Abstract

The quality of life of 188 young adults with moderate or severe mental retardation was examined. Schalock and Keith's (1993) Quality of Life Questionnaire (QOL-Q) was used as the primary outcome measure. Young adults who had exited high school had significantly higher overall quality of life scores than did those who were still attending school. Individuals who had jobs in the community also had significantly higher levels of quality of life. Although young adult adaptive functioning was the single largest indicator of the QOL-Q index total score and three of the four subscales, it was not related to scores on the Satisfaction subscale. Here, family- and environment-related variables played a greater role. Implications and directions for future research are discussed.

Quality of life has emerged as an important concept in reference to individuals with mental retardation and developmental disabilities. The application of this concept, defined a multitude of ways (Felce, 1997; Halpern, 1993), has been used to improve the lives of persons with disabilities (Schalock, 1997). Traditionally, the concept of quality of life has been applied to adult populations of individuals with mental retardation. In particular, with adult populations, measures of quality of life have been used to determine the extent to which an individual's needs and desires are being met (Felce, 1997; Schalock, Bonham, & Marchand, 2000). This can include assessment of service needs for adults with mental retardation and/or assessment of individual outcomes, such as empowerment, independence, social belonging, community integration, and satisfaction (Schalock & Keith, 1993).

The concept of quality of life is equally important for younger populations of individuals with mental retardation, including transition age young adults. Transition can be defined as “the life changes, adjustments, and cumulative experiences that occur in the lives of young adults as they move from school environments to more independent living and work environments” (Wehman, 1996, p. 7). For many young adults with mental retardation, particularly mild or moderate, transition is the time when they are leaving the school system and moving toward more adult roles. Schools and families can play an active part in helping these young adults achieve their postschool goals, including increased independence and autonomy, satisfying vocational pursuits, community living opportunities, and increased social networks and supports. For young adults with more severe mental retardation, transition often involves a shift from public school special education services to the adult service delivery system (Kraemer & Blacher, 2001). Postschool outcomes previously identified as germane to the transition process may be less relevant than quality of life issues for these individuals who are lower functioning. Thus, this is a unique period in the lifespan to investigate quality of life, particularly for individuals with more severe mental retardation; however, little empirical work exists.

Families are critical to the well-being of individuals with mental retardation and can have a tremendous impact on their quality of life (Seltzer & Krauss, 2001). In terms of transitioning young adults, investigators have noted that parents and families, not just the young adult with mental retardation, are affected by transition and, thus, should be actively involved in the transition process (Morningstar, Turnbull, & Turnbull, 1996; Szymanski, Hershenson, & Power, 1988).

One would expect many factors to be related to a young adult's quality of life during the period of transition, including individual characteristics, variables pertaining to the family, and variables relating to the environment. A conceptual model for understanding these relationships is presented in Blacher (2001). Essentially, by evaluating transition outcomes and processes within a multidimensional framework, one can gain a better sense of the breadth of the transition experience and identify an array of variables that influence quality of life as a transition outcome. Previous work has indicated that many variables can affect transition outcomes for students with disabilities, including family background, the quality of a student's school program, and the nature and quality of transition services received (Halpern, 1993). In addition, Hughes and Hwang (1996) have suggested that environmental factors can interact with individual characteristics to affect an individual's quality of life.

Although consensus is lacking on a specific definition of quality of life (Borthwick-Duffy, 1989; Felce, 1997; Rusch & Millar, 1998; Taylor & Bogdan, 1990), it is generally viewed as multifaceted, comprising both objective and subjective dimensions (Cummins, 2000; Felce, 1997; Goode & Hogg, 1994; Hatton, 1998; Hughes, Hwang, Kim, & Eisenman, 1995). In a review Halpern (1993) suggested that most investigators have assessed quality of life objectively, that is. by direct observation and measurement (Goode, 1990). For example, factors relating to employment, educational attainment, and social networks have been measured (Halpern, 1993). However, proponents of quality of life models for individuals with disabilities in general, and mental retardation in particular, have increasingly stressed the importance of subjective indicators. These include aspects relating to personal well-being and life satisfaction (Hatton, 1998). Thus, the growing practice of examining quality of life as a transition outcome for individuals with mental retardation is an attempt to capture not only conventional outcomes, such as employment, but also other subjective aspects of transition, such as autonomy, self-determination, and personal well-being.

The most commonly used instrument in assessing quality of life among individuals with mental retardation is Schalock and Keith's (1993) Quality of Life Questionnaire—QOL-Q (Cummins, 1997, 2001). This measure includes both objective and subjective dimensions of quality of life and has been shown to be reliable in populations of individuals with mental retardation, including those with severe mental retardation (Schalock, Lemanowicz, Conroy, & Feinstein, 1994). This measure contains four subscales: Satisfaction, Competence/Productivity, Empowerment/Independence, and Social Belonging/Community Integration. The QOL-Q has been used as an outcome in assessing indicators of quality of life in populations of individuals with mental retardation (e.g., Schalock, Lemanowicz, Conroy, & Feinstein, 1994; Schalock et al., 2000); however, these populations are typically older adults (mean age in the 40s), with most living out of the family home (Seltzer & Krauss, 2001). In general, it is relatively unknown what the quality of life looks like for transition-age young adults with moderate or severe mental retardation. Moreover, indicators of quality of life for this population are in need of investigation.

Our purpose in the present study was to examine quality of life for a sample of young adults with moderate or severe mental retardation during transition. The research questions used were: (a) What does quality of life look like for a sample of young adults with moderate or severe mental retardation during transition? (b) What young adult, family, and environmental variables predict an overall quality of life index for these individuals? (c) Do different variables predict subjective versus objective dimensions of quality of life for this sample during transition? In the present study, Schalock and Keith's (1993) QOL-Q total score was used as an overall index of quality of life. The Satisfaction subscale was utilized as a subjective dimension of quality of life, with the other three subscales encompassing more objective dimensions (Cummins, 2001).

Method

Participant Selection

The sample consisted of 188 families with transition-age sons or daughters who had moderate or severe mental retardation. These families were selected from an intact sample of 212 families with young adult children who had mental retardation; the 24 young adults not included in this study were younger than 18 years old, and the QOL-Q is not appropriate for individuals under the age of 18 (Schalock & Keith, 1993). Participating families were obtained with the help of the Regional Center system in California, a state-wide network in which all identified individuals with mental retardation are registered. Regional Center staff mailed letters of invitation to families who met the criteria of having young adults between the ages of 16 and 26 with moderate to severe mental retardation and parents who could speak and complete questionnaires in English. Regional Centers continued to send out letters until the response criterion (e.g., 100 in one mailing, 150 in another) was reached. The intact sample of 212 consisted of families who could be reached and interviewed.

Sample Characteristics

Mothers were respondents in 86% in the interviews. Fathers served as respondents in 3% of the interviews, both mother and father in 9%, and grandparents in 2%. The mean age of mothers was 49 years (range = 38 to 70), and more than two thirds of them (67%) were employed. Two thirds of mothers had completed at least 2 years of college. The majority of the families (69%) were two-parent, and 73% of the sample was Anglo. Family income varied tremendously, with 58% of the families earning over $50,000 per year.

The target individuals were 188 young adults with moderate or severe mental retardation. Their mean age was 21 years (range = 18 to 26); 49% were male. Although all young adults qualified for Regional Center services in California under the primary diagnosis of mental retardation, disability as reported by the parent varied. The most commonly reported additional diagnoses of disabilities were cerebral palsy (33%), Down syndrome (15%), and autism (10%). Seventy-six percent of the young adults were ambulatory. The majority still lived at home with their parents (83%). Those young adults that lived out of the family home (17%) were still very connected with their families and had lived out of the home for no longer than 5 years. In addition, research has shown that even when individuals with mental retardation move out of the family home, families are still very involved in their lives (Baker & Blacher, 1993; Blacher & Baker, 1994; Blacher, Baker, & Feinfield, 1999).

Forty-five percent (or 85 young adults) had exited high school at the time of the interview. During their school years, all the young adults attended public schools, primarily in Southern California. For those 103 who had not exited school at the time of the interview, 65% attended segregated special education classes all day, 10% attended one or two general education classes and the rest special education classes, and 25% spent some of their day in the community (i.e., at a job site, practicing skills in the community) and the rest of the day in segregated special education classes. For those young adults who had exited the school system, 40% were working in a sheltered workshop or day activity center; 20%, in a paid supported work environment; and 40% were not working at all.

The young adults evidenced significant difficulties in adaptive behavior as measured by the Vineland Adaptive Behavior Scales—the Vineland (Sparrow, Balla, & Cicchetti, 1984). The mean standard score for the Adaptive Behavior Composite was 26 (standard deviation [SD] = 10.8), with two thirds of the sample (n = 126) in the severe range of functioning (Vineland standard scores of 20 to 35). The mean developmental age equivalency for the group was 3 years, 4 months. For the sample as a whole, relative strengths were in the Socialization domain (mean standard score = 30.5), with the area of greatest need being communication (mean standard score = 22.5) (see Table 1).

Table 1

Young Adult Characteristics (N = 188)

Young Adult Characteristics (N = 188)
Young Adult Characteristics (N = 188)

Maladaptive behavior, as measured by the Maladaptive/Problem Behavior section of the Scales of Independent Behavior-Revised (Bruininks, Woodcock, Weatherman, & Hill, 1996), was evident. Twenty percent of the sample (n = 38) engaged in moderate to serious levels of maladaptive behavior.

Instruments

The measures administered pertained to (a) family and young adult characteristics; (b) family impact, stress, and coping; (c) family involvement; (d) school programming/transition experiences; and (e) quality of life.

Family and young adult characteristics

The Family Data Sheet, a demographic questionnaire that has been used previously with these families (e.g., Blacher, Hanneman, & Rousey, 1992), consists of questions regarding demographic information for both mother and father as well as questions related to the young adult, such as age, gender, diagnosis, ethnicity, and ambulation.

The Vineland was administered to caregivers as a structured interview. This measure yields four domain scores, Communication, Daily Living Skills, Socialization, and Motor Skills and an overall Adaptive Behavior Composite. In this study we utilized the Adaptive Behavior Composite score. The alpha reliability coefficient for this score is .92. Alphas for the subscales range from .79 to .86.

We also used the Scales of Independent Behavior-Revised (SIB-R) Problem Behavior Scale (Bruininks et al., 1996). The Problem Behavior Scale of the SIB-R provides a General Maladaptive Index comprised of eight problem areas organized into three broad Maladaptive Behavior Indexes (Internalized, Asocial, and Externalized). Each of the eight behaviors is rated according to frequency of occurrence and severity. Scores can range from +10 (good) to −74 (extremely serious). The mean for “normal” samples is 0 (SD = 10). A cut-off score of −22 and below was used to classify maladaptive behavior as moderately serious to serious. For the purpose of the present study, only the General Maladaptive Index composite score was used. The mean and SDs for the General Maladaptive Index can be found in Table 1.

Family impact, stress, and coping

To measure this domain, we used four instruments. The Family Impact Questionnaire (Donenberg & Baker, 1993) is a self-administered 50-item Likert-scale measure that assesses parent perception of their young adult child's impact on the family relative to the impact most young adult children have on the family. It yields six subscales, but only two were used in the present study: Negative Feelings About Parenting (11 items) and Positive Feelings About Parenting (7 items). On this measure respondents rate statements (e.g., “My young adult is more stressful,” “My young adult is easier to care for”) according to a 4-point scale, with 0 indicating not at all and 3 indicating very much. For the current sample, the Positive Feelings and Negative Feelings About Parenting subscale alphas were .85 and .75, respectively.

We also used the Questionnaire on Resources and Stress-Short Form—QRS-F (Friedrich, Greenberg, & Crnic, 1983), a 52-item questionnaire designed to measure the influence of an individual with a disability on the life of other family members. Although the measure includes four factor scores, we used only the Parent and Family Problems subscale. On this measure parents respond to true/false statements (e.g., “Other members of the family have to do without things because of ______”). Higher scores are indicative of greater stress. For the current sample the alpha for the Parent and Family Problems subscale was .87.

We employed the Family Crisis Oriented Personal Evaluation Scales—F-COPES (McCubbin, Olson, & Larsen, 1981), a 29-item questionnaire that assesses coping strategies in families. Questions are presented on a 5-point Likert-scale, with 1 indicating strong disagreement and 5 indicating strong agreement. It consists of five subscales and a total scale. The total scale was used in the present study. The alpha for the total scale was .79.

Support, an informational questionnaire, has been used previously with these families (e.g., Blacher et al., 1992). It consists of 20 questions, not intended to serve as a scale, regarding the types of support (emotional and financial) parents receive from government agencies, community groups, family, and friends. It is also used to survey their level of satisfaction with the types and amount of support they receive. The following items from Support were used in the present study: (a) caregiver's happiness with the amount of support received from family, (b) caregiver's happiness with the amount of support received from nonfamily, (c) number of people caregiver can count on for help with the young adult, and (d) number of respite hours received.

Family involvement

To determine family involvement, we employed the Parent Involvement in Transition Planning, a 17-item questionnaire designed specifically for this study. It assesses parents' level of involvement and opportunity to be involved in the transition planning process. Questions consist of dichotomous and Likert-scale items designed to investigate three types of involvement: behavioral (8 items, e.g., kinds of involvement parents actually had in transition planning), cognitive (3 items, e.g., how much parents thought about various aspects of transition during the transition process, such as work, community living, and access to social activities), and emotional (3 items, e.g., how much parents worried about various aspects of transition during the transition process, such as work, community living, and access to social opportunities). Specific items used in the present study were (a) parent involvement in transition planning (i.e., behavioral involvement), (b) parent knowledge of adult services, and (c) parent overall satisfaction with involvement in transition planning.

School programming/transition experiences

To investigate school programming/transition experiences, we used the Transition Experiences Survey. This measure, described more fully in Kraemer and Blacher (2001), is a 48-item interview protocol that is used to determine the young adult's current and previous involvement in programming related to the transition from school to adult life. The questionnaire includes both open- and closed-ended questions. Closed-ended questions consist of dichotomous and Likert-scale items. Domains surveyed include Employment, Community Living, and Socialization. Parent expectations regarding future employment options and future community living environments were also explored. The following seven items from this survey were used in the present study: (a) school status, (b) paid work experience in school, (c) current paid work, (d) parent work preference for son/daughter, (e) community living instruction at school, (f) involvement in social activities outside of the home; and (g) size of young adult's social network. This instrument is really a guided interview protocol. Although the questions have face validity, this instrument does not contain scales appropriate for reliability analyses.

Quality of life

The QOL-Q was used in the present investigation as an outcome measure. It contains 40 criterion-referenced items that reflect four aspects of a person's quality of life: Satisfaction, Competence/Productivity, Empowerment/Independence, and Social Belonging/Community Integration. Each question is rated on a 1- to 3-point Likert scale. The Quality of Life Index is the sum of all 40 items and can range from 40 (low quality of life) to 120 (high quality of life). This scale can be completed by the young adult him- or herself or, if the young person is not cognitively able to complete the form, a parent or caregiver can do so. In the present study, a parent (or grandparent in 2% of the cases) completed the questionnaire. The alpha for the overall index was .87 and for the subscales, alphas ranged from .64 to .92. Here, we considered the Satisfaction subscale of the QOL-Q a more subjective dimension of quality of life, with the other three subscales encompassing more objective dimensions (Cummins, 2001; Stancliffe, 1999).

Procedure

A small packet containing three of the more sensitive measures (FIQ, QRS-F, and F-COPES) was mailed to the family prior to the interview. These measures were sent ahead of time so that the family could complete the questionnaire privately, without the interviewer or young adult being present. These measures were typically completed a week or so before the interview. When the interviewer arrived for the in-person interview, he or she answered any questions the family had on the measures in the mailed packet.

The remaining measures were administered during an in-person interview at the families' own home. Most interviews were scheduled at a time when the target young adult could be present to facilitate the completion of the Vineland and so that he or she could participate in the process to the greatest extent possible. The primary interviewers for this study were the first and second authors. Interviewers received training in the administration of instruments and in procedures, safety, and etiquette.

The interviewer read the questions for all in-person parent measures and recorded responses by hand; administration of the Transition Experiences Survey and Parent Involvement in Transition Planning were also audiotaped in order to ensure accuracy of the information reported for later coding and to record any qualitative information provided by the parent during the interview. On average, each interview lasted 2 hours, and participants received an honorarium for their involvement.

Analyses

Mean QOL-Q total and subscale scores were examined for the entire sample as well as by young adult school and work status. In addition, stepwise multiple regression was employed to determine which young adult, family, and environmental variables contribute to the quality of life during transition of young adults with moderate or severe mental retardation. Specifically, the total score from the QOL-Q was used as an outcome in regression analyses. Regression analyses were also conducted using the four subscales of the QOL-Q, separately, as outcomes.

The process used in determining variables to enter into each of the predictor analyses was multifaceted and conceptually driven. First, pertinent items on each of the measures were classified into one of three domains: young adult, family, and environment variables. Young adult variables were those items relating to the target individuals: adaptive behavior, maladaptive behavior, age, gender, and ambulation. Family variables pertained to family demographics (family income, maternal education, and socioeconomic status), family adjustment and coping (positive and negative impact of the youth on the family, coping skills), family support (happiness with level of support received from family and friends, number of respite hours received), and involvement in transition planning (parent involvement: behavioral involvement, knowledge of adult services, satisfaction with transition planning). Environment variables were those pertaining to the schools' role in transition planning, the instruction received by the young adults, the young adults' social networks and opportunities for social involvement, and whether the young adults were currently in paid employment.

Correlations were run between variables in each domain and each of the dependent measures. Those items that were significantly correlated with one or more of the outcome measures were entered into regression models. If there were two variables that had a strong intercorrelation, r > .50, the variable with the strongest bivariate correlation with the dependent measure was retained. For the reported regression analyses the Ns are smaller because not all parents completed the packet that was mailed to them.

Results

What does quality of life look like for a sample of young adults with moderate or severe mental retardation during transition?

Table 2 illustrates the mean QOL-Q total and subscale scores for the entire sample. The highest subscale mean was for Satisfaction (M = 23.0, SD = 3.8). This table also displays QOL-Q total and subscale mean scores for those young adults still in school and for those who exited. As shown in the table, those young adults who had exited the school system had significantly higher QOL-Q total scores than did those still in school, F(1, 187) = 9.0, p < .05. They also had significantly higher Empowerment/Independence subscale scores, F(1, 187) = 20.7, p < .001. There were no significant differences between the two school status groups on the other three subscales of the QOL-Q. When the two school status groups were compared on adaptive behavior, no significant group differences were found (in-school and exited Ms = 27.3 and 24.6, respectively).

Table 2

Quality of Life Scores by School Status and Work Outcomes

Quality of Life Scores by School Status and Work Outcomes
Quality of Life Scores by School Status and Work Outcomes

The quality of life scores of the 85 young adults who had exited public school were compared across three work outcomes: (a) not working, (b) working in a sheltered workshop or involved in a day activity program, and (c) holding paid jobs in the community (either with or without support from a job coach). Data presented in Table 2 indicate that the young adults who were working in the community had significantly higher QOL-Q total scores as well as higher scores on the Competence/Productivity, Empowerment/Independence, and Social Belonging/Community Integration subscales. Tukey post hoc analyses indicated that young adults working in the community had higher QOL-Q total scores than did either those not working or involved in sheltered workshops/day activity centers. Other post hoc analyses are reported in Table 2. When the three work groups were compared on adaptive behavior, the community work group was significantly higher with regard to adaptive behavior than either the sheltered work/day activity group or nonworking group, F(2, 83) = 8.9, p < .001. Thus, an analysis of covariance was performed comparing the three work groups on quality of life, using adaptive behavior as a covariate. Although the F values attenuated somewhat, the same significant relationships held (see Table 2).

One important issue involved with using the QOL-Q is the influence of paid employment on the Competence/Productivity subscale. That is, individuals who are in paid employment can receive a maximum score of 30 on this subscale, whereas individuals not in paid employment can receive a maximum score of 14, regardless of their satisfaction with their current employment/activities. The three work groups in the present study were affected differently by this scoring procedure. Individuals in the work group (n = 17) all had a maximum possible score of 30. Those in the not-working group (n = 34) all had maximum possible score of 14, and those in the sheltered work/day activity center group (n = 34) were a mixture of individuals with the low maximum (unpaid day activity users) and the high maximum (paid sheltered workshop employees). As a result, the difference in scores between the three work groups on the Competence/Productivity subscale, as well as the QOL-Q total score, may be a function of the scoring procedure on the Competence/Productivity subscale of the QOL-Q. Therefore, a composite of the three remaining subscales was created so as not to unfairly bias those individuals with little or no paid work experiences. This was also done by Eggleton, Robertson, Ryan, and Kober (1999). As shown in Table 2, the Satisfaction–Empowerment/Independence–Social Belonging/Community Integration composite scores did not significantly differ across the three work groups.

What young adult, family, and environmental variables predict an overall quality of life index for these individuals?

Table 3 displays the correlations of variables that significantly entered into one or more of the stepwise regression models. Variables that significantly predicted the young adults' overall quality of life were the Vineland Adaptive Behavior Composite, the negative impact on parenting composite of the FIQ, parents' knowledge of adult services available in the community, and number of friends the young adult's social network contained. Over 40%, R2 = .44, of the variance in the QOL-Q Index score was explained. Higher functioning young adults with larger social networks who had parents with more knowledge of adult services and less negative young adult impact had higher overall quality of life scores (see Table 4).

Table 3

Correlation Matrix of Young Adult, Family, and Environment Variables That Significantly Entered Into One or More Regression Models

Correlation Matrix of Young Adult, Family, and Environment Variables That Significantly Entered Into One or More Regression Models
Correlation Matrix of Young Adult, Family, and Environment Variables That Significantly Entered Into One or More Regression Models
Table 4

Multiple Stepwise Regression Analyses for Variables Predicting Overall Quality of Life (QOL-Q Index Score) (N = 146)

Multiple Stepwise Regression Analyses for Variables Predicting Overall Quality of Life (QOL-Q Index Score) (N = 146)
Multiple Stepwise Regression Analyses for Variables Predicting Overall Quality of Life (QOL-Q Index Score) (N = 146)

Do different variables predict subjective versus objective dimensions of quality of life for this sample during transition?

Three variables predicted the Satisfaction subscale, a subjective dimension of quality of life, of the QOL-Q. Negative impact, happiness with the amount of family support, and size of young adult's social network best predicted the satisfaction items on the QOL-Q. That is, less negative impact on parenting, greater happiness with family support, and larger young adult social networks predicted higher scores on the Satisfaction subscale of the QOL-Q; however, the model only accounted for 16% of the variance (see Table 4).

Adaptive behavior was the most significant predictor for each of the three remaining subscales of the QOL-Q, classified in this study as objective dimensions of quality of life. As indicated in Table 4, in addition to adaptive behavior, parental knowledge of available adult services, young adult gender, and parental involvement in transition planning entered significantly into the model for the Competence/Productivity subscale, overall R2 = .29. Male young adults with higher adaptive functioning who had parents who were knowledgeable of adult services and involved in their transition planning scored higher on the Competence/Productivity subscale of the QOL-Q. In terms of predicting the Empowerment/Independence subscale, the young adult's adaptive functioning and parental knowledge of adult services significantly accounted for 31% of the variance (see Table 4).

For the Social Belonging/Community Integration subscale, six variables entered into the model: young adult adaptive functioning, the family's capacity to cope, the size of the young adult's social network, satisfaction with help from nonfamily members, young adult's paid work experiences while in school, and negative impact on parenting. Young adults with higher adaptive functioning, larger social networks, paid work experiences in school, and parents with less stress/negative impact and more capacity to cope scored higher on the Social Belonging/Community Integration subscale. The overall model accounted for 43% of the variance (see Table 4).

Discussion

Our intent in the present investigation was to examine quality of life for a sample of young adults with moderate or severe mental retardation during transition. Mean QOL-Q total scores for the entire sample were similar to scores obtained by Schalock and Keith (1993). In the present study, those young adults who had exited the school system had higher QOL-Q total scores than did those still in school, with no significant mean differences in adaptive behavior. The exited young adults also had higher scores on the Empowerment subscale. This may suggest that for individuals who are still in school, versus those who are out of school, environment or family factors play a greater role than participant characteristics. For instance, once young adults have exited the school system and “graduated,” their parents may feel as if they provide them with more choice and independence, or it is possible that nonschool environments afford more choice-making and opportunities for autonomy. We recently corroborated this finding in a qualitative study (McIntyre, Kraemer, & Blacher, 2002), with a subset (N = 30) of the same sample, wherein mothers were asked to define quality of life for their young adult child and to describe the extent that their son or daughter had quality of life. Mothers with young adults who had exited the school system were more likely to incorporate themes of independence in their definitions of quality of life as compared to mothers of those young adults who were still in school.

Young adults who had exited the school system and were in paid work in the community, with or without support, had higher QOL-Q total and subscale scores, with the exception of the Satisfaction subscale, than those working in sheltered workshops/day activity centers and/or those not working at all. This is not surprising because the working group had significantly higher adaptive behavior scores compared to the other two groups, and adaptive behavior was the greatest predictor of the QOL-Q total score and the more objective subscales of the QOL-Q. However, even when adaptive behavior was controlled for in analyses, the paid work group had significantly higher QOL-Q and objective subscale scores, thus indicating that other variables besides adaptive behavior differ between the groups and may affect quality of life outcomes.

When the three work groups were compared on the created Satisfaction–Empowerment/Independence–Social Belonging/Community Integration composite, which did not contain the Competence/Productivity or “work” subscale, there were no differences between the groups. This suggests that work status only affects the “work” aspect of quality of life but has little impact on other dimensions of quality of life, or it may support a scoring anomaly of the Competence/Productivity subscale of the QOL-Q, wherein individuals who do not have a paying job have an artificially low ceiling on the subscale. It could, however, also be related to the overall level of functioning of the present sample; results of Eggleton et al. (1999), whose sample was much higher functioning and older than ours, were different. They found that working individuals had significantly higher QOL-Q SES composite scores than did those not working or in a sheltered workshop. Thus, it will be important for future researchers examining the impact of employment on quality of life to control for personal characteristics, such as adaptive behavior, and use a measure that is equally weighted for a variety of work groups, both paid and unpaid.

Although adaptive behavior was the single largest predictor of the three more objective subscales of the QOL-Q, other family-related variables did enter significantly into the models. Specifically, parent's knowledge of adult services, parent's involvement in transition planning, negative impact, family coping, and satisfaction with the amount of help from nonfamily members significantly entered in one or more of the objective subscale models. This supports the importance of involving families in the transition process (Hanley-Maxwell, Whitney-Thomas, & Pogoloff, 1995; Morningstar et al., 1996). By involving families in planning for their son's or daughter's future and providing supports to the family system during transition, service providers may be able to positively affect quality of life, or at the very least parents' perception of it.

It is not surprising that adaptive behavior of the young adult did not predict the more subjective domain of Satisfaction. Schalock et al. (2000) also found that personal characteristics, such as IQ, age, communication problems, and ambulation, had no direct effect on the Satisfaction subscale of the QOL-Q. Items relating to family stress and social resources were, however, significant indicators of the Satisfaction subscale in the present sample. Life satisfaction in general relates more to personal well-being and supports and, therefore, is more individualized and idiosyncratic (Hanley-Maxwell et al., 1995; Taylor & Bogdan, 1990). Indeed, many investigators have found a low correlation between objective and subjective dimensions of quality of life (Cummins, 2000, 2001; Schalock et al., 2000) and question whether a total quality of life score makes any conceptual sense (Ager & Hatton, 1999; Hatton, 1998; Schalock, in press).

In this study we examined indicators of objective and subjective dimensions of quality of life for a sample of transition-age young adults with severe mental retardation; however, it is also valuable to examine the nature of the critical indicators. Schalock et al. (1994) classified indicators of quality of life as either stable status variables or less stable intervention variables. In the present study, adaptive behavior could be considered a stable status variable (i.e., a variable that is slow to change), with variables such as family coping, parent involvement in transition planning, and knowledge of adult services considered less stable intervention variables. Thus, although adaptive behavior is a strong predictor of individual quality of life, it is not the sole predictor. Indeed, we are encouraged that other family-related variables that are more amenable to intervention (e.g., parent involvement in transition planning) seem to also affect the quality of life of young adults with mental retardation during transition.

There are many difficulties inherent in attempting to measure quality of life in general and with individuals with moderate or severe mental retardation in particular. A primary methodological concern in measuring quality of life in persons with moderate or severe mental retardation is the use of proxy respondents (Schalock et al., 1994; Stancliffe, 1999). In this case, individuals who are unable to respond for themselves have a proxy respond on their behalf. The assumption is that the proxy is knowledgeable about the person and can provide an adequate response on behalf of that individual. Although this requires speculation on the part of the proxy, the literature has provided evidence that the use of proxy respondents is an acceptable alternative (e.g., Schalock & Keith, 1993; Stancliffe, 1999), particularly when all or the majority of the items are objective, as is the case with the QOL-Q. Thus, the use in the present study of mothers and other family members as proxies for the young adults on the QOL-Q is satisfactory. As Stancliffe (2000) noted:

Use of proxies can be justified when the questionnaire used is known to possess empirically well established consumer: proxy agreement. The bulk of evidence about the Quality of Life Questionnaire (QOL-Q) (Schalock & Keith, 1993), indicates that it is such a scale. (p. 90)

One limitation of the present study is the possible confound between family variables and quality of life outcomes, given that family members were the respondents for both family-related measures and the QOL-Q. This could be particularly relevant for the Satisfaction subscale, a more subjective measure of quality of life, where one would have the greatest doubts that proxy reports are valid. It is certainly possible that family members who feel less negative impact of the young adult on the family, more supported by others, knowledgeable of adult services, and involved in transition planning would report that their son's or daughter's quality of life was better. Indeed, it could be that the respondent's view of the family and of his or her offspring's quality of life might be one and the same.

In future work investigators should continue to explore the multiple dimensions of quality of life for young adults with mental retardation during transition. In particular, approaches should be developed that enable individuals with more severe mental retardation to play some role in expressing their views about their personal well-being and life satisfaction. Perhaps the use of open-ended questions utilizing qualitative methodologies will better capture the subjectivity of life satisfaction and allow for more diverse and individualized responses.

NOTE: This paper was supported, in part, by Grant HD121324 (J. Blacher, principal investigator.) from the National Institute of Child Health and Human Development.

References

References
Ager
,
A.
and
C.
Hatton
.
1999
.
Discerning the appropriate role and status of “quality of life” assessment for persons with intellectual disability: A reply to Cummins.
Journal of Applied Research in Intellectual Disabilities
12
:
335
339
.
Baker
,
B. L.
and
J.
Blacher
.
1993
.
Out-of-home placement for children with mental retardation: Dimensions of family involvement.
American Journal on Mental Retardation
98
:
368
377
.
Blacher
,
J.
2001
.
Transition to adulthood: Mental retardation, families, and culture.
American Journal on Mental Retardation
106
:
173
188
.
Blacher
,
J.
and
B. L.
Baker
.
1994
.
Family involvement in residential treatment of children with retardation: Is there evidence of detachment?
Journal of Child Psychology and Psychiatry
35
:
505
520
.
Blacher
,
J.
,
B. L.
Baker
, and
K. A.
Feinfield
.
1999
.
Leaving or launching? Continuing family involvement with children and adolescents in placement.
American Journal on Mental Retardation
104
:
452
465
.
Blacher
,
J.
,
R.
Hanneman
, and
A. M.
Rousey
.
1992
.
Out-of-home placement of children with severe handicaps: A comparison of approaches.
American Journal on Mental Retardation
96
:
607
616
.
Borthwick-Duffy
,
S.
1989
.
Quality of life: The residential environment.
In W. E. Kiernan & R. L. Schalock (Eds.), Economics, industry, and disability: A look ahead (pp. 351–363). Baltimore: Brookes
.
Bruininks
,
R. H.
,
R. W.
Woodcock
,
R. F.
Weatherman
, and
B. K.
Hill
.
1996
.
Scales of Independent Behavior-Revised comprehensive manual.
Itasca, IL: Riverside
.
Cummins
,
R. A.
1997
.
Self-rated quality of life scales for people with an intellectual disability: A review.
Journal of Applied Research in Intellectual Disabilities
10
:
199
216
.
Cummins
,
R. A.
2000
.
Objective and subjective quality of life: An interactive model.
Social Indicators Research
52
:
55
72
.
Cummins
,
R. A.
2001
.
Living with support in the community: Predictors of satisfaction with life.
Mental Retardation and Developmental Disabilities Research Reviews
7
:
99
104
.
Donenberg
,
G.
and
B. L.
Baker
.
1993
.
The impact of young children with externalizing behaviors on their families.
Journal of Abnormal Child Psychology
21
:
179
198
.
Eggleton
,
I.
,
S.
Robertson
,
J.
Ryan
, and
R.
Kober
.
1999
.
The impact of employment on the quality of life of people with an intellectual disability.
Journal of Vocational Rehabilitation
13
:
95
107
.
Felce
,
D.
1997
.
Defining and applying the concept of quality of life.
Journal of Intellectual Disability Research
41
:
126
135
.
Friedrich
,
W. N.
,
M. T.
Greenberg
, and
K. A.
Crnic
.
1983
.
A short-form of the Questionnaire on Resources and Stress.
American Journal of Mental Deficiency
88
:
41
48
.
Goode
,
D. A.
1990
.
Thinking about and discussing quality of life.
In R. L. Schalock & M. J. Begab (Eds.), Quality of life: Perspectives and issues (pp. 41–57). Washington, DC: American Association on Mental Retardation
.
Goode
,
D. A.
and
J.
Hogg
.
1994
.
Towards an understanding of holistic quality of life in people with profound intellectual and multiple disabilities.
In D. A. Goode (Ed.), Quality of life for persons with disabilities: International perspectives and issues (pp. 197–207). Cambridge, MA: Brookline Books
.
Halpern
,
A. S.
1993
.
Quality of life as a conceptual framework for evaluating transition outcomes.
Exceptional Children
59
:
486
498
.
Hanley-Maxwell
,
C.
,
J.
Whitney-Thomas
, and
S. M.
Pogoloff
.
1995
.
The second shock: A qualitative study of parents' perspectives and needs during their child's transition from school to adult life.
Journal of the Association for Persons with Severe Handicaps
20
:
3
15
.
Hatton
,
C.
1998
.
Whose quality of life is it anyway? Some problems with the emerging quality of life consensus.
Mental Retardation
36
:
104
115
.
Hughes
,
C.
and
B.
Hwang
.
1996
.
Attempts to conceptualize and measure quality of life.
In R. L. Schalock (Ed.), Quality of life: Volume 1: Conceptualization and measurement (pp. 51–62). Washington, DC: American Association on Mental Retardation
.
Hughes
,
C.
,
B.
Hwang
,
J.
Kim
, and
L. T.
Eisenman
.
1995
.
Quality of life in applied research: A review and analysis of empirical measures.
American Journal on Mental Retardation
99
:
623
641
.
Kraemer
,
B. R.
and
J.
Blacher
.
2001
.
Transition for young adults with severe mental retardation: School preparation, parent expectations, and family involvement.
Mental Retardation
39
:
423
435
.
McCubbin
,
H. I.
,
D.
Olson
, and
A.
Larsen
.
1981
.
Family Crisis Oriented Personal Scales (F-COPES).
In H. I. McCubbin, A. I. Thompson, & M. A. McCubbin (1996), Family assessment: Resiliency, coping, and adaptation—Interventions for research and practice (pp. 455–507). Madison: University of Wisconsin System
.
McIntyre
,
L. L.
,
B. R.
Kraemer
, and
J.
Blacher
.
2002
.
Quality of life for young adults with severe intellectual disability: Mothers' thoughts and reflections.
Manuscript submitted for publication
.
Morningstar
,
M. E.
,
A. P.
Turnbull
, and
H. R.
Turnbull
III
.
1996
.
What do students with disabilities tell us about the importance of family involvement in the transition from school to adult life?
Exceptional Children
62
:
249
260
.
Rusch
,
F. R.
and
D. M.
Millar
.
1998
.
Emerging transition best practices.
In F. R. Rusch & J. G. Chadsey (Eds.), Beyond high school: Transition from school to work (pp. 36–59). Belmont, CA: Wadsworth
.
Schalock
,
R. L.
1997
.
The conceptualization and measurement of quality of life: Current status and future considerations.
Journal on Developmental Disabilities
5
/
2
:
1
21
.
Schalock
,
R. L.
in press
.
Moving from individual to family quality of life as a research topic.
In A. P. Turnbull, I. Brown, & H. R. Turnbull (Eds.), Family quality of life: An international perspective. Washington, DC: American Association on Mental Retardation
.
Schalock
,
R. L.
,
G. S.
Bonham
, and
C. B.
Marchand
.
2000
.
Consumer based quality of life assessment: A path model of perceived satisfaction.
Evaluation and Program Planning
23
:
77
87
.
Schalock
,
R. L.
and
K. D.
Keith
.
1993
.
Quality of Life Questionnaire manual.
Columbus, OH: IDS
.
Schalock
,
R. L.
,
J. A.
Lemanowicz
,
J. W.
Conroy
, and
C. S.
Feinstein
.
1994
.
A multivariate investigative study of the correlates of quality of life.
Journal on Developmental Disabilities
3
/
2
:
59
73
.
Seltzer
,
M. M.
and
M. W.
Krauss
.
2001
.
Quality of life of adults with mental retardation/developmental disabilities who live with family.
Mental Retardation and Developmental Disabilities Research Reviews
7
:
105
114
.
Sparrow
,
S. S.
,
D. A.
Balla
, and
D. V.
Cicchetti
.
1984
.
Vineland Adaptive Behavior Scales.
Circle Pines, MN: American Guidance Service
.
Stancliffe
,
R. J.
1999
.
Proxy respondents and the reliability of the Quality of Life Questionnaire empowerment factor.
Journal of Intellectual Disability Research
43
:
185
193
.
Stancliffe
,
R. J.
2000
.
Proxy respondents and quality of life.
Evaluation and Program Planning
23
:
89
93
.
Szymanski
,
E. M.
,
D. B.
Hershenson
, and
P. W.
Power
.
1988
.
Enabling the family in supporting transition from school to work.
In P. W. Power, A. Dell Orto, & M. B. Gibbons (Eds.), Family interventions throughout chronic illness and disability (pp. 216–233). New York: Springer
.
Taylor
,
S. J.
and
R.
Bogdan
.
1990
.
Quality of life and the individual's perspective.
In R. L. Schalock & M. J. Begab (Eds.), Quality of life: Perspectives and issues (pp. 27–40). Washington, DC: American Association on Mental Retardation
.
Wehman
,
P.
1996
.
Life beyond the classroom: Transition strategies for young people with disabilities (2nd ed).
Baltimore: Brookes
.

Author notes

Authors:Bonnie R. Kraemer, PhD, Visiting Research Associate (on leave from the University of New Mexico) ( bonniek@unm.edu.), Laura Lee McIntyre, MA, Doctoral Student, and Jan Blacher, PhD, Professor, Graduate School of Education, University of California, Riverside, Riverside, CA 92521-1028. Requests for reprints should be sent to the first or third author