact on counselors' attitudes toward self-employment as a rehabilitation strategy.

The Self-Employment Evaluation Scale was developed to measure VR counselors' attitudes toward self-employment as a rehabilitation strategy. This instrument demonstrated good reliability and consistent covariation in predicted directions with other variables in this study. Taken together, these data suggest the SEES functioned as a reliable and valid indicator of counselor attitudes toward self-employment strategies.

In this study, an attempt was made to determine a few of the key components that lead to attitude development in VR counselors. A multiple regression of items believed to be instrumental in attitude development onto the SEES provides clues to important influences on VR counselor attitude development. First, the counselors' past experience with self-employment strategies is the strongest predictor of attitude, accounting for 38% of the variance in counselor attitudes. While this result is not surprising, it raises the question of whether VR counselors in most states receive adequate training and support to use self-employment strategies successfully. If not, it seems likely that early failure experiences with self-employment strategies would lead to negative attitudes and consequently, decreased use of self-employment strategies. Given adequate training and support, early experiences may be more positive, resulting in more positive attitudes and higher rates of self-employment closure.

The second plausible influence on attitude development is the atmosphere regarding self-employment strategies in the VR counselor's office. In most states the VR system is highly organized and hierarchical. For instance, individual plans are prepared by counselors in standard formats. Additionally, counselors' performances are evaluated by supervisors and may be publicly compared. Such procedures express organizational standards or expectations for performance and imply contingencies. All staff are aware of these standards and their standing. They also become aware of the strategies that are believed to contribute the most and the least to achieving those standards. As an example, in our early discussions, some counselors indicated the emphasis in their office was on placements and self-employment strategies were generally viewed as inefficient or costly. Arnold & Seekins (1994), however, present data that self-employment closure costs compare favorably to other closures.

The final influence on VR counselors' attitudes toward self-employment identified in this study may come indirectly from a state's policy regarding the use of self-employment. Policies governing the use of self-employment vary widely across states (Arnold & Seekins, 1994). The counselors in this study indicated remarkable variability in their ratings of how facilitative their own state's policy is for pursuing self-employment strategies. With over 30% of the variance in the counselors' ratings of office atmosphere being accounted for by ratings of their states' policies governing self-employment, these policies probably have a substantial impact on counselors' use of self-employment strategies. The results of this present study suggest that the model adopted by the state VR administration will impact the level of optimism in the local VR office about using self-employment strategies. Unfortunately, this impact may be translated into decreased availability and accessibility of self-employment strategies to VR consumers.

In most states, the methods used to reach closure during the VR process are determined in large part by the VR counselor. If counselors approach the rehabilitation process with tendencies not to use self-employment strategies, the consumer wanting to pursue self-employment will likely find the process difficult and discouraging. The results of this study clearly suggest that the ease with which a consumer can pursue self-employment (as rated by the counselor) is related to counselors' attitudes.

The four variables predicted a priori to show effects from counselors' attitudes were first tested individually. These data indicate that counselors' attitudes affected their ratings of the rehabilitation process when they used self-employment strategies. Thus, the counselors' rating of the availability of self-employment as a rehabilitation strategy was related to their attitudes as was their ratings of the need to be protective of the state's resources, the difficulty of the process people must go through to attempt self-employment, and the counselor's own personal reaction to a consumer's request to pursue self-employment. These four variables may represent how facilitative counselors are to the process of pursuing self-employment. If so, then nearly 40% of the variance in how much consumers pursuing a self-employment closure are facilitated in the process can be predicted by the counselors' attitudes toward self-employment strategies. Because counselors' attitudes appear to be closely related to the broader VR context (e.g. office atmospheres, state policies), changes in the system may be most useful for making self-employment a more equitable choice for VR consumers.

Residents of many rural areas face difficult economic circumstances. While various economic development strategies are debated (Miller, 1985), entrepreneurial approaches, including self-employment, are advocated as an option of particular utility in rural areas (Malecki, 1988; Miller, 1985). This is consistent with trends in the general labor market.

Rural VR counselors might contribute to the economic development of the rural communities they serve by assisting consumers interested in starting their own businesses. Rural consumers might also benefit indirectly if counselors also participated in the generic activities of rural economic development to help create jobs. This latter step would be a new role for VR in rural communities.

The role of self-employment rehabilitation strategies in vocational rehabilitation may be dependent on important characteristics such as population density in the VR region. Thus, the use of self-employment strategies is probably more important in rural areas, where competitive employment options may be limited, than in more urban ones. While it is recognized that self-employment strategies may not be appropriate for many VR consumers, self-employment should certainly be as available as any other VR strategy. With current self-employment closure rates for states between .5% and 7%, it is clear that self-employment is not as available for many consumers as it could be. These research results suggest changes in the VR service system and pose the question of what role VR counselors might legitimately play in the economic development of the rural areas they serve. As part of our research, we have identified a number of creative approaches to using self-employment as a vocational rehabilitation closure. These have ranged from simple referrals to ad hoc arrangements with local community/economic development programs that approximate a program of supported self-employment.

This study was exploratory in nature and the results presented above should be investigated further. The cross-sectional and self-report nature of the data in this study can only suggest a causal link between state policies, counselor attitudes, and the use of self-employment strategies. Firm conclusions await more rigorous field studies that may compare policies across a smaller number of states and the relationship of these policies to self-employment strategy use. Additionally, as states develop more facilitative policies in the future, longitudinal studies of the effects could easily be employed. Such studies would be valuable to administrators in other states contemplating policy changes.

Consumer self-determination and freedom of choice is valued by most VR counselors. However, VR counselors operate within the broader context of the VR system. This study provides evidence that the broader VR system might change its policies to facilitate access to self-employment options, especially in rural areas.

References

Antonovsky, A. (1987). Unraveling the mystery of health. London: Jossey-Bass.

Arnold, N., & Seekins, T. (1994). Self-employment as a vocational rehabilitation closure: An examination of state policies. Journal of Disability Policy Studies, 5(2).

Becker, E. H. (1984). Self-employed workers: An update to 1983. Monthly Labor Review, July, 14-18.

Belous, R.S. (1989). The contingent economy: The growth of the temporary, part-time, and subcontracted work force. Washington, DC: National Planning Association.

General Accounting 0ffice (1993). Vocational Rehabilitation: Evidence for Federal Programs' effectiveness is mixed.

Gaithersburg, MD: U. S. General Accounting Office.

Griffin, R.W. & Moorhead, G. (1986). Organizational behavior. Boston: Houghton Mifflin Co.

Malecki, E.J. (1988). New firm startups: Key to rural growth. Rural Development Perspectives, 4 (2), 18-23.

Miller, J.P. (1985). Rethinking small business as the best way to create rural jobs. Rural Development Perspectives, 1 (2), 9-12.

Nunally, J.C. Psychometric Theory. New York: McGraw-Hill.

Rural Institute on Disabilities (1993). How do people make their living in rural areas? Missoula: The University of Montana.

Rehabilitation Act Amendments of 1992, P.L. 102-569.

Rehabilitation Services Administration (1988). Case service Reports. Washington, DC: Department of Education, Office of Special Education and Rehabilitation Services.

Schein, E.H. (1972). Management development as a process of influence. In Turner, J.H. & Filey, A.C., & House, R.J. (Eds.) Studies in organizational behavior (419-429). Glenview, IL: Scott Foresman and Co.

Seekins, T. (1992). Rural and urban employment patterns: Self-employment as a metaphor for rural vocational rehabilitation. Missoula, MT: Rural Institute on Disabilities, The University of Montana.

Silvestri, G. T. (1991). Who are the self-employed? Employment profiles and recent trends. Occupational Outlook Quarterly, Spring, 26-36.

---

Acknowledgements

This work was supported by grant #G0087C0228. This study is part of a larger effort supported by NIDRR to develop effective strategies for providing rehabilitation services in rural areas. The opinions expressed in this document are those of the authors and do not necessarily represent the policy of the Department of Education, and you should not assume endorsement by the Federal government. The authors wish to acknowledge the work of their colleagues Nancy Arnold, Peg Plimpton, and Cheryl Vandenberg on this project.

Questions? Would you like to receive periodic updates about our research and training activities? Do you have comments or suggestions about this site? E-mail your requests, comments and suggestions to dspas@ruralinstitute.umt.edu or call 888-268-2743 and ask for the Information Coordinator.