Hearing loss

Many persons with disabilities are excluded from full participation in the nation's labor force (Louis Harris & Associates, 1998). This is a major reason for the existence of the federal-state vocational rehabilitation (VR) program. The primary goal of this program, under the direction of the Rehabilitation Services Administration (RSA), is to assist those with disabilities obtain, or retain, employment. For over 75 years, VR has improved the lives of persons with disabilities by providing an individualized program of services leading to employment outcomes that result in notable gains in their economic status (Department of Education, 1999). Despite the value of this program, many persons with disabilities experience difficulties in obtaining jobs even after the provision of rehabilitation services. For example, of the consumers closed during fiscal year 1997 who received VR services, 61% obtained some form of employment, while the remaining 39% did not (Capella, 2001). Several authors have evaluated the efficacy of the VR program, either overall or by individual state program, primarily targeting employment as the key outcome variable (e.g., Butterworth, Schalock, & Gilmore, 1998; Gibbs, 1990; Platt, 1995; Walls, Misra, & Majumder, 2002). Although not evaluated as frequently, another relevant question is whether the VR services received and related outcomes differ by type of disability.

In particular, this is a relevant question for individuals with hearing loss (Note: for brevity, this term will be used to represent all persons who are deaf, late-deafened, hard of hearing, or deaf-blind), as success in obtaining employment through VR services by this group has decreased substantially (Rehabilitation Services Administration [RSA], 2000). Even though the prevalence of hearing loss has increased (Ries, 1994; RSA, 2000), RSA recently noted a 10-year decline in the total number of successfully rehabilitated VR consumers with hearing loss (RSA, 2000). The percentage of consumers with hearing loss, of the total number of consumers successfully rehabilitated each year, has also declined during this period. This trend prompted RSA to call for state agencies to examine their policies and procedures related to caseloads of consumers with hearing loss. Research is necessary to investigate this issue. The University of Arkansas Rehabilitation Research and Training Center for Persons who are Deaf or Hard of Hearing has been mandated to conduct research in this area during its current 5-year funding period. This study represents the first piece of research in a series of projects.

Investigating outcomes for persons with hearing loss is also important when one considers the large number of individuals in this group. According to a recent Vital and Health Statistics report, published by the National Center for Health Statistics, 8.3% of the population in the United States has a hearing loss (Adams, Hendershot, & Marano, 1999). Approximately 11.5 million persons between the ages of 18 and 64 have some level of hearing loss (Adams et al., 1999). Given that they are of working age, these people would potentially be candidates for VR services. Despite the high incidence of hearing loss in the population, the percentage of persons with hearing loss served by VR is relatively low in terms of the total population of all those who receive services.

The low levels of participation and decrease in successful outcomes for consumers with hearing loss could be related to a variety of factors. One hypothesis is that the VR system may not be meeting the needs of this group of consumers. This failure could result in fewer successful closures, as well as fewer persons from this group seeking VR services. A section of the 19th Institute on Rehabilitation Issues document, "Serving the Underserved--Principles, Practices, and Techniques" focused on persons who are hard of hearing (Corthell & Yarman, 1992). This document reported that this population frequently does not receive appropriate services, as VR counselors are often inexperienced and not aware of the multiple services persons who are hard of hearing may need. One avenue to begin assessing the hypothesis that the VR system is not meeting the needs of the entire group of consumers with hearing loss is to make comparisons between them and VR consumers with other disabilities. This type of assessment will provide information that may lay the groundwork for additional research. It is also a good starting point because data is currently available from RSA to make the comparisons of interest.

Other authors previously examined the employment outcomes of persons with hearing loss served by VR through analysis of extant case service data on the state or national level (El-Khiami, 1986; Moore, 2001a, 2001b; Moore & Schroedel, 2001a, 2001b). Overall, VR services lead to positive employment outcomes for significant numbers of this population. For example, Walls and Fullmer (1997) compared the types of employment obtained by consumers with different disabilities. Their results indicated that persons with hearing loss obtained jobs that were comparable to other disability groups. Moore (2001a) examined RSA-911 data for fiscal year 1996 to determine, among other objectives, whether certain VR services were significantly associated with successful closures for persons with hearing loss. All four services he evaluated (assessment, restoration, counseling and guidance, and job placement) were associated with successful closure. Moore (2001b) also examined fiscal year 1997 RSA-911 data to evaluate differences in services received and employment outcomes for successfully closed consumers in three categories: deaf, hard-of-hearing, and late deafened. The relationship between VR services and work status at closure was also investigated, with the finding that college or university training, business or vocational training, on-the-job training, and job placement were predictive of competitive employment at closure for this population.

Moore and Schroedel (2001a) recently took another step in evaluating VR service delivery to individuals with hearing loss, by providing descriptive information on the demographic characteristics, VR services, and case outcomes for subgroups of this population, as well as providing comparisons with the general population of VR consumers. I took the next step in evaluating VR services and outcomes for consumers with hearing loss, by using statistical analyses to compare their experiences and outcomes to those of the general population of persons served by VR (i.e., persons from other disability groups).

The purpose of this project was to evaluate VR service delivery for consumers with hearing loss. Acceptance rates, characteristics of persons served, services provided, and resulting employment outcomes were examined through analysis of national case services data. This study compared participation and outcomes of the target population with the general population of rehabilitation consumers with other types of disabilities. The primary research questions investigated were:

1. How do acceptance rates of consumers with hearing loss compare to consumers with other disabilities?

2. How do demographic characteristics of consumers with hearing loss compare to consumers with other disabilities?

3. How do specific services received by consumers with hearing loss compare to consumers with other disabilities?

4. How do costs of services received by consumers with hearing loss compare to consumers with other disabilities?

5. How do employment outcomes of consumers with hearing loss compare to consumers with other disabilities?

The demographic variables evaluated in this study were (a) sex, (b) age, (c) education level, (d) race, (e) severity of disability, and (f) work status at application. All 17 services listed in the RSA-911 data were included in the evaluation of Research Question 3, as well as total number of services received. Employment outcomes were evaluated in terms of (a) whether the consumers obtained employment or not, and, if closed successfully, (b) type of employment obtained (e.g., competitive employment, self-employment, homemaker), and (c) whether medical insurance was available through a job at closure.

Method

Data for the study were obtained from the national RSA-911 database for fiscal year (FY) 1997, which is maintained by the Rehabilitation Services Administration (RSA). This database includes information about each person whose case was closed during that year. In addition to primary and secondary disability type, RSA-911 variables from three major categories were extracted from the database: consumer characteristics, services obtained, and employment outcomes. The variables included in the study were selected because they were considered the most relevant to the research in terms of identifying discrepancies in consumer characteristics, services, or outcomes for consumers with hearing loss versus other consumers.

For the analysis of differences in acceptance rates, the entire population of VR consumers in the FY 1997 data was used (N = 586,021). For the majority of the remaining analyses, the consumer population was those who had been closed successfully or unsuccessfully after receiving VR services (i.e., Status 26 or Status 28 closures, respectively; N = 347,731). The number of consumers with hearing loss (as either their primary or secondary disability) was determined by creating a subset based on disability codes (i.e., codes between 231 and 299; n = 23,992). In order to obtain equal sized groups, a random sample of 23,992 consumers was drawn from the population of consumers with other types of disabilities closed in Status 26 or 28. These two samples were combined, for a total sample size of 47,984 consumers. This sample was used to address most of the analyses associated with Research Questions 2 through 5. For a few analyses, a subset of this sample was used. Differences in terms of the first two measures of employment outcomes were assessed for both the complete sample of 47,984 persons and a reduced sample consisting of only persons who were not competitively employed at application (n = 35,385). For the third measure of employment outcomes, availability of medical insurance, another reduced sample was used. Persons from the full sample who were closed successfully and did not have medical insurance available through an employer at application (n = 26,760) were included in this analysis.

Information on demographic characteristics for each group is provided in the analyses for Question 2; therefore, it is not repeated here. However, the types of disabilities reported for persons in the comparison group are not included. Because this information may be of interest to the reader, type of disability for consumers in the other disability group (n = 23,992) is provided in Table 1.

Due to the large sample sizes used for these analyses, most significance test results were expected to be statistically significant. Therefore, in order to deem a finding important or meaningful, practical significance of the differences between the groups also had to be established. Practical significance was evaluated with a measure of association or an effect size for each statistical significance test.

Analyses were conducted with SAS[R] Version 8.2. Chi-square was used to evaluate research questions involving categorical variables. Measures of association used in evaluating practical significance for these analyses were phi coefficients and/or odds ratios. A phi coefficient is a measure of association derived from the Pearson chi-square statistic, with values ranging from -1 to 1 for 2 X 2 tables. However, because the variables of interest here are nominal, the direction of the association is irrelevant and only absolute values will be reported. Odds ratios are another measure of association, available for 2 X 2 tables only. When the variables being tested are independent of each other, the odds ratio will equal one. Differences between groups based on the continuous variables (i.e., age, education level, number of services, and cost of case services) were evaluated with the analysis of variance procedure. Cohen's d was the effect size used to measure practical significance of these findings.

Results

Acceptance Rates and Demographic Characteristics

Because of the large number of significance tests conducted, the statistical results will not be listed here. The reader is instead referred to the table associated with the research question. Effect sizes or measures of association and descriptive statistics for each group are provided in the tables. Question 1 investigated whether differences existed between the groups in terms of acceptance rates. A significantly higher percentage of consumers with hearing loss were accepted for services.

Question 2 involved evaluating differences between groups on several demographic characteristics. Significant differences were found in terms of sex, with a higher percentage of females served in the hearing loss group than in the other disability group. Consumers in the hearing loss group were significantly older, on average, than consumers in the other disability group. Significant differences were found between the groups based on race, with a substantially higher percentage of Whites and lower percentage of African-Americans served in the hearing loss group. There were also significant differences between the groups in terms of work status at application. The most relevant difference was that consumers with hearing loss were more likely to be competitively employed at application than those with other disabilities. Although there were statistically significant differences between the groups on severity of disability, the difference in percentages was small. There were not significant differences between the groups in terms of education level, with the average educational level of the groups virtually identical. Detailed information (including group percentages) for the categorical variable analyses presented above is provided in Table 2. Table 3 provides information for the analyses associated with the continuous variables.

Case Service Variables

Question 3 addressed VR services received. The total number of services received differed between the groups, as consumers with hearing loss obtained significantly more services. No practically significant differences between groups were found on receipt of the following services: (a) Job finding services, (b) Job placement, (c) On-the-job-training, (d) Counseling and guidance--substantial, (e) Other services, and (t) Rehabilitation engineering. A significantly higher percentage of consumers with hearing loss received the following services: (a) Restoration, (b) Assistive technology devices, (c) Assistive technology services, (d) Assessment, and (e) Personal assistance services (which includes interpreters). A significantly higher percentage of consumers with other disabilities received the following services: (a) College or university training, (b) Business and vocational training, (c) Adjustment training, (d) Miscellaneous training, (e) Maintenance, and (f) Transportation. Data on the receipt of VR services, including percentages of each group receiving the service, are provided in Table 4.

Question 4 addressed whether differences existed in terms of amount of money spent on case services. Results indicated that significantly more money was spent on consumers with hearing loss than consumers with other disabilities. Refer to Table 3 for statistical test results and associated group means and standard deviations.

Employment Outcomes

Question 5 addressed differences in employment outcomes for the two groups. When evaluating the entire sample, a significantly higher percentage of consumers with hearing loss were closed successfully. There was also a difference in the same direction when evaluating the reduced sample (i.e., only those not competitively employed at application). However, the actual percentage differences were smaller with the reduced sample, as indicated by the odds ratio value being closer to one.

Differences in employment outcomes were also assessed by comparing type of successful closure for those closed in Status 26. For the full sample, significant differences were found. Primary differences were that a smaller percentage of consumers with hearing loss obtained competitive employment and a larger percentage of these consumers were closed as homemakers. These differences were much more pronounced when comparing the reduced sample. Consumers with hearing loss had a substantially smaller percentage closed in competitive employment, while having a substantially larger percentage closed in homemaker status.

An additional evaluation of employment outcomes was conducted involving the availability of medical insurance through an employer at close. No significant differences between the groups were found on this variable. Table 5 provides results for the statistical tests covering employment outcomes, as well as group percentages for each analysis.

Follow-up Analyses

Significant differences between the groups were found for three service variables that involve training: (a) College or university training, (b) Business or vocational training, and (c) Miscellaneous training. Because training is an important service in terms of future employability, even small differences in these areas are relevant. Consumers with hearing loss received all three of these services less frequently. One reason for this difference may have been that more consumers with hearing loss than consumers with other disabilities were competitively employed at application. Therefore, additional analyses based on information obtained from the initial analyses investigated for this study were conducted.

The full sample of 47,984 was reduced to include only those persons not competitively employed at application (as was done with the analyses of employment outcomes), resulting in a sample size of 35,385. Chi-square analyses were conducted again with this reduced sample for the three training service variables. Results indicated no practically significant differences between groups in terms of (a) College or university training, [chi square] (1, N = 35,385) = 4.31, p = .04 ([PHI] = .01; OR = .94), or (b) Miscellaneous training, [chi square] (1, N = 35,385) = 20.31, p < .0001 ([PHI] = .02; OR = 1.13). Significant differences between the groups in terms of receipt of business or vocational training remained, [chi square] (1, N = 35,385) = 97.50, p < .0001 ([PHI] = .05; OR = 1.37).

Discussion

The results indicate that there are significant differences in several areas of the VR experience for consumers with hearing loss compared to consumers with all other disabilities, as well as differences in terms of the two groups' demographic characteristics. A substantially higher percentage of persons with hearing loss were accepted for services compared to persons with other disabilities. In fact, the odds of being accepted for VR services were 2.7 times higher for persons with heating loss (as evidenced by the odds ratio of .37 associated with this contingency table).

In terms of demographic characteristics, consumers with hearing loss were significantly older on average than consumers with other disabilities. This finding might be expected as hearing loss is often associated with aging. Differences that were not expected included a larger percentage of consumers with hearing loss being female (51% of the heating loss group, while only 45% of the other disability group were female). According to the U.S. Census Bureau (2000), the incidence of heating loss in the population is considerably higher in males than females. The finding in this study raises a question of whether a proportion of men with hearing loss are not being served by VR. It is important to investigate why, in a population that men represent a clear majority, a higher percentage of women are being served. Another significant difference between the groups was found in terms of race. A higher percentage of consumers with hearing loss were White while a lower percentage were African-American, compared to consumers with other disabilities (differences in percentages of other racial groups were minimal). However, in this case, there is evidence that the prevalence of hearing loss is lower in the African-American population than in the White population in the U. S. (Adams et al., 1999).

Another difference in terms of demographic characteristics was that a substantially higher percentage of consumers with hearing loss were competitively employed at application. This difference in percentage was large, with 21% more consumers with hearing loss working in the labor force at application. One possible reason for this finding is that these consumers are older and may have lost some or all of their hearing after a long history of employment. In this case, they would perhaps need assistance with accommodations and assistive technology to maintain employment. However, some people in this situation would also need assistance in finding alternative employment if their jobs required hearing at a level no longer possible for them. Some evidence for this hypothesis is provided by evaluating the work status at application for consumers classified by RSA as having a post-vocational hearing loss. A substantially higher percentage of these consumers, compared to those classified as having a pre-lingual or pre-vocational hearing loss, were competitively employed at application (46% versus 32%). However, both of these percentages are substantially higher than the percentage of consumers with other disabilities who were competitively employed at application (i.e., approximately 16%).

Although there were significant differences in terms of the services received by each group, many of these differences could be anticipated based on type of disability. The services received more frequently by consumers with hearing loss are ones that are likely to be needed by this population. For example, restoration, which involves medical interventions such as surgery or therapy, is often provided to improve hearing or communication ability. Obtaining hearing aids or other assistive hearing devices would fall under assistive technology devices and services, and personal assistance services includes the use of interpreters. Assessment, which includes diagnosis and evaluation services, was also received slightly more frequently by this group, although the vast majority of each group received the service.

When evaluating the entire sample, slight to moderate differences in terms of the receipt of some training service variables were found. However, when evaluating only those consumers who were not competitively employed at application, these differences decreased. In fact, the only difference that remained significant was the percentage of consumers provided business or vocational training. It seems sensible that persons who are competitively employed at application will not have as much need for training services. It follows that this variable primarily accounted for the initial differences between the groups.

There was a significant difference between the groups in amount of money spent on consumers' cases, with an average of $486 more spent on cases for consumers with hearing loss. Although this is a moderately large amount, it must be evaluated in the context of the variability of case cost. Extremely high standard deviations were found for the average case cost of each group, with the standard deviation of the hearing loss group being the largest. This indicates that the hearing loss group had even greater variability than the other group, which may mean that several high outliers increased the average for this group. The large variability also resulted in a small effect size for this difference. Therefore, although a difference in the group means does exist, it is attenuated by the high variability of the average case cost for each group.

The differences found between the groups in terms of employment outcomes are important. The comparisons made with the reduced sample (of only those not competitively employed at application) will be discussed, as this is a more accurate portrayal of differences between groups based on receipt of VR services. A substantially higher proportion of consumers with hearing loss than with other disabilities were closed successfully. However, evaluation of the type of employment outcomes obtained indicates that fewer consumers with hearing loss were closed competitively (77% versus 86% for the other disability group) and more were closed as homemakers (17% versus 8%). Differences between the other closure statuses were small. This finding is relevant because competitive employment is generally considered the most desirable closure available. Although a higher percentage of consumers with hearing loss (of those not competitively employed at application) are closed successfully, many are not working in the labor force at the time of case closure. If non-competitive work was the goal of the consumer at application, this outcome is appropriate. Although it seems unlikely that a substantially greater percentage of consumers with hearing loss would have the career goal of homemaker, compared to consumers with other disabilities, it is not possible to assess this as information on occupational goal at application is not available in the RSA-911 data. However, some evidence that homemaker may not have been the original occupational goal of this many consumers with a hearing loss is that only 5% of the sample were classified as homemakers at application.

As stated previously, this study can only be considered a starting point in investigating the issue of the reduction in number and percent of successful closures for VR consumers with hearing loss, and it has limitations. A primary limitation of the study is the use of an ex-post facto design, which does not allow establishment of causal relationships. Having a large national database such as RSA-911 available for this type of research is certainly valuable. However, using pre-existing data limits the conclusions that can be drawn from the analyses. Another limitation is the restricted number of variables available for comparison. Although many variables are included in the RSA-911 database, several additional variables would have been of interest in this study, such as prior work history, occupational goal at application, and whether a degree or certificate was obtained (if the person received training).

Conclusions

The purpose of this investigation was to determine whether differences existed between VR consumers with hearing loss and consumers with other disabilities in terms of acceptance rates, demographic characteristics, services received, and employment outcomes. The impetus for the investigation of such differences was that the number and percent of consumers with hearing loss closed successfully by VR consistently decreased during a recent 10-year period (RSA, 2000). Based on the results of this study, being accepted for services does not appear to be an issue for the majority of persons with hearing loss (90% were accepted, compared to 77% for the other group).

In terms of demographic characteristics, a larger percentage of women than men with hearing loss were served by VR. This finding is relevant as a substantially larger percentage of the population of persons with hearing loss are men (U.S. Census Bureau, 2000). This may mean that there are a number of men with hearing loss who are not being served by VR. In terms of services received, the only potential bias against persons with hearing loss is that fewer of this group received business or vocational training, compared to consumers with other disabilities, even after taking employment status at application into account. Of course, this would represent a bias against consumers with hearing loss only if they needed the service to reach their job goal but it was not provided to them.

A relevant finding of this study related to employment outcomes was that a higher percentage of consumers with hearing loss were closed successfully, although fewer of those closed successfully were placed in competitive employment. Despite a high percentage of successful closures being a positive outcome, it does not explain why there has actually been a 10-year trend of a decline in the number of consumers with hearing impairment obtaining employment outcomes. Also, when a substantial percentage of these successful closures are being closed as homemakers, this may not necessarily be the optimal outcome for these consumers. The question of why a substantially greater percentage of consumers with hearing loss were closed as homemakers is relevant and deserves more attention.

To further investigate this consistent decrease in number of consumers with hearing loss being closed successfully, analyses could be conducted that compare the number of persons applying for services each year, during the 10-year period of the decline and before, to determine whether the number of persons with hearing loss who seek services is also consistently decreasing. If this is found to be the case, research involving direct contact with persons with hearing loss should be conducted, to determine why they have or have not sought services from VR. If it is not found to be the case, the decrease in successful closures may have originated in a smaller percentage of consumers with hearing loss obtaining an employment outcome, compared to the years prior to the noted decrease. Further research is needed in this area to understand the documented decrease in successful outcomes. Research beyond extant data studies, such as case audits and direct contact with consumers and non-consumers, would provide more detailed information and is certainly needed. This knowledge will help VR to ensure that consumers with hearing loss receive the assistance they require to obtain successful outcomes.

References

Adams, P. F., Hendershot, G. E., & Marano, M. A. (1999). Current estimates from the National Health Interview Survey, 1996. Vital and Health Statistics, 10(200). Hyattsville, MD: National Center for Health Statistics.

Butterworth, J., Schalock, R., & Gilmore, D. (1998). Rates of vocational rehabilitation system closure into competitive employment. Mental Retardation, 36(4), 336-337.

Capella, M. E. (2001). Vocational rehabilitation programs: Relationships among measures of effectiveness and state-to-state comparisons. Journal of Rehabilitation Administration, 25(1), 19-28.

Corthell, D. W., & Yarman, D. W. (1992). Serving the underserved--Principles, practices, and techniques, 19th Institute on Rehabilitation Issues. Menomonie, WI: University of Wisconsin-Stout.

Department of Education. National Institute on Disability and Rehabilitation Research: Correction notice for the final long-range plan for fiscal years 1999-2003, Federal Register, 64, 234 (Dec. 7, 1999).

El-Khiami, A. (1986). Selected characteristics of hearing-impaired rehabilitants of general VR agencies: A sociodemographic profile. In D. Watson, G. Anderson, & M. Taff-Watson (Eds.), Integrating human resources, technology, and systems in deafness (pp. 136-144). Silver Spring, MD: American Deafness and Rehabilitation Association.

Gibbs, W. E. (1990). Alternative measures to evaluate the impact of vocational rehabilitation services. Rehabilitation Counseling Bulletin, 34(1), 33-43.

Louis Harris & Associates (1998). N.O.D./Harris survey of Americans with disabilities. New York: Louis Harris & Associates.

Moore, C. L. (2001 a). Racial and ethnic members of under-represented groups with hearing loss and VR services: Explaining the disparity in closure success rates. Journal of Applied Rehabilitation Counseling, 32(1), 15-23.

Moore, C. L. (2001b). Disparities in job placement outcomes among deaf, late-deafened, and hard-of heating consumers. Rehabilitation Counseling Bulletin, 44(3), 144-150.

Moore, C. L. & Schroedel, J. G. (2001a). A national profile of the vocational rehabilitation of Americans with hearing loss. Unpublished monograph. Little Rock: University of Arkansas, Rehabilitation Research and Training Center for Persons who are Deaf or Hard of Hearing.

Moore, C. L. & Schroedel, J. G. (2001b). A profile of Kentuckians with hearing loss served by the state vocational rehabilitation agency. Unpublished report. Little Rock: University of Arkansas, Rehabilitation Research and Training Center for Persons who are Deaf or Hard of Hearing.

Platt, J. J. (1995). Vocational rehabilitation of drug abusers. Psychological Bulletin, 117, 416-433.

Rehabilitation Services Administration (2000). Information on the provision of vocational rehabilitation services to individuals with hearing loss (deaf and hard of hearing) [Information Memorandum RSA-IM-00-21, March 21, 2000]. Washington, DC: Department of Education.

Ries, P. W. (1994). Prevalence and characteristics of persons with hearing trouble: United States, 1990-91. Vital and Health Statistics, 10(188). Hyattsville, MD: National Center for Health Statistics.

U. S. Bureau of the Census (2000). Statistical abstract of the United States (120th ed.). [On-line document]. Available from http://www.census.gov/prod/2001pubs/statab/sec03.pdf.

Walls, R. T., & Fullmer, S. L. (1997). Competitive employment: Occupations after vocational rehabilitation. Rehabilitation Counseling Bulletin, 41, 15-25.

Walls, R. T., Misra, S., & Majumder, R. K. (2002). Trends in vocational rehabilitation: 1978, 1988, 1998. Journal of Rehabilitation, 68(3), 4-10.

Michele E. Capella, Ph.D., CRC, RRTC on Blindness & Low Vision, P.O. Box 6189, MS State, MS 39762. Email: mcapella@colled.msstate.edu

COPYRIGHT 2003 National Rehabilitation Association
COPYRIGHT 2003 Gale Group