The quality of life construct has gained prominent attention in human services over the last 20 years. We investigated whether quality of life differences exist between adults with developmental disabilities and the general population. Differences were found in scales measuring well-being and decision-making as well as other more specific variables. The two groups also differed in overall quality of life; those with developmental disabilities had lower quality of life. A logistic regression model comprised of the life dimensions differentiated between the groups with over 90% accuracy. Overall results of this static group comparison indicate that adults with developmental disabilities are at a significant disadvantage with regard to quality of life in comparison with the general population.
Quality of life is a construct that has begun to receive a great deal of attention in the developmental disability field. Development and subsequent testing of instruments that measure global and specific components of quality of life now provide the ability to examine services and their relationship to outcomes for the people receiving them. However, there are a number of questions related to using quality of life as an outcome indicator. Are we measuring quality of life for people with developmental disabilities only against others with disabilities, and, if so, what do we consider to be success? Should we not be measuring indicators of quality of life for the population as a whole? Historically, researchers examining quality of life between individuals with and those without disabilities has been sparse.
Although relatively limited, results from researchers who have compared those with disabilities and the general population have shown that individuals with disabilities have less autonomy (Kishi, Teelucksingh, Zollers, Park-Lee, & Meyer, 1988) and a lower quality of life (Lehman, Ward, & Lnn, 1982; National Organization on Disability, 2000; Ruth, 2002; Ruth & Struxness, 1994). Those who have more severe disabilities may, in fact, experience the largest gaps (Chubon, Clayton, & Vandergriff, 1995; National Organization on Disability, 2000). One comparative study between individuals with developmental disabilities and the general population conducted in the early 1990s (Ruth & Struxness, 1994) served as the basis for the development of a widely used instrument (Core Indicators consumer survey) that is now utilized by nearly half of state developmental disability authorities. Yet, these same states still use general population data that were collected in 1993 as the yardstick to compare current quality of life data of those with developmental disabilities (Ruth, 2002). Clearly, there is a need for more current general population data to determine how those with developmental disabilities are faring.
Perhaps the most familiar ongoing comparative research is conducted bi-annually by the National Organization on Disability (NOD). The NOD Harris poll surveys individuals with and those without disabilities with regard to employment, income, education, health care, access to transportation, entertainment/socializing, participation at religious services, political participation, and life satisfaction (National Organization on Disability, 2000). These results are based on people who self-reported physical or mental disabilities or health problems. According to the most recent Harris poll, less than one third of people with disabilities work (full- or part-time) compared to 79% of those who do not have disabilities. Twenty percent have not graduated from high school as opposed to 9% of those without disabilities. One third reported household incomes below $15,000, nearly three times more than in the general population. What is perhaps the most startling statistic is that 33% of those with disabilities are very satisfied with life in general in comparison to 67% of people without disabilities. This gap grows even wider when considering that only 26% of those with severe disabilities are very satisfied with their lives (National Organization on Disability, 2000). Quality of life for people with disabilities, as a whole, has shown some moderate increases in the late 1990s and early 21st century but to a lesser degree than for the general population (National Organization on Disability, 2000).
Not everyone has been quick to embrace quality of life studies particularly aimed at people with mental retardation because of potential to further demonstrate differences with the general population (Taylor & Bogdan, 1996). In addition, Edgerton (1996) argued that the subjective components of quality of life have been shown to be relatively unchanging in individuals. Although highs and lows of life may cause some brief alterations in one's perceptions of happiness, these levels will eventually return to baseline. This translates into the fact that some people are just inherently happier than others. We are probably all familiar with some individuals who have a characteristically sunny outlook on life, even when events are stacked against them. As a result, the more objective aspects of a person's existence can provide less static indicators of the construct. Although this may lead to the conclusion that the objective dimension of quality of life is the only one that can be changed through provision of services, this may not always be the case. The subjective global and domain specific dimensions are alterable in that inaccurate self-perceptions can be changed through interventions, such as counseling or medication (Livneh, 2001; Roessler, 1990). Indeed, one's perceptions may change markedly through rehabilitative efforts (Schalock et al., 2002).
When examining outcomes within service delivery systems, some may argue that what is actually being measured is lifestyle outcomes as opposed to quality of life. However, the rapidly growing body of work of Schalock and his colleagues attests to the fact that researchers can, indeed, examine the more encompassing construct, quality of life (Schalock, Alonso, & Braddock, 2002). The underlying power of quality of life studies appears to lie in the reality that regardless of professional or personal perspective, the construct is recognizable and sensitizing. Schalock (1997) asserted that certain principles are inherent to any quality of life outcomes study, regardless of the population being studied. He argued that the key elements of the quality of life construct can and should be measured. For those with severe, complex disabilities, it is merely a matter of choosing the most appropriate methodologies (Schalock, 1997). First and foremost, conducting any quality of life research with those who have disabilities first involves discarding stereotypes (Taylor & Bogdan, 1996) and utilizing techniques to obtain the perceptions of the participant. Perhaps more than any other construct in the social and medical sciences, quality of life can and should be considered a concept that is unique to the human experience, one that crosses lines of disability, gender, age, and socioeconomic status. When, however, the core factors of quality of life are employed, results can and should be generalizable to the general population (Cummins, McCabe, Romeo, & Gullone, 1994; Schalock, 1996). Quality of life has many layers and many intricacies, but it is also a construct that can yield very informative and important results for all people. When researchers employ sound methodological practices, the information they yield is not only theoretical and practical; they can provide data that can be used to guide policy development for human service delivery systems. In this study we sought to develop an increased understanding of the quality of lives of people with and without developmental disabilities. Specifically, we hypothesized that differences exist with regard to well-being, autonomy, community participation, rights and access to services, as well as an overall difference in quality of life.
The consumer sample was composed of 502 adults with mental retardation or other developmental disabilities who participated in the Core Indicators Consumer Survey between August 2001 and February 2002 in a southern state. This represents an 80% response rate. Selection of the sample of those with developmental disabilities receiving state-funded services was randomly drawn through the data-management team for the state's Division of Mental Retardation. All consumers and guardians (as applicable) who took part in the survey gave their written consent to participate. Of the 502 consumer participants, 194 proxies responded on behalf of consumers in those instances when the severity of the consumer's intellectual disability precluded him or her from participating in the research or giving verbal or written consent to do so. Written guardian consent was obtained prior to any interviewing activities.
General population group
Participants in the telephone survey were 590 adults who were living in the community and had access to a telephone. The survey was administered May through June 2002. A 36.4% eligible response rate was achieved. This is somewhat low (Dillman, 2000) but acceptable for this mode of survey administration (Tuckel & O'Neill, 2002). We selected participants by using the Waksberg random-digit dialing procedure, ensuring that everyone with a telephone had an equal chance of being selected, regardless of whether the phone number was publicly listed (Waksberg, 1978).
The Core Indicators consumer survey was used as the basis for this study. This instrument was developed and piloted in 1997 by 7 member states of the National Association of State Directors of Developmental Disability Services and the Human Services Research Institute, in conjunction with national methodology and development disability professionals (Human Services Research Institute, 2001), for use with those who have developmental disabilities and are receiving state-funded services through their state developmental disability authority. The number of states taking part in this effort has increased to nearly half of all states (Human Services Research Institute, 2003). The instrument has undergone test–retest reliability testing with individuals who have developmental disabilities (.80). In addition, consistency of some scales has been tested. The Choice scale was found to be highly consistent, α = .96, and community inclusion's internal consistency was also respectable, α = .76 (Human Services Research Institute, 2004).
The instrument addresses the multidimensionality of quality of life and includes items related to relationships, safety, health, choice-making opportunities, community participation, well-being and satisfaction, and rights. Responses may be obtained through self-report or by proxy. Although satisfaction and personal perception items are only valid when reported by the individual receiving services, items related to community participation, services, and choice may be answered by proxy if the severity of the consumer's intellectual disability prevents his or her self-report. This is considered an acceptable practice by researchers in the field of developmental disability methodology (McVilly, Burton-Smith, & Davidson, 2000; Schalock, 1996). Most items of the direct interview are yes/no or utilize a 3-point scale.
We collected demographic and contact information by randomly contacting agencies who serve adults with developmental disabilities. Project staff contacted the consumer or guardian (as appropriate) to request participation in the survey. If written consent was received, the instrument was administered, primarily as a face-to-face interview. Incomplete, inconsistent, or invalid responses from the consumer group were removed as outlined by Core Indicators consumer survey protocol (Human Services Research Institute, 2001). This resulted in a sample of 502 consumers. Of this number, 308 individuals responded on their own behalf; 194 proxies answered on behalf of consumers within previously outlined guidelines. Because some surveys were completed by proxy, items related to well-being and satisfaction could not be answered.
General population group
To administer the instrument via telephone to the general population, we made some alterations. Items pertaining to home support staff, support coordination, advocates and guardians, attendance at self-advocacy meetings, and ability to use the telephone were omitted. A pilot telephone survey was conducted to determine ease of administration. Question order and the construction of some items were altered as a result of the pilot survey, primarily because demographic information that was provided by agencies for the consumer group was now integrated into the telephone survey. Questions asked of the general population group that were not used with the consumer group covered education level, household income, occupation, presence of disability (and open-ended descriptions thereof), and comments at the conclusion of the survey. Following pilot-testing, personnel from a university survey research center conducted the telephone survey. They used a 22-line telephone bank and a pool of trained interviewers. A total of 590 general population participants took part. With respect to those in the general population group surveys excluded from analysis, 3 people who were rated as having poor understanding of the survey were not included in data analysis. In addition, 1 respondent's self-reported age was 17. Ten people terminated the interview prior to providing their ages. The final sample was 576 general population participants.
Analysis of Data
Data were analyzed to determine whether differences in quality of life as measured across the chosen life dimensions existed between the two groups. Variables were assessed for normality and frequency distribution. Variable means, standard deviations (SDs), mode, and skewness were examined. There was a gender bias in the general population sample (33.5% males vs. 56.2% males in the consumer group). Therefore, a random subsample providing approximately equal gender distributions between the general population and consumer groups was drawn, resulting in 344 participants (56.1% male) selected for the gender-adjusted general population subsample. Data were analyzed for both the full general population sample and the gender-adjusted subsample. With few exceptions, no differences appeared between the full sample and gender-adjusted sample. In this study we present the gender-adjusted results.
Frequencies, means, and SDs were reported on each item for each group. We performed t tests at the item and scale level to determine whether significant differences between groups existed. Because each of the four scales (Autonomy, Community Participation, Well-Being, and Access–Rights) was considered to represent distinct dimensions of quality of life, a total quality of life score was calculated to give an overall composite measure. A t test was also calculated on the total score statistic. Because we made no assumptions with regard to whether the consumer group would score higher or lower than the general population group on individual quality of life items, t tests were two-tailed. Because tests were performed as an item and composite measure, the significance level was set conservatively at .05/ 40. The numerator, .05, represents the alpha typically accepted error rate in much social research. The denominator represents the total number of items used to comprise the scales (36) plus the number of scales (4). This resulted in the alpha level being set at .0013. The 194 proxy respondents did not complete items related to well-being. Therefore, no Well-Being or total Quality of Life Scale scores were calculated for these individuals.
In order to make inferences about the populations being studied, we conducted multivariate techniques. Because the Well-Being scale was not conducted with proxies, 194 cases were considered missing, resulting in 308 individuals in the consumer group and 344 general population participants being included in multivariate analysis. Prior to conducting the regression analysis, we examined all variable means, SDs, skewness, and correlations to ensure that the variables were appropriate to use. Logistic regression was chosen because it enables prediction of group membership of each dichotomous outcome variable based upon predictor variables. We ran the logistic regression using each of the quality of life dimension scales to determine whether any were significant contributors to the dependent variable.
Internal consistencies of the instrument were established for this study. Cronbach's alphas were computed to determine the cohesion of survey items that composed the Life Dimension scales. Because people with developmental disabilities provided the basis for development of the instrument, we used the consumer group to determine the alpha levels of the four Life Dimension scales, which were based on items that had common characteristics in accordance with Schalock's (1996) determinants of those global areas that comprise quality of life for an individual. Table 1 contains items used to create the scales. To be included as a life dimension scale, at least half of the items that made up that scale must have been responded to. We applied this rule to avoid overgeneralizing the importance of individual items within the scale and to ensure that the mean was truly reflective of the items it contained.
Further analysis was conducted on missing cases to determine whether differences existed. Those in the consumer group who had proxy respondents had more severe disabilities than those who responded on their own behalf, χ2 (5, N = 502) = 85.22, p < .001. Because of the relative brevity of the instrument, some areas were combined to create more reliable scales. This is the case for the Access– Rights scale, α = .60, which contained 11 items regarding access to transportation, services, privacy, and medical and dental care. The Autonomy scale, α = .80, included 8 items relating to choice, from everyday decisions to those that have major impacts on one's life. The Well-Being scale, α = .53, was comprised of 11 items related to happiness, relationships, and satisfaction with specific activities. The Community Participation scale, α = .62, consisted of 6 items signifying involvement with community activities. Recent reliability estimates conducted using the national Core Indicators sample across all participating states resulted in similar, slightly higher internal consistencies (Human Services Research Institute, 2004). However, not all items used to create those scales were included in the present study. Not all instrument items were included in the scales, but item-level t tests are presented for each item.
Because a random sample was drawn to adjust for the gender discrepancy between the groups, the gender frequencies were nearly identical. Consumers averaged 40 years of age versus 45 for the general population, t = 5.1, p < .001. There were no racial differences between groups, p = .05. With regard to marital status, vast differences were apparent, χ2 (2, N = 843) = 494.3, p < .001. Ninety-five percent of consumers were single as opposed to 20% of the general population group. Over 60% of the general population were married compared to 3% of consumers. Almost 20% of the general population had been married previously versus 2% of consumers.
The nature of the sample was such that all individuals in the consumer group had mental retardation or other developmental disabilities. Many had multiple disabilities. Nearly one third also had mental illness; 16%, cerebral palsy; 5%, autism; and over 40%, some other disability in addition to a developmental disability. Approximately 20% of the general population indicated having a disability. Of those who specified the disability, 3% indicated mental illness as their primary disability. However, most (14.8%) fell into the “other” category, which included chronic pain, arthritis, diabetes, orthopedic impairment, hearing impairment, visual impairment, cardiac impairment, hepatitis, amputation, skin disorder, and cancer. Of those in the general population sample, virtually all lived in their own, privately owned residence or with family (99%). One respondent resided in a retirement home and 3 lived in military housing. These 4 cases were classified as “other.” The consumer group had more varied living arrangements. Approximately 40% owned, rented, or lived with family. Another 40% lived in group homes (staffed residences), with foster families, or in agency-owned housing. Nearly 15% of consumers lived in institutions or nursing homes.
Quality of Life Dimension Item Responses
Table 1 shows means, SDs, and t-test results for each quality of life dimension item. An item was considered to be significantly different between the two groups at p = .0013. With regard to well-being, the consumer group was more lonely, wanted to work more hours, and was more afraid at home. Those in the consumer group were also less likely to have friends. Every item related to choice was significantly different between consumers and the general population, p < .001. Those who were in the consumer group found themselves less likely to choose where they lived, with whom they lived, daily schedule, what free time activities in which to participate, and what to purchase with their own money. Consumers also had fewer options to choose from with regard to day activities and place of residence. In terms of community integration and participation, no differences were found between those with developmental disabilities and the general population. Several differences were found with regard to rights restrictions. Consumers were more likely to have their mail opened without giving their permission as well as less likely to have privacy when guests would come to visit. In terms of access, consumers were more likely to have difficulty in obtaining needed transportation services.
Quality of Life Dimension Scales
Four scales that captured related quality of life dimensions were created from survey items. These scales represented Well-Being, Access–Rights, Autonomy, and Community Participation. T tests were generated to determine whether differences existed between the groups at a scale level. When comparing the general population group and the consumer group (see Table 2), we found significant differences between two of the four scales. The Autonomy scale showed great dichotomy between the two groups, with the general population scoring nearly one-half point higher than consumers on the scale. This was expected because the general population group had an advantage over consumers in every choice-related item on the instrument. With regard to Well-Being, the consumer group again fell short. The Access–Rights and the Community Participation scales were not significant at the more conservatively set alpha of p = .0013. All four scales were positively correlated, p < .01, because increases in one likely indicated increases in the other (see Table 3). A composite overall Quality of Life dimension score was significantly different between the groups, with the consumer group receiving a lower average total scale score. This indicates that differences do, indeed, exist with regard to these quality of life dimension scales.
Because the Well-Being scale was only calculated for consumer group members responding for themselves (n = 308), the logistic regression was run using this smaller subset of the consumer group. Significant positive correlations were found between almost all of the scales (see Table 3). However, because none of these correlations was above .40, concerns associated with multicolinearity and associated relationships between predictor variables were lessened (Licht, 1995).
Logistic regression techniques were used to determine whether the independent variables could be used to predict the probability that a particular observation belonged either to the consumer group or to the general population group. The independent variables used in the logistic regression were well-being, community participation, access, and autonomy. Age, race, marital status, and gender were also entered into the logistic regression to see whether they played a role in distinguishing between the groups. The logistic regression model did prove to be predictive, χ2 (10, N = 651) = 582.15, p < .001. It was able to explain between 59% to 79% of the variance that was present in the sample. This model correctly predicted group membership 90.8% of the time. The regression classified both groups well, with a slight advantage given to those in the consumer group (92.2% vs. 89.5%). Table 4 contains regression results. Well-being, community participation, and autonomy were found to be significant predictors, along with marital status. Access, gender, and race were not significant at the .0013 level. With regard to predictive power of the independent variables, as Autonomy and Well-Being scores increased, individuals were more likely to belong to the general population group. Higher community participation scores were more indicative of belonging to the consumer group. In comparison to single respondents, those who were married or who had been married were more likely to belong to the general population group. The marital variable was the strongest predictor, followed by well-being and autonomy. Because of the nature of the results, direct comparison of predictive value of community participation was not possible; however, in this model it was also a very strong predictor.
Examination of the differences present between those with developmental disabilities and the general population is indicative of discrepancies that may be present across the state. We found that people with developmental disabilities experienced a significantly lower quality of life. It is quite apparent that they have less choice, from the mundane day-to-day decision-making of what to do in their free time to the more important life decisions of where to live and work. Because a high proportion of those in the general population group were married, it was expected that respondents would also indicate that they might find themselves with less complete decision-making authority if they felt that they were part of a decision-making team with their spouse or family. Even in light of this, the general population group members still considered themselves to be far more autonomous than did those in the consumer group. Examining levels of community participation, although roughly equivalent, leads to further questions as to the shape of the consumers' activities in integrated settings. For example, these results show that one would expect either group to go out to eat at the same level; however, a person with a developmental disability who goes out is far less likely to have decided where to go. This finding is re-emphasized by Kishi, Teelucksingh, Zollers, and Park-Lee (1988), whose work showed that for people with developmental disabilities in community settings, opportunities for choice-making were fewer when compared with those of the general population.
Results also show that individuals in the consumer group were at a disadvantage with regard to access to transportation and some basic human rights. Consumers experienced more rights restrictions in the form of less privacy with guests and others opening their mail for them. Our results also reinforce the fact that inadequate transportation is a grave concern for people with disabilities (National Organization on Disability, 2000). Given the highly rural nature of the southern state in which this study was conducted, we recognize that transportation could be an issue for everyone. However, these results indicate that when a significant disability is present, a lack of transportation is an even greater problem, which could further explain the discrepancies found in employment, where a significantly higher percentage of the consumer group indicated that they wanted to work more hours. How can one be expected to maintain a job or some kind of day activity outside of the home if there is no consistent means to get there? Although additional examination of the impact of employment status would prove valuable, this data set did not contain vocational items beyond participants wanting to work more than they currently did.
Overall well-being as measured by this instrument was significantly different between the groups. This was a disheartening, but not totally unexpected finding. Those with disabilities were more likely to be lonely, more likely to be afraid in their homes, and less likely to have friends. Therefore, though consumers in this study were going out in their community at equal levels with their nondisabled counterparts, one must again question the means through which this occurred if people were taking part in what can be considered social activities but are still lonelier and have fewer meaningful relationships than do members of the general population.
At a more global level, the general population experienced what could be perceived as a significantly better overall quality of life. The statistical model used to determine whether the chosen life dimensions could accurately predict whether a person was a member of the general population group or a consumer participant was an extremely strong tool. This reinforces the fact that the life dimensions between the two groups were so different that, 9 out of 10 times, they alone could determine to which group a person belonged. The largest contributor to the overall variance, however, was marital status. In fact, if the marital status variable is removed from the logistic regression, well-being was the most significant predictor by far (although this also led to the percentage accurately classified to decrease by 10%).This finding alone is a sad indicator of the discrepancies that are present in the lives of people with developmental disabilities and those in the general population. Because in this study we sought to find answers to the question of whether differences existed between the two groups, it appears that whether one uses an item, scale, or more holistic view, many quality of life differences are apparent between individuals with developmental disabilities and other residents in their state.
Although our overall intent in this study was to outline differences between the consumer and general population groups, we also found interesting similarities. A very surprising finding was that there were very few appreciable differences in terms of community participation between groups. People with and those without developmental disabilities were equally likely to take part in a variety of offerings present in their cities and towns. This is somewhat conflicting with the National Organization on Disability Harris poll findings (2000) showing that people with disabilities enjoy less participation in leisure activities. However, the Harris poll was not limited to those with developmental disabilities. Also, community participation items in this instrument were yes/no, whereas the Harris poll considered frequency of activities. It is quite likely that had frequency been included in the present study, the inequities would have resurfaced.
Limitations are inherent in this type of study. Findings may be substantially different for people with developmental disabilities who are not receiving any services or by those in service systems that operate outside the state's publicly funded developmental disability agencies. Because states vary considerably in service provision for individuals with developmental disabilities, care should be taken when generalizing to other states. Though the methodology permitted interview by telephone (with discretion), the consumer group was largely surveyed in a face-to-face interview format. On the contrary, all general population participants consented to answer the survey on the telephone. This dichotomous methodology does not equate to a completely complementary treatment of the two groups. In addition, the consumer survey could be answered by proxy on behalf of the consumer, whereas the general population survey was conducted solely as a self-report. Though proxy responses are acceptable (Schalock, 1996), this, too, is an unequal treatment of the groups. Internal consistencies (alpha levels) calculated in this study were moderate for the consumer group.
The present study exhibits several methodological trade-offs. Psychometric test results of the instrument is, at present, limited. However, we believe that its widespread use and relative ease of transferability to the general population proved a good starting point for comparative research on quality of life. We included people with a broad spectrum of diagnostic labels, from mild to profound mental retardation. Although we could have limited this study to those with mild to moderate mental retardation, those with more severe disabilities would have been excluded. We opted, instead, to not exclude those who are so often ignored. This choice meant that further limitations were present because of a mixed methodology via use of proxies versus self-respondents. Although this decision resulted in missing data in the Well-Being dimension, those with more severe disabilities were still represented in all the other scales. In addition, the consumer group was also heterogeneous in terms of residence, with 15% living in institutional settings. We recognize limitations in this regard, but to exclude those in institutional settings would have clouded the true picture of overall experiences of people with developmental disabilities. We expect that this will, however, give impetus for further analysis of subgroups within the sample.
Implications for the Future
Replication of this study on a broader scale could be used to determine whether findings hold up nationally. Replication over time would provide valuable information about trends that are occurring. This study provided some interesting insight into quality of life and subsequent differences experienced by those with developmental disabilities. Although the differences that were found are certainly well worthy of future study, so too are some of the similarities. Although no significant differences were found in the health-related items, one must still define what constitutes acceptable levels of participation in exercise and recency of medical and dental preventive care. It is important to recognize that these findings do not indicate that those in the disability group were healthy or exemplary in their levels of community participation. Rather, it could mean that those in the disability group were as unhealthy, uninvolved, or likely to experience rights restrictions as those in the general population.
At a surface level, the two groups in this study appear to be separated by a single difference: presence of a developmental disability. However, the groups also represented other differences in composition that are worthy of further exploration. Nearly 20% of the general population participants indicated having some kind of disability. An examination of this subset could yield important information. This group can be compared with the rest of the general population sample as well as with the consumer group. Because this analysis would include people with a variety of disabilities, results could also be compared with the most recent National Organization on Disability Harris poll (2000).
Though the methodology employed in the present study was not a flawless one, it draws a picture of difference in quality of life for persons with developmental disabilities. Given the wide adoption of this instrument, it has the potential to yield very powerful data. Provision of general population quality of life data adds richness to consumer data. It serves as a very accessible comparison, whether for families of individuals with disabilities, human service providers, legislators, and the public at large. But perhaps most importantly, this study can serve as a foundation for the future. It is important not only to establish that differences are present for certain groups of community members, but to take that knowledge and develop strategies to erase those gaps. We did not design this study to solely highlight the chasms that exist within our society but, rather, to provide data that can be used to help to bridge them. Clearly, studies such as this one provide further proof that providers of services must be cognizant of quality of life outcomes for those they serve. Merely providing services does not equate to a level quality of life playing-field as compared to others in the communities in which those with developmental disabilities live.
The authors acknowledge Fred Danner and Ralph Crystal, who served on the first author's doctoral committee and contributed to the development of this research. Additional thanks are given to Kevin Lightle, Director, Kentucky Division of Mental Retardation, for his active support of this research.
Authors: Kathy Sheppard-Jones, PhD (email@example.com), Disability Program Administrator, 209 Mineral Industries Building; Harold Kleinert, EdD, Executive Director; H. Thompson Prout, PhD, Professor (Department of Educational & Counseling Psychology), Interdisciplinary Human Development Institute, University Center for Excellence, University of Kentucky, Lexington KY 40506-0051