Race as a Correlate of Vocational Rehabilitation Acceptance: Revisited.

The new millennium seems certain to bring profound changes in the physical appearance of the typical United States (U.S.) citizen (Rehabilitation Services Administration, 1993; U.S. Bureau of the Census, 1996). By the year 2010, White Americans will find themselves in the minority in several U.S. cities (U. S. Department of Labor, 1987). Leong (1991) reported that the United States will comprise more racial minorities and women in the workplace, yielding a more diverse workplace. As part of this overall forecast, more minority customers are likely to seek vocational rehabilitation (VR) services, a troubling trend since race appears to influence the overall disability rate between minorities and non-minorities (Allen, 1976; Bowe, 1984; Brown, 1993; Ficke, 1992; Hayes-Bautista, 1992; U.S. Department of Education NIDRR, 1992; Walker, Adbury, Maholmes, & Rackley, 1992). African Americans, for example, are more likely to present with both disabilities and significant disabilities than White Americans (Allen, 1976; Marshall, 1987; Walker, 1988). Since this demographic transformation is already underway in the United States, one can predict that VR counselors will see more racial minorities than ever, making the issue of racial and ethnic eligibility a critical concern within the VR system.

According to Van Der Veer, Van Ijzendoor, and Valsiner (1994), "one of the most well-known rules of scientific methodology states that researchers should endeavor after repeatable or replicable experiments and results" (p. 11). Replication is often called an interesting paradox because researchers usually recite the need for replication (Wheaton, 1995; Wilson, 1999, 2000), however, studies are seldom replicated (Hacking, 1983). Citing Hacking's ironic observation, Miedema and Biesta (1994) concluded that a significant amount of lip service is paid to replication, but only a limited amount of theoretical reflection appears on the topic in the standard texts and reference books. The authors adduced that the scarcity of study replications mentioned by Miedema and Biesta, can also be observed, or lack of, in professional journals in rehabilitation education and counseling.

Principally, because so many studies of VR outcomes among African Americans and White Americans recommend replication in other states using Rehabilitation Services Administration (RSA)-911 data (Dziekan & Okocha, 1993; Wheaton, 1995; Wilson, 1999, 2000), the investigation at hand set out to replicate the Wilson's (1999) investigation on VR acceptance among African Americans and White Americans with disabilities. Finally, of the eight studies that have examined race and VR acceptance since 1980, none has constituted a replication of a previous study of acceptance by race.

Race and VR Acceptance

The first study to investigate discrepancies between African Americans and White Americans was conducted by Atkins and Wright in 1980. Atkins and Wright found African Americans were accepted less for VR services than White Americans in most of the Rehabilitation Services Administration (RSA) regions in the United States. However, after recalculating the data from the study of Atkins and Wright, Bolton and Cooper (1980) questioned whether a 5.5% discrepancy was sufficient to support the conclusion that African Americans and White Americans received unequal treatment in the VR system. Nevertheless, the study of Atkins and Wright retained its historical authority because it was the first investigation of race and VR acceptance. Subsequently, several research teams set out to confirm or refute the assertions adduced by Atkins and Wright. Interestingly enough, Atkins and Wright failed to apply a statistical test to race and VR eligibility to see whether the proportions observed between African Americans and White Americans were significant.

Over a decade later, Herbert and Martinez (1992) sought to determine whether race/ethnicity (American Indian/Alaskan Native, Asian/Pacific Islander, Black, or White) correlated with case service statuses. These statuses were 08 (closed not accepted for VR services), 26 (rehabilitated), 28 (closed other reason after the Individual Plan for Employment [IPE]), and 30 (closed other reasons before the IPE). Herbert and Martinez reported that White Americans had a significantly higher acceptance rate than African Americans. Moreover, African Americans and Hispanics were more likely to be found ineligible for VR services than other minority groups. Notwithstanding the contrast in samples used by Atkins and Wright (1980) and Herbert and Martinez, the results reported by the latter authors reinforced those reported earlier by Atkins and Wright (i.e., African Americans tended to be accepted less for VR services than White Americans).

Consequent and subsequent to the study of Herbert and Martinez (1992), Dziekan and Okocha (1993) explored the accessibility of VR services between minorities (African Americans, Hispanics, Native Americans, and Asian Americans) and the majority (White Americans). The results reported by Dziekan and Okocha paralleled those reported by Atkins and Wright (1980) and Herbert and Martinez (1992), in that African Americans were less likely to be found eligible for VR services than White Americans. Feist-Price (1995) reported the same results two years later. Accordingly, the results revealed by the first four investigators suggested a discrepancy between VR acceptance and race.

Contrary to the four prior studies on VR acceptance and race, Wheaton (1995) concluded that "the proportions of European Americans [White Americans] and African Americans found eligible for VR services are not significantly different statistically" (p. 228). In contrast to past investigations, Wheaton employed a different hypothesis (symmetrical) and sampling procedure (homogeneity of proportions) in his study. Similarly, Peterson (1996) found VR acceptance among African Americans, White Americans, American Indians, Eskimos or Aleuts, Asians or Pacific Islanders, Hispanics, and "others" to be independent. Subsequent to the studies of Wheaton and Peterson, Wilson (1999) found no differences between African Americans and White Americans in VR acceptance. Thus, while the later findings of investigators (Peterson, Wheaton, and Wilson) challenged the earlier findings of Akins and Wright, these later findings also confirmed a need to continue and extend eligibility research, particularly in other states. For example, the relationship between VR acceptance and variables such as racial membership, socioeconomic status, and vocation remains unclear. Thus, as both Wheaton and Wilson recommended, a need for replicating VR studies that examine acceptance rates seems not only compelling, but also necessary.

Finally, addressing some of the limitations in earlier VR investigations, Wilson (2000) used a multivariate analysis to investigate VR acceptance based on race, education, work status, and source of support at application. Primary source of support and race were the only two variables statistically significant in the regression model. While Wilson employed a different test statistic and methodology than other researchers (for example, Atkins & Wright, 1980; Dziekan & Okocha, 1993; Wheaton, 1995) who investigated VR acceptance, his results were congruent with longstanding observations--namely, African Americans tended to be accepted less for VR services than White American with disabilities (Atkins & Wright, 1980; Feist-Price, 1995; Herbert & Martinez, 1992). Although the reasons underlying VR eligibility decisions continue to attract interest, evidence suggests a continuing association between VR acceptance and race. This study was founded on the following research question:

Research Question

Is there a difference in the acceptance rates between African Americans and White Americans in the state of Michigan?

Method

Data Collection

Data from this study came from the national RSA-911 database in Washington, DC. We randomly selected Michigan from the list of midwestern states to replicate the investigation by Wilson (1999). Note: Both VR and the Bureau of Visual Service agencies were included in the sample. In addition, the coding procedures for RSA-911 data conformed to federal guidelines established by RSA in 1995 (RSA, 1995). To decrease the possibility of coding error, we generated and examined descriptive statistics for outliers and suspicious patterns. Not one was observed.

Variables

Racial status (the independent variable). Racial status served as the independent variable with two levels (African American or White American). Race is defined as the race reported by customers on their application for VR services (RSA, 1995). In this investigation, "Blacks" are referred to as African Americans and "Whites" are referred to as White Americans. Because the study at hand was a replication of the Wilson's (1999) study, we excluded Asian Americans and Native Americans from the investigation as well.

Acceptance for VR services (the dependent variable). Because all the closure statuses from the national RSA-911 data were labeled 1 through 6, it became necessary to group the categories by the 1995 RSA definitions of acceptance and none acceptance for VR services. Thus, the criterion variable of VR acceptance included two levels: Statuses 08 from 02, statuses 08 from 06, and status 38 from 04 were coded as 1, not accepted for VR services. Statuses 28 and 30 were coded as 2, accepted for VR Services.

Education at referral (the control variable). Education at referral became a controlled variable with three levels: less than a high school degree (LT HS), high school degree (HS) degree, and greater than a high school degree (GT HS). The GT HS category included attendance and training at vocational and trade schools, colleges, and universities. To rule out education as an extraneous variable, only customers with a high school education were included in this investigation because more customers indicated that they completed HS than any other educational level in Michigan and in the national population of VR customers. Thus, the findings can be generalized to a large population of Michigan customers. Finally, because Wilson's (1999) investigation did not control for the education variable, building the education variable into the study seemed like a logical step. This was the only modification in the replication.

Participants

The sample included 17,961 customers who sought VR services from Michigan from October 1, 1997, to September 30, 1998. We then identified persons with no missing data on the independent and dependent variables of race and VR acceptance, respectively. The final subsample included African Americans (n=1,453) and White Americans (n=3,122) from the population of customers with no missing values (25% of population) on the variables under investigation. The only variable for which we controlled was education.

Procedure

The personal computer version of the Statistical Package for the Social Sciences (SPSS) was used to calculate both the descriptive and inferential statistics in the investigation.

Analysis. The chi-square test of independence was used to analyze the dichotomous data in the investigation. As Fraenkel and Wallen (1993) observed, "the chi-square test is based on a comparison between expected frequencies and actual, obtained frequencies. If the obtained frequencies are similar to the expected frequencies, then researchers conclude that the groups do not differ" (p. 201).

To examine the association between the independent variable (race) and dependent variables, acceptance or non-acceptance (Status 02 from 08, 08 from 06, and 38 from 04), we used the phi coefficient, which is the appropriate measure when independent and dependent variables are categorical and dichotomous (2 by 2), as in the investigation at hand (SPSS, 1997).

Results

Results of the chi-square test. Race and VR acceptance emerged as dependent (statistically significant) of each other: [chi square] (1, n=4,575)=9.811; p [is less than] .002). Additionally, a significant association emerged between race and VR acceptance (phi coefficient =.046), which when squared, denotes that race accounted for approximately .002% of the variance in VR acceptance (see Table 1).

Table 1
Vocation Rehabilitation Acceptance Decision by Race In Michigan

                        Acceptance Status

                    Not Accepted     Accepted        Total

Racial Group          n      %       n      %       n      %

African Americans     410   28.2   1,043   76.1   1,453   100
White Americans       746   23.9   2,376   71.8   3,122   100
Total               1,156   25.3   3,419   74.7   4,575   100

Note: [chi square] (1, n=4,575) =9.811; p < .002; phi
coefficient =.046; phi squared = .002.

Discussion

The results of this investigation revealed that in the state of Michigan, African Americans with a high school education are less likely to be accepted for VR services than White Americans with a high school education. Although past research teams that investigated race and VR acceptance did not control for the education variable (the exception being Wilson, 2000), these findings concur with the results presented in past studies of race and VR acceptance (e.g., Dziekan & Okocha, 1993; Feist-Price, 1995; Herbert & Martinez, 1992; Wilson, 2000). Also, these results challenged findings from other researchers who reported no statistical difference between race and VR acceptance (Peterson, 1996; Wheaton, 1995; Wilson, 1999). A total of nine studies have examined race and VR acceptance since 1980, including this study. Moreover, of these nine studies, five found race and VR acceptance dependent (statistical significance). While the remaining three reported race and VR acceptance as independent (not statistically significant). The first study to address this issue, Atkins and Wright (1980), which is the seminal investigation on race and VR eligibility, included no statistical test to verify the posited association between race and VR acceptance. Thus, the majority of studies that examined race and VR acceptance suggest that African Americans, and by inference, other minorities (e.g., see Herbert & Martinez, 1992), are less likely to be accepted than White Americans for VR services. Since the variance explained in the dependent variable is small, the results of this study should be interpreted with caution.

Because the chi-square test statistic is influenced by the sample size (SPSS, 1997), the sample size of this study might have increased the prospects of finding significance. Also, because a relatively large number of VR customers with no missing data on the variables under investigation were accepted for VR services, overfitting (sensitivity) might have occurred. This process is called the "additive effect" of the sample size (Wilson, 2000). The application guide of SPSS (1997) states that "the size of the chi-square depends not only on the differences between the observed and expected counts but also on the sample size. Try making a file with 100 or even 10 copies of the data and watching how the chi-square statistic increases with the sample size" (p. 71). Meaning, as the sample size increases, the probability of finding statistical significance will also increase (Grimm & Yarnold, 1995). Given the advantages and disadvantages of the sample size, the size of the sample is considered an advantage for this study. Thus, the results of this study seem to represent most customers (e.g., African Americans and White Americans with a high school education) in Michigan. It is, however, conceivable that drawing a proportional sample from the population of VR customers might have yielded different results.

Watson and Collins (1993) suggested that the lack of cultural competence is a primary reason why racially and ethnically diverse populations have different experiences compared to their White American counterparts in the VR system. It is obvious that "professional multicultural rehabilitation competencies and standards are necessary if persons with disabilities from diverse ethnic backgrounds are to be well served by rehabilitation counselors" (Middleton, et al., 2000, p. 220). Because persons from diverse backgrounds are increasing in numbers, VR counselors cannot afford to neglect an ethical responsibility to serve all persons with disabilities in a like manner, regardless of race, gender, or sexual orientation. As Mahalik, Worthington, and Crump (1999) recently reported, not only do non-minority counselors have worldviews that contrast with those of the minority clients they serve, but also minority counselors tend to have worldviews that contrast with those of the minority clients. Thus, we recommend more training, especially educational training (e.g., workshop & seminars), for all VR counselors, African American and White American, so that they all become more competent in serving diverse populations.

Because contemporary research suggests that the VR system is a microcosm of the larger society (Rubin, Pusch, Fogarty, & McGinn, 1995; Sue, 1994; Wilson et. al., 1999), it is possible that the negative perceptions of VR counselors unfairly judge African Americans seeking VR services (Rosenthal & Berven, 1999). To assist people with disabilities through the VR system successfully, and to assist minorities specifically, VR counselors must first acknowledge any stereotypes they embrace. Confronting one's stereotypes and biases is essential for VR counselors with preexisting views regarding minorities (Rosenthal & Berven). More importantly, not only are some of these stereotypes negative, but also they are resistant to change even in the face of contrary information about a particular stereotype or group (Rosenthal & Berven). Thus, it was reported by Rosenthal and Berven that VR counselors (VR masters students at the time of the study) seem to view African Americans as having less potential than White Americans. Based on the findings of Rosenthal and Berven, it is hypothesized that some White Americans VR counselors find some African Americans ineligible for VR services, having unfairly based their decisions on stereotypes and biases. According to Bennett (1995), "the fact that cultural differences are often associated with racial differences confirms myths and stereotypes associated with race" (p., 44). Augoustinos and Walker (1995) supported this social cognition hypothesis by Bennett. One should note that other variables besides race explain VR acceptance decisions. Race is, of course, only one factor in a myriad of factors essential to examining possible intergroup tensions. In a national survey on intergroup relations, Smith (2000) observed that:

   Intergroup relations go well beyond race relations. While race relations
   are an important component of intergroup relations, they are merely part of
   the larger whole. Moreover, race relations do not always represent the
   greatest problems in intergroup relations. On some dimensions,
   non-racial/ethnic groups such as Muslims, gays and lesbians, and the
   disabled are judged to be in more extreme circumstance. (p. 9)

Although cultural mistrust can be a constructive reaction in certain situations, some African Americans may approach the system for services thinking that VR counselors are not going to be fair or objective. Alston and Bell (1996) reported that:

   Cultural mistrust is one of several characteristics of African Americans
   with disabilities that may influence the manner in which they approach the
   rehabilitation system and interact with its professionals. It is important
   to note that cultural mistrust is not intrinsically negative. Ideally, the
   likelihood of rehabilitation entry and success for African American clients
   will be greatly enhanced by professional awareness concerning cultural
   mistrust and knowledge of strategies to control its effects on the
   rehabilitation process (p. 19).

Cultural mistrust by African Americans may, of course, create a self-fulfilling prophecy (Thomas & Sillen, 1972) that leads to negative behaviors in the form of missed appointments and early termination. These behaviors may result in an unsuccessful closed status (for example, uncooperative or unable to locate) when one is closed unrehabilitated in the VR system. Just as White Americans may have negative perceptions of African Americans, African Americans may have negative perceptions of White American counselors who try to help them become eligible for VR services. Because some African American customers have encountered difficulties with White Americans (Smith, 2000), they may behave toward White American VR counselors in a manner consistent with their negative past experiences with White Americans. This may lead to another self-fulfilling prophecy. Nevertheless, one should endeavor, generally, not to blame the victim in instances of strained communications. Because clients and therapists hold different world views (Bennett, 1995; Mahalik, Worthington, & Crump, 1999), strained communication within the confines of the customer/counselor dyad can become an obvious concern. However, as Smith (2000) reports, being gay or lesbian, for example, may complicate intergroup relations as much as being African American. According to the research conducted by Smith, though, people in the United States believe that African Americans experience the most discrimination of racial and ethnic minority groups (racial/ethnic groups included in the Smith study: Asians, African Americans, Hispanics, Native Americans, & Whites). Although explanations for acceptance discrepancies remain imprecise (Wilson, 2000), VR agencies are challenged to continue the process of inclusiveness for all people with disabilities. Finally, based on the small association between the independent and dependent variables in the present study, it is highly unlikely that VR decisions are solely made based on one's race or ethnicity. More research is needed in this area.

Limitations of the Study

Because this study was ex post facto, obvious threats to internal validity may have altered the results. As Fraenkel and Wallen (1993) noted, "the major threat to the internal validity of a causal-comparative study is the possibility of a subject-characteristics threat" (p. 322). The customer's ages might be considered for inclusion in future studies. Two variables uncontrolled for in this investigation included age and customer income.

Beyond the general need for replication and statistical controlling for certain independent variables, this investigation also contains limitations related to external validity. As Campbell and Stanley (1963) stated, external validity "always turns out to involve extrapolation into a realm not presented in one's sample" (p. 17). Specifically, this study was limited to prospective VR customers who sought VR services, the state of Michigan, and African American and White Americans with a high school education who self-reported race on their applications for VR services. Strictly speaking, the results reported here apply only to the state and the sample from which the customers were drawn. More research is needed to generalize the findings of VR acceptance and race throughout the United States--a now familiar refrain. Because the amount of variance explained (.002) in the independent variable (race) by the dependent variable is small, it is unlikely that VR counselors in the state of Michigan are making eligibility decisions based solely on race.

Future Research

Because a preponderance of past research drew only upon state RSA-911 data, we recommend that future researchers seek national resources from which to draw samples in order to generalize the findings of VR acceptance and race to the entire United States. Although the only previous nationally documented study by Atkins and Wright (1980) did not indicate whether there was a significant difference between African Americans and White Americans in VR acceptance, evidence from that study suggested that African Americans were less likely to be eligible for VR services than their White American counterparts. Bolton and Cooper (1980) immediately challenged some of the conclusions of this pioneering study. Moreover, given that the data used in Atkins and Wright's study are now 21 years old, one can assume with less confidence that African Americans are still being found ineligible in higher proportions than White Americans in the United States. To illustrate this point, although most of the VR acceptance studies reported that African Americans were less likely to be found eligible than White Americans, data have not been gathered from 41 states to substantiate this view of VR acceptance in the U.S. In short, we do not know if the conclusion set forth by Atkins and Wright is universally true-- that "compared to Whites, Black VR applicants were [are] not only more likely to be screened out (found ineligible) but, if made eligible for services, Blacks were [are] less likely to be rehabilitated" (Atkins & Wright, 1980, p. 44). We need to determine whether the results presented by Atkins and Wright over 20 years ago still apply today throughout the United States. Perhaps, a national instead of a state study on race and VR acceptance will answer this most important question.

As several researchers have indicated (Wheaton, 1995; Wilson, 1999, 2000), another area of future research might focus on investigating VR acceptance and race in other states. While the external validity of past VR eligibility research is limited to the number of states where researchers have actually explored race and VR acceptance, research in other states might reveal a more definable trend. We contend that one can only generalize to the states from which population samples were drawn. Thus, the extant studies on race and VR acceptance have yet to indicate a generalizeable pattern. However, if the number of states providing study results total 30 or more, and if the majority concluded that race and VR acceptance are dependent, then one could adduce a more reliable pattern of race and VR acceptance for the entire country. It is apparent that to generalize an acceptance trend in the United States, researchers will have to use the national RSA-911 database (or other databases that yield like data), or produce more state investigations regarding VR acceptance and race. Until then, we cannot generally adduce the noted discrepancies in VR acceptance from this study.

Controlling for cultural competence, counselor ethnicity, and stereotypes are recommended for future researchers. Although stereotypes and counselor's ethnicity are found to influence the way one perceives and reacts towards other races (Alston & Bell, 1996; Association of Community Organizations for Reform Now, 1996; Ayers, 1969; Sue, 1994), we suggest including these variable in future investigations to weed out extraneous variance to the research design. Furthermore, including these variables in a multivariate equation will not only yield valuable information, but also give the amount of variance accounted for when all other variables are controlled in analysis. Finally, because Wilson (2000) found that primary source of support was correlated with VR acceptance, future researchers may want to include primary source of support in their multivariate analysis. As with many studies of race and VR acceptance, this investigation revealed more questions than answers.

Conclusion

Given that several investigators of VR eligibility have also recommended further replication work (see Wheaton, 1995; Wilson, 1999), study replication is clearly in order. However, replication is also a paradox because researchers continually repeat the need for study replication, nevertheless, studies are seldom replicated (Hacking, 1983). Thus, the purpose of this study was a to replicate the Wilson (1999) investigation and to examine whether African Americans and White Americans with disabilities differ in VR acceptance in yet another state. Results showed a statistically significant difference between race and VR acceptance: [chi square] (1, n=4,575) =9.811; p [is less than] .002). However, the small association between race and VR acceptance (phi coefficient=.046) indicates that less than 1/2 of 1 percent of the variance in acceptance is explained by race. Although this study was a replication, the results challenge earlier results reported by Wilson (1999), Wheaton (1995), and Peterson (1996), which revealed no differences in acceptance rates by race. Based on the results reported here, other researchers might well use different resources to see whether patterns in VR acceptance persist throughout the United States.

Acknowledgement

The authors would like to express a sincere thanks to the United States Department of Education, Rehabilitation Services Administration, for the use of National 911 Data.

References

Allen, K. G. (1976). First finding of the 1972 survey of the disabled: General characteristics. Social Security Bulletin, 39, 18-37.

Alston, R. J., & Bell, T. J. (1996). Cultural mistrust and the rehabilitation enigma for African Americans. The Journal of Rehabilitation, 62, 16-20.

Association of Community Organizations for Reform Now. (1996, July). How to keep your classroom White. Harper's, 293, 18-19.

Atkins, B. J., & Wright, G. N. (1980). Three views: Vocational rehabilitation of Blacks: The statement. Journal of Rehabilitation, 46, 40, 42-46.

Augoustinos, M., & Walker, I. (1995). Social cognition. An integrated introduction. London: Sage.

Ayers, E. G. (1969). White racist attitudes and rehabilitation. Rehabilitation Counseling Bulletin, 13(1), 53-60.

Bennett, C. I. (1995). Comprehensive multicultural education: Theory and practice (3rd ed.). Massachusetts: Allyn & Bacon.

Bolton, B., & Cooper, P. G. (1980). Three views: Vocational rehabilitation of Blacks: The comment. Journal of Rehabilitation, 46, 41, 41-49.

Bowe, F. (1984). U.S. census and disabled adults. Hot Springs: University of Arkansas, Arkansas Rehabilitation Services, Arkansas Rehabilitation Research and Training Center.

Brown, J. G. (1993). In T. Wright & P. Leung (Eds.), Substance abuse. Meeting the unique needs of minorities with disabilities. National council on disabilities: A report to the President and the Congress. Proceedings from the national council on disabilities (p. 41). Washington, DC: National Council on Disability.

Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin Company.

Chideya, F. (1995). Don't believe the hype: Fighting cultural misinformation about African Americans. New York: Penguin Group.

Dziekan, K. I., & Okocha, A. G. (1993). Accessibility of rehabilitation services: Comparison by racial-ethnic status. Rehabilitation Counseling Bulletin, 4, 183-189.

Feist-Price, S. (1995). African Americans with disabilities and equity in vocational rehabilitation services: One state's review. Rehabilitation Counseling Bulletin, 39, 119-129.

Ficke, R. C. (1992). Digest of data on persons with disability. Washington, DC: National Institute on Disability and Rehabilitation Research.

Fraenkel, J. R., & Wallen, N. E. (1993). How to design and evaluate research in education. New York: McGraw-Hill.

Grimm, L., & Yarnold, P. (1995). Reading and understanding multivariate statistics. Washington, DC: American Psychological Association.

Hacking, I. (1983). Representing and intervening. New York: Cambridge University Press.

Hayes-Bautista, D. E. (1992). Latino health indicators and the underclass model: From paradox to new policy models. In A. Furino (Ed.). Health policy and the Hispanic (pp. 32-47). Boulder, CO: Westview.

Herbert, J. T., & Martinez, M., Y. (1992). Client ethnicity and vocational rehabilitation case service outcome. Journal of Job Placement, 8, 10-16.

Leong, F. T. L. (1991). Guest editor's introduction. Career development of racial and ethnic minorities [special issue]. Career Development Quarterly, 39, 196-198.

Mahalik, J., Worthington, R., & Crump, S. (1999). Influence of racial/ethnic membership and therapist culture on therapists worldview. Journal of Multicultural Counseling and Development, 27, 2-17.

Marshall, M. (1987, October). Fighting for their rights. Ebony, 62(5), 68-70.

Middleton, R. A., Rollins, C. W., Sanderson. P. L., Leung, P. Harley, D. A., Ebener, D., & Idrogo, A. L. (2000). Endorsement of professional multicultural rehabilitation competencies and standards: A call to action. Rehabilitation Counseling Bulletin, 43, 219-240.

Miedema, S., & Biesta, G, J. (1994). Constructivism and pragmatism: How to solve the problematic relation between methodology and epistemology in the debate about replication. In Van Der Veer, R, Van Ijzendoorn, M., & Valsiner, J. (Eds.), Reconstructing the mind: Replicability in research on human development (pp. 71-92). New Jersey: Ablex Publishing Corporation.

Peterson, G. E. (1996). An analysis of participation, progress, and outcome of individuals from diverse racial and ethnic backgrounds in the public vocational rehabilitation program in Nevada. Unpublished doctoral dissertation, University of Northern Colorado.

Rehabilitation Services Administration (RSA). (1993). Rehabilitation Act of 1973 as amended by the Rehabilitation Act of 1992. Washington, DC: U. S. Department of Education.

Rehabilitation Services Administration. (1995). Reporting manual for the case service report (RSA-911) (RSA-PD-95-04). Washington, DC: Rehabilitation Services Administration.

Rosenthal, D.A., & Berven, N.L. (1999). Effects of client race on clinical judgement. Rehabilitation Counseling Bulletin, 42, 243-264.

Rubin, S.E., Pusch, B.D., Fogarty, C., & McGinn, F. (1995). Enhancing the cultural sensitivity of rehabilitation counselors. Rehabilitation Education, 9, 253-264.

SPSS. (1997). SPSS base 7.5 applications guide. Chicago: SPSS.

Smith, T W. (2000). Taking american's pulse II: A survey of intergroup relations. National Opinion Research Center. NORC, University of Chicago. The National Conference for Community and Justice.

Sue, D.W. (1994). The challenge of cultural diversity: Overcoming barriers. In P. Leung & R.A. Middleton (Eds.), National Association of Multicultural Concerns, 20-27.

Thomas, A., & Sillen, S. (1972). Racism and psychiatry. New York: Carol Publishing Group.

U.S. Bureau of Census (1996, February). Population projections of the United States by age. sex, race, and Hispanic origin; 1995 to 2050 (P25-1130. pp. 8-17). Retrieved April 7, 1998, from the World Wide Web: http://www.census.gov/prod/1/ popp25-1130/p251130b.pdf.

U.S. Department of Education National Institute on Disability and Rehabilitation Research (NIDRR). (1992). Disability statistics abstract number 4. Washington, DC: Author.

U.S Department of Labor. (1987). Projections 2000. Occupational Outlook Quarterly. 3(3), 11-12.

Van Der Veer, R, Van Ijzendoorn, M., & Valsiner, J. (Eds.). (1994). Reconstructing the mind: Replicability in research on human development. New Jersey: Ablex Publishing Corporation.

Walker, S. (1988). Toward economic opportunity and independence: A goal for minority persons with disabilities. In S. Walker et al. (Eds.), Building bridges to independence. Proceedings of national conference on employment successes, problems and needs of Black Americans with disabilities.

Walker, S., Adbury, C., Maholmes, V., & Rackley, C. (1992). Prevalence, distribution and impact of disability among ethnic minorities. In future frontiers in the employment of minority persons with disabilities (pp. 10-24). Washington, DC: Howard University Research and Training Center.

Watson, A. L., & Collins, R. (1993). Culturally sensitive training for professionals. Journal of Vocational Rehabilitation, 3, 38-45.

Wheaton, J. E. (1995). Vocational rehabilitation acceptance rate for European Americans and African Americans: Another look. Rehabilitation Counseling Bulletin, 38, 224-231.

Wilson, K. B. (1999). Vocational rehabilitation acceptance: A tale of two races in a large midwestern state. Journal of Applied Rehabilitation Counseling 30(2), 25-31.

Wilson, K. B. (2000). Predicting vocational rehabilitation eligibility based on race, education, work status, and source of support at application. Rehabilitation Counseling Bulletin, 43, 97-105.

Wilson, K. B., Jackson, R., & Doughty, J. (1999). What a difference a race makes: Reasons for unsuccessful closures within the vocational rehabilitation system. American Rehabilitation, 25, 16-24.

Keith B. Wilson
The Pennsylvania State University

Debra A. Harley
University of Kentucky

Reginald J. Alston
University of Illinois-Urbana Champaign

Keith B. Wilson, Ph.D., CRC, Department of Counselor Education, Counseling Psychology, and Rehabilitation Services, 308 CEDAR Building, University Park, PA 16802-3110. Email: [e-mail]