Version classiqueVersion mobile

A Quarter-Century of Normalization and Social Role Valorization

Robert J. Flynn
Raymond Lemay

Part 4: Links Between Normalization, Social Role Valorization, Social Science Theory, and Empirical Research

11. Normalization and residential services: The Vermont studies1

Sara N. Burchard

Texte intégral

  • 1 This research was supported in part by the Vermont State Department of Mental Health (DMH) and the (...)

1Normalization has had an immeasurable impact on human services, education, and the social fabric of North America since its introduction 25 years ago. It has revolutionized thinking about service delivery across the entire spectrum of human services. Normalization has changed how we view people with needs, how we view their place in our society, and how we view the role of human services. Public policy has moved from a position of removing people with mental-health, educational, or social needs in order to “fix” or contain them, to one of viewing the role of public policy and services as bringing supports to them and their families (Smull & Bellamy, 1990) in order to assist them to participate in and benefit from the services and opportunities offered in their communities. Normalization has promoted our most recent civil-rights movement in the USA, culminating in the passage of the Americans With Disabilities Act of 1990. Normalization may prove a major factor in health-care reform as the USA attempts to find more useful and cost-effective ways of meeting the health, mental-health, and supportive-care needs of its citizens.

2The Normalization philosophy of inclusion, participation, and treatment en venue has gained momentum, increasing in scope and affecting persons and services far beyond those persons who are called mentally retarded or developmentally disabled. The philosophy and its effects on services have spread to the “treatment” of children with severe emotional disturbance and persons affected by chronic mental illnesses, permanent brain injury, long-term health problems, aging, or terminal illnesses. The legal and social recognition that “separate is not equal,” and certainly not better, and that educational or other services to people need not be segregated to be individualized, has come a very long way since Brown and the Board of Education in 1959. That people are happier, do better, receive more respect and individualization, and have a better quality of life if they remain out of congregate, institutionalized settings and “special places” is a message that has been heard by many in direct-service and policy positions. However, the reality has not automatically followed the recognition nor even the legislation.

3The ideas, derived from Normalization philosophy, of bringing services to people in their homes, schools, and communities and keeping people at home in their families as much as possible, whether they be medically fragile, developmentally disabled, or otherwise in need of support, have had a far-reaching impact on service philosophy and service provision. However, the impact has not been even in its implementation and acceptance. There exist strong forces in professional practice and training that focus on “specialties,” that locate the “disability” in the person, that treat or fix the person in isolation from his or her life and family. There are also financial constraints and political realities that have prevented the implementation of Normalization practices for many persons.

4While service philosophies and service systems have changed due to Normalization philosophy, the issues of implementation remain. Are our services designed to enhance the inclusion of persons with disabilities in recreation, leisure, work, education, and other services in their communities, supporting them to gain and maintain respect, social supports and friendships, and autonomy and independence, and to have access to lifestyles of their choice, similar to those available to other community members? It was from efforts to examine questions such as these that the Vermont studies grew.


5The program of research in which my colleagues and I have been engaged over the past 15 years at the University of Vermont resulted directly from the impetus provided by Normalization for the development and “humanization” of community-based services for persons with mental retardation. This program of research has been inextricably interwoven with evaluating Normalization practices as the latter have been followed and applied in the State of Vermont. A national leader in promoting and implementing practices based upon Normalization principles, Vermont was recently identified as first in the USA in terms of its rate of inclusion (79.4%) of students with disabilities in regular educational activities (Annual Report to Congress, 1991). Vermont has also been among the most progressive states in terms of its financial support for community services (Braddock &Fujiura, 1991; Braddock, Hemp, Fujiura, Bachelder, & Mitchell, 1989). Vermont closed its only institutional center for persons with mental retardation during the winter of 1994-1995, guaranteeing that public dollars go to support small community-based services.

6The research described in this chapter was undertaken in an effort to support and evaluate the community residential services in Vermont as they were developing. Because the Vermont service system was predicated on the implementation of Normalization principles (Nirje, 1970; Wolfensberger, 1972), an examination of how Normalization principles are actually applied has been an essential part of our research program. Unlike research undertaken to examine the question “Does Normalization work?” (Zigler, Hodapp, & Edison, 1990), our work has asked “How well has the state been accomplishing its social-policy objectives?” In effect, we have been interested primarily in how well Normalization principles have been implemented and only secondarily in the ways in which Normalization implementation has expressed itself in client outcomes. The research began in 1979 with a request from the Mental Retardation Services branch of the Vermont State Department of Mental Health (DMH) to develop staff competencies for persons working in the delivery of the then-new community mental-retardation services. Our program of research has evolved over the past 15 years to include an empirical examination of issues central to themes of Normalization.

7The State of Vermont launched its drive to transform the delivery of mental-retardation services from an institution-based to a community-based system in the mid-1970s. The goals of community services were from the beginning framed from the perspective of the principle of Normalization as articulated by Wolf Wolfensberger (1972). DMH and advocates brought Wolfensberger to Vermont on several occasions, sent key policy-makers and providers to PASS (Wolfensberger & Glenn, 1975) and Normalization workshops, and had PASS evaluations completed for several local services in order to give coherent direction to the development and implementation of the new service system.



8Against this backdrop, members of the Psychology Department at the University of Vermont were asked to develop staff competencies for the newly developing, small community homes (by state mandate, residences were to house no more than six persons), in order to provide local agencies with criteria for hiring and training staff. Although nationally many were developing competencies and training based on expert opinion and academic training programs (Fiorelli & Keating, 1979), a different procedure was employed in Vermont. A field-based methodology was selected to generate and identify essential staff and manager competencies by capitalizing on the information and experience of those most directly involved in the delivery of community services. A large pool of potential competencies was generated, based on the methods of job analysis (delineating of the tasks that staff members performed in the course of their work, how much time these tasks took, and how important they were), critical-incident reporting (describing incidents that made staff feel especially competent or successful in carrying out their jobs, as well as incidents in which they especially felt in need of greater skill), structured interviews with direct-care staff and managers about how to support individuals to achieve maximum independence and integration into community living, and a review of competencies derived from expert opinion. Individuals throughout the state directly involved in providing community services for persons with mental retardation participated in this process.

9The final step in identifying critical competencies was to have a broad sample of state mental-retardation professionals, including academicians, administrators, paraprofessionals, and advocates, use a forced-choice procedure to sort the competency statements into categories (”absolute prerequisite,” “desirable prerequisite,” “needs to be acquired with training,” and “not essential”). This resulted in the identification of a core of 21 essential competencies for staff and managers of community residences for persons with mental retardation (Thousand, Burchard, & Hasazi, 1986).

10There were several instructive and surprising results from this field-based, bottom-up generation of competencies. First, despite our research bias toward identifying skills that could be readily operationalized, the final pool of items contained many statements that could only be characterized as attitudes rather than behavioral skills. Second, those items identified by the vast majority of respondents as absolutely essential for service providers were primarily attitudes rather than skill-based competencies. Third, the most absolutely prerequisite competencies were closely related to Normalization philosophy, namely, high regard and respect for the individuals being served, provision of age-appropriate activities, enhancing consumer status, individualization, and choice. And fourth, the (re)discovery was made, after content analysis, that the core absolutely prerequisite competencies included Carl Rogers’s (1969) famous basic elements essential for developing a successful helping relationship. The resulting competency statements were organized under two general headings based upon content: a set of value-based, humanistic Normalization/Person Orientation Competencies, and a set of Teaching/Technical Skill Competencies. The set of 21 core competencies are presented in Table 11.1. As can be seen on inspection, 18 of the 21 core competencies belong to the value-based Normalization/Person Orientation category.

11While the more commonly used methods of competency generation, namely, expert opinion and job analysis, generated skill and knowledge-based competency statements, the open-ended, field-based methods of interview and critical-incident analysis generated competencies related to interpersonal interaction and values. These were subsequently identified by all levels of informants as the most crucial.


12If the identified attitudes, skills, and knowledge were in fact critical competencies for service providers, then they should be reflected in improved services for consumers. Two studies were conducted to validate the relationship of staff competencies to service outcomes, a concurrent and a predictive validation study.

13In order to determine whether the quality of service recipients’ programs were related to their providers’ competence in Normalization practices and values, interpersonal skills, and teaching and training, it was necessary to identify, define, and develop measures of outcomes that reflected program goals. Since the articulated policy upon which all state programs were founded was implementation of Normalization and the developmental model, it was essential to measure those aspects of the residential programs under study. If programs were implementing Normalization, developmental training, and support in an effective manner, this should be reflected in service recipients’ greater personal independence, community participation, integration, well-being, and satisfaction. Table 11.2 shows the constructs that we undertook to measure and the instruments used to measure them. It also shows the methods used to examine the relationship between staff competence and program quality in the 14 small group residences that existed within the state at that time (Burchard, Pine, Gordon, Joffe, Widrick, & Goy, 1987).

Level I: Absolutely Prerequisite for All Positions (13 Competencies)

TABLE 11.1. CORE COMPETENCIES FOR STAFF AND MANAGERS OF SMALL COMMUNITY RESIDENCESLevel I: Absolutely Prerequisite for All Positions (13 Competencies)

Level II: Desirably Prerequisite for All Positions (8 Competencies)

Level II: Desirably Prerequisite for All Positions (8 Competencies)

14At the time of this study, there were few existing ways of measuring program Normalization that could be employed within our limited means. The resources to provide PASS (Wolfensberger & Glenn, 1975) or PASSING (Wolfensberger & Thomas, 1983) evaluations of all 14 programs were not available. A review of the literature found few published studies that had incorporated measures of program Normalization other than PASS (Eyman, Demaine, & Lei, 1979) or a local adaptation of PASS (Hull & Thompson, 1980, 1981). Other available measures of environmental factors related to Normalization, such as block treatment, lack of individualization, personal choice, autonomy, and physical appearance, were designed to identify institutional practices (Pratt, Luszcz, & Brown, 1980). These measures were insensitive to differences among Vermont programs, which had all been developed on the same model. All were small residences (six or fewer residents), located within residential neighborhoods, with day or work programs for residents and an emphasis on age-appropriate activities and appearances. Therefore, a structured interview was developed for residence staff that indirectly assessed the degree to which the program they directed incorporated important principles of Normalization: individualization, privacy, autonomy, responsibility, and age-appropriate activities, rhythms, and lifestyle (Burchard et al., 1987). Although this instrument did not cover every important aspect of effective Normalization implementation, it did incorporate elements thought to differentiate among the Vermont programs. Other elements, such as age-appropriate dress and the physical appearance of rooms, homes, and the neighborhood, were consciously addressed by all of the programs, which therefore differed little on these aspects.



15To provide additional measures of Normalization practices and to cross-validate the staff Normalization interview, three other kinds of measures were incorporated into the study. Observational measures of staff and resident behavior in the home were used to assess age-appropriate activities, responsibility, and independence. A structured interview with residents about their lifestyle, including self-perceived independence, autonomy, responsibility, community access, privacy, and interpersonal relationships within the residence, was used to evaluate resident satisfaction and Normalization issues (Burchard, Pine, & Gordon, 1990). The final method of evaluating program implementation of Normalization principles was a measure of recent community activities. This measure, based on an interview with staff, asked the number and type of out-of-home activities in which each resident had participated during the previous two weeks. Using a prompt sheet of activities, staff were asked whether residents had gone as a group, singly, or in small numbers, and where the activity had taken place. Scores for programs were based upon the number of activities that had taken place in integrated settings with additional points for nongroup (individualized) activities. Only activities that involved the opportunity for contact or participation with nondisabled peers were scored.

16Measures of developmental-model implementation were based on an examination of participants’ individual program plans (mandated by the state to reflect training and progress in daily and community-living skills) and ratings of program quality made by state reviewers evaluating the programs for quality-assurance purposes. Finally, personal adjustment was evaluated by means of a structured interview with service recipients, in which a simple forced-choice format was used to evaluate self-perceived well-being. This measure was adapted for use with individuals with mental retardation (Seltzer, 1980) from an interview used to evaluate the general psychological well-being of adults from the general population (Dupuy, 1978).

17Group-home manager competence was evaluated in two ways: by means of supervisor ratings of the manager’s performance on each of 53 competencies identified as essential for manager job performance in the previous study (Thousand et al., 1986), and by means of an objective written assessment given to the manager. The written task involved analyzing a setting in which clients received services, designing a program to meet the clients’ needs, and then demonstrating knowledge and skill in defining objectives, creating and collecting data, and behavior management.

18The competency-validation process involved 14 small group homes with 78 residents and 14 managers, the latter also serving as direct-care providers. The results showed that manager competencies were highly and logically related to measures of program quality (Burchard et al., 1987). Variations in managers’ competencies translated into measurably different program outcomes for residents in terms of program Normalization, community integration, developmental programming, resident activities, and resident well-being.

19Managers with greater competency in technical skills had programs with more data-based Individualized Program Plans (IPP) and received higher ratings from state reviewers, who analyzed managers’ written records as an important part of their biannual evaluations. These managers’ programs, however, had lower scores on program Normalization, based on data gathered during the structured interviews, and residents in their programs were observed to engage in less personal and home care. Across all participants, engaging in these age-appropriate adult activities was related to higher self-reported well-being and to being in a program with a higher Normalization score (structured-interview data). Residents with severely handicapping conditions who lived with technically competent managers were observed to engage in more independent and self-initiated behavior. Managers with higher scores on Normalization/person orientation competencies had residents who were more actively involved in accessing community activities in more individual ways, one aspect of Normalization implementation included in our Normalization measures. It should be noted that competence in Normalization/person orientation areas and competence in teaching and technical skills were not necessarily mutually exclusive. In fact, in the subsequent predictive validation study, these two sets of competencies were highly correlated, which was not due merely to resident level of disability.

20As small community programs in Vermont were about to expand rapidly, there was an opportunity to examine competencies in a predictive manner. Local agencies throughout Vermont included competency-based assessment procedures provided by our research group along with their usual procedures in interviewing potential managers. The candidate-assessment procedures included the written evaluation described above, a structured interview with candidates, a competency rating scale given over the phone to a person serving as the candidate’s reference, and a similar competency rating from the candidate’s previous employer. The agencies also used whatever procedures they had in place for candidate selection and made their decisions based upon their own criteria and/or the exigencies of the moment. Approximately 10 months later, measures of manager competence and program quality were assessed in the same way as in the original study, except that no direct observations were made of resident activities in the homes.

21Seventy-two persons applying to become managers were evaluated. Twenty-five managers were hired, of whom 18 subsequently participated in the follow-up evaluation. They were managers and co-managers of small community residences for two to six residents of all levels of disability.

22The results of the predictive study replicated the findings from the concurrent validation study (Burchard, Pine, Widrick, & Creedon, 1985). Managers with higher scores on Normalization competencies (value-based person-orientation and interpersonal skills), regardless of the level of disability of their residents, had programs a year later in which residents were more actively accessing the community. These managers’ programs also received higher program-Normalization scores on the Normalization interview. Managers with higher technical competencies upon hiring, on the other hand, had better data-based individual program plans and higher satisfaction ratings from state reviewers 10 months later. The concurrent and predictive relationships found between the live-in residence-managers’ competencies and program outcomes are shown in Table 11.3.



Note. The predictive correlations were based on data collected 10 months after the managers were hired. More than one correlation in a cell reflects use of several competency measures during candidate assessment. There were 14 small homes in the concurrent study and 16 in the predictive study.
a Directional hypothesis. b Self-care, cooking, and house-care activities were significantly related to program-Normalization scores (r = 0.28) and to well-being (r = 0.42)


23In order to evaluate programs from the consumer’s perspective, data from the concurrent and predictive studies for those individuals living in small group homes who had been able to participate in the resident-satisfaction interview were pooled. There were 57 such individuals living in 12 small community group homes staffed primarily with one or two resident managers. Those individuals living in homes with more frequent opportunities for community-integrating activities had higher total scores on residence satisfaction, another measure of Normalization and covering the issues of autonomy, individualization, privacy, community access, and responsibility. There were 34 group-home residents who were unequivocal about wanting or not wanting to continue living in their residence and who were able to support their preference with a positive or negative reason. Wanting to stay was associated with programs with greater activation (activities at home, r = 0.57), positive social relations with other residents in the home (r = 0.41), and positive relations with the manager (r = 0.44) (Burchard, Pine, & Gordon, 1984). Those individuals who wanted to stay in their current residence were those whose managers were evaluated as having greater competence (r = 0.57), both technical and Normalization-related, and whose programs provided greater opportunities for individualized participation in community-based activities (r = 0.64). A regression analysis showed that while residents’ personal characteristics were unrelated to wanting to stay in their respective group homes, manager Normalization competence explained 42% of the variance in the desire to stay, to the exclusion of other variables (Burchard et al., 1990).


24This series of studies made it very clear that staff and manager competencies, particularly sensitivity to resident needs and to Normalization issues, had important effects on program quality. Staff competence in Normalization values, attitudes, and practices was reflected in the promotion of Normalization goals and in clients’ activities and satisfaction with their living situation. These studies also showed that community access and participation, important corollaries of Normalization, are valued by residents and contribute to their satisfaction with their living environment.


25As service reform gained momentum throughout the 1970s and 1980s, a wide array of community mental-retardation services grew up (Bruininks, Rotegard, Lakin, & Hill, 1987). Most service systems were based upon the philosophy of Normalization and the developmental model. Despite disagreements about definitions and applications of Normalization philosophy (Zigler et al., 1990; Wolfensberger, 1980), the goal of community services was and is to promote independent functioning and a normalized lifestyle: what, where, how, and with whom people with disabilities spend their time, including physical and social integration and opportunities to engage in meaningful work for meaningful wages, alongside and in the same manner as persons of similar age in the broader community.

26A key question was (and still is) the extent to which the array of community living arrangements and programs met the goals of the new social policy. That is, to what extent did these services provide opportunities to live, work, and recreate, and to exercise choice, independence, and individuality in the same manner as and alongside same-aged peers? Had the community programs stemming from the impetus and ideas of Normalization philosophy possibly created new forms of isolation and segregation? To what extent had the barriers of exclusion and differentness actually been broken down? And to what extent were the recipients of the new community supports being served to their satisfaction?

27Based upon the success of our earlier studies and the obvious desirability of examining issues of program quality from a perspective that includes the individuals served, we proposed to conduct a longitudinal study of community adjustment among adults with mental retardation in Vermont. The types of residential settings selected were prevalent options across the nation for persons with mental retardation: small group homes (GHs); semi-independent, supervised apartments (SAs); and natural families (FHs). These types of settings differed widely from one another in terms of organizational structure and amount of support afforded, providing considerable variability in aspects of community living that are important elements of Normalization. The degree of personal independence, autonomy, and choice thus promised to vary across types of setting, as did the kind, degree, intensity, and duration of social support and the availability of peers for social activities and contact. The inclusion of persons living with their natural families furnished a useful comparison group for evaluating the quality of life and other program outcomes of individuals living in residences provided by the service system. In addition, a longitudinal study of community adjustment would provide an opportunity to elaborate our measures of those elements of Normalization implementation that would be most apt to differ among residential settings and provide an opportunity to examine the relationship of Normalization constructs to positive outcomes from the resident’s perspective. It would also be possible to examine which settings and characteristics of settings best promoted Normalization outcomes.

28Community adjustment is a very complex, multidimensional construct and requires the inclusion of a broad range of variables if researchers are to evaluate “how well people are doing” in community programs supposedly designed to promote their meaningful participation in their communities (Emerson, 1985). Our study proposed to examine a broad range of personal, environmental, and psychosocial variables, over a 3-year period, among persons living in these three types of residences. Its purpose was to evaluate how well Vermont programs were meeting the goals of the service system, from the perspective of the persons served as well as from that of the service system.

29Natural-family homes were added since they were and continue to be the most frequent “placement” for individuals with disabilities. When examining lifestyle Normalization and psychosocial variables such as adjustment, friendships, social-support networks, and continuity of relationships, this group constituted a reasonable comparison group.


30Participants for the longitudinal study were recruited on a statewide basis from provider agencies and advocacy organizations. Persons between the ages of 23 and 55 who had lived in their current residence for at least 8 months and would be able to participate in a simple interview format were recruited. There were 157 such individuals identified throughout the state: 57 in 20 group homes, 47 in 35 supervised apartments, and 52 living with their families. Of these, 133 agreed to participate. The participants included almost the entire population of persons in group homes and supervised apartments in Vermont when the research began. There was no way of determining how many persons were living with their natural families. Instead, such persons were recruited in proportion to the number of persons residing in group homes and apartments in their respective regions. An examination of family characteristics showed that their educational status and financial situations were similar to that of the range of households in Vermont.

31All agency-sponsored residential programs were under state mandate and review to provide services based upon the philosophy of Normalization and the developmental model (Wolfensberger, 1972). Persons living in agency-operated settings were required to have suitable out-of-residence daytime activities lasting at least 4 to 6 hours per day, to be in a “homelike” milieu housing no more than six persons with disabilities, and to have a written plan for receiving services and developing skills. Within each geographical region, supervised apartments and group homes were administered by the same agency, which, in many cases, was also responsible for providing residents’ day, work-activity, or community-employment program. Persons living with their family members and receiving services (usually day or work programs) also received them from the agencies in question. Hence, the philosophy of service provision was similar across all areas of the state and types of residential settings. Group homes (GHs) served from four to six persons, providing 24-hour supervision and training with one or two staff persons. Supervised apartments (SAs) served one or two (in one case, three) persons, providing supervision and training as needed by one or two staff persons but with no on-site, live-in supervision. Family homes (FHs) were not regulated by the state, and careproviders did not necessarily subscribe to any particular philosophy of service nor did they routinely provide training (Burchard, Hasazi, Gordon, & Yoe, 1991).


32The model for examining community adjustment and integration is shown in Table 11.4. As predictor variables, the model includes personal characteristics (age, gender, level of disability, length of prior institutionalization, social integration and environmental characteristics (residence type, Normalization of the physical environment, careprovider competencies or attitudes). Criterion variables included Normalization-related outcomes (lifestyle Normalization, type and extent of work, physical integration, social integration, and independent performance of daily and community-living skills) and personal-adjustment outcomes (behavior adjustment, type and extent of social-support network, satisfaction with residence, work, social support, and well-being). Information for the predictor and Normalization outcome variables was obtained by personal structured interview with careproviders and record reviews while all personal adjustment outcomes except behavior ratings were obtained by private interview with the participant. Measures were obtained three times, at approximately annual intervals.

33Environmental Normalization was assessed with a 38-item rating sheet completed by researchers based upon personal observation of the residence and neighborhood to assess proximity to services and Normalization factors related to the physical location and exterior and interior appearance. Careprovider competencies were measured by means of supervisor ratings. Because parents had no supervisors to rate their competencies, a 29-item self-rating scale of Normalization attitudes was constructed to reflect their attitudes and practices toward promoting independent functioning, personal responsibility, community integration, and age-appropriate activities of their adult family member with disabilities.




Personal characteristics

Lifestyle-Normalization indicators

Level of disability
Length of prior institutionalization
Social integration

Performance of adaptive behavior
Residence-lifestyle Normalization
Community integration:
-Physical integration
-Community employment

Environmental characteristics

Personal adjustment indicators

Normalization of environment
Environmental opportunity
Careprovider competency
Parent Normalization attitudes

Residence satisfaction
Work satisfaction
Social support networks
Personal well-being
Severity of problem behaviors

34The Community Adjustment Scale (Seltzer & Seltzer, 1976), completed by caretakers, was used to assess performance of independent behavior (129 yes/no items) and environmental opportunity (45 yes/no items). Lifestyle Normalization was assessed with the 34-item structured careprovider interview tapping residents’ age-appropriate activities, responsibilities, autonomy, and rhythms that had been used in our earlier studies (Burchard et al., 1990). Measures of integration were based upon the number, type, social context, and source of initiation (structured by staff versus self-or peer-initiated) of out-of-residence, nonwork activities in which the participant had engaged during the previous 2 weeks, as enumerated by the careprovider.

35Physical integration was operationalized as the frequency of activities that took place in the community where contact with other community members was possible and probable. Social integration was operationalized as the frequency of community activities carried out in the company of a nondisabled peer. Community employment was defined as paid work done in a community-integrated setting 20 to 40 hours per week (Burchard et al., 1991).

36Measures of the personal-adjustment criterion variables included a 10-item careprovider rating of the frequency and severity of problem behavior, a 22-item Residence-Satisfaction Interview used previously to evaluate the participant’s view of the independence, autonomy, responsibility, activation, community access, and personal relations in his or her residence, and an 18-item Personal Well-Being scale (Burchard et al„ 1990). Work satisfaction was assessed with a 6-item forced-choice scale. The constituents of participants’ social networks were generated from an interview adapted from Weinberg (1984). The persons from whom study participants received instrumental, emotional, and social support were enumerated, and participants’ satisfaction with the frequency of contact and support provided by each network member was evaluated (Burchard, Rosen, Gordon, Yoe, Hasazi, & Simoneau, 1992).

37Personal characteristics thought to affect the criterion measures were used as covariates: age, gender, and level of cognitive challenge. Previous institutionalization had no relationship to the criterion variables. The study sought to determine the extent to which community programs were meeting the social-policy goals of Normalization and habilitation and to assess the lifestyle satisfaction and well-being of the service recipients, from their self-reports. Data were collected from participants and their careproviders by trained graduate and postgraduate research assistants in annual interviews over a 3-year period.

38The variables providing information on how well and to what extent Normalization practices were being incorporated into services were environmental Normalization: location and internal and external appearance of the residence; lifestyle Normalization: individualization, autonomy, personal responsibilities, choice, age-appropriate activities, and daily and seasonal rhythms; community work; physical and social integration: the number, type, social context, and initiation of out-of-home nonwork activities; and residence-lifestyle satisfaction: personal report concerning Normalization of lifestyle and satisfaction. (For more detailed descriptions of the nature, reliability, and validity of the measures, see Burchard et al., 1990, 1991, 1992. The data were analyzed according to a variety of procedures, including ANOVAs, ANCOVAs, correlations, and regressions.)


39The initial status of the study participants (38 in FHs, 54 in GHs, and 41 in SAs) was examined with regard to their lifestyle Normalization and personal adjustment. There were no differences in any outcomes related to prior institutionalization. However, because there were proportionately more persons with moderate retardation living in group homes than with their families or in apartments, level of mental retardation was included as a covariate in all analyses to statistically control for differences in disability.

40Comparisons of program quality across the three different settings (see Table 11.5) showed that group home residents (GHs) were most like family home residents (FHs) on the majority of lifestyle-Normalization indicators. Apartment residents (SAs) experienced greater independence, residence-lifestyle Normalization, and physical and social integration than did persons in the other settings (Burchard et al., 1991). Although it may seem odd that persons living in their own family’s home received lower scores on residence Normalization than persons in agency-operated settings, it must be remembered that these were adults living with fathers, mothers, or relatives who assumed the primary adult roles within the home. So, in general, FH residents experienced less autonomy, choice, and independence and fewer age-appropriate activities than did those in the agency-run homes. It should also be noted that although persons in apartments experienced more social integration, on average, than did those in group homes, the rate of activities over a 2-week period outside the home with nondisabled peers (who were neither staff nor family members) was extremely low. For most persons, there were no such activities (Burchard et al., 1992), the average rate of which ranged between less than once a week (GHs) and three times a week (SAs). (See Tables 11.6 and 11.7).



Note. Physical integration = average weekly activities in the community. Social integration = average weekly community activities with nonhandicapped companions. Careproviders could not report activities (because the latter were too numerous) for 10 persons in apartments. Standard deviations are in parentheses. Community employment = part-time or full-time employment for wages in a nonsegregated setting.
abc Superscripts denote significant differences between group means. d Performance, lifestyle-Normalization, and social-integration measures were correlated with level of retardation, respectively, r =-.45,-.26, and-.30.
*p <.05. **p <.01. ***p <.001

41While persons living in SAs were leading the most normalized lifestyles of the three groups, they were most like persons living in their own homes with their own families (FHs) on the self-reported personal-adjustment indicators (Table 11.6). They reported greater residence satisfaction and well-being and rated their residence higher than did persons living in GHs, although careprovider ratings of behavior adjustment did not differ between the two groups.

42Based upon comparisons of self-reported social networks (Table 11.7), participants were similar across settings in viewing network members as primarily supportive (87% satisfied), in having few (and primarily no) reciprocal relationships, and in identifying only about one relationship in which they saw themselves as a helper vis-à-vis a network member. Group-home residents had the largest networks and the most peers in their networks, due to their group-living situation. Persons in family homes had the fewest peer-friends (an average of 2), fewer of them dated (24%), and they had the fewest number of staff members in their networks (an average of 1). The persons most frequently cited as a source of support by participants were kin of those in family homes, and staff and peers about equally by those living in agency-run settings (Burchard et al., 1992).


43These results showed quite consistent and significant differences with regard to the social policy goals of independent functioning, Normalization of lifestyle, and community integration between the three settings. Individuals who lived in SAs experienced the greatest residence-lifestyle Normalization, community access and integration, and independent performance of skills. Although persons in FHs were no different with regard to ability, their independent skill performance was similar to that of those in GHs while their residence-lifestyle Normalization was even less. These differences remain after using covariance techniques to control for ability differences and are systematically residence related. The organizational structures of the three settings, amount of supervision, number of coresidents, and role and relationship of supervising adult all contributed to the differential residence-related outcomes found here (Burchard et al., 1991).

44There were also significant between-setting differences in social characteristics and personal adjustment. On the outcome measures of adjustment and satisfaction, family-home residents had the highest average scores. Here, however, persons in SAs were more similar to persons in FHs than to those in GHs. It appears that group-living settings involving multiple housemates not of one’s own choosing and live-in staff (i.e., group homes) entail less autonomy, choice, community integration, and personal satisfaction, and no greater social support, than other kinds of settings.


45To identify factors, beyond organizational features of the three different types of settings, that might be related to greater attainment of the program goals of Normalization and independent functioning, regression analyses were conducted. To be conservative, variables known to have a potentially considerable influence on the criterion, such as level of intellectual challenge, were entered first as a block for purposes of statistical control (Seltzer, 1986). Then, in order to determine which of a set of logically related predictors accounted for a significant increment in the variance of the dependent variable, the set of variables in question was entered using a backward-regression procedure. This is an exploratory procedure that eliminates variables that have only a negligible influence on the criterion. For all subjects combined, and after personal competence had been entered (25% of the variance), location, residence-lifestyle Normalization, and opportunity (having environmental opportunities and the autonomy to choose to engage in the activities available) together explained an additional 44% of the variance in independent functioning (in which 69% was accounted for in all). These factors of location, residence-lifestyle Normalization, and opportunity are all elements of Normalization as articulated by Wolfensberger (1972).



Note. The first five variables are self-reports, the last two are informant reports based on careprovider report. Standard deviations are in parentheses.
abc Superscripts denote significant differences between group means.
*p <.05. **p <.01. ***p <.001

46In a regression model predicting residence-lifestyle Normalization from environmental factors, level of disability and place of residence were entered as a block first. Only residence type was a significant predictor of residence-lifestyle Normalization, accounting for 31% of the variance. When additional environmental variables were entered (backward), a combination of community employment, location, and opportunity accounted for an additional 35% of the variance. Where a person lived, both organizationally (residence type) and physically (location, access), opportunities provided or permitted by supervisors and other influential individuals in the person’s life, and the opportunity to work were thus crucial for predicting residence-lifestyle Normalization. Gender, level of disability (for participants among whom there were none with significant physical or profoundly handicapping conditions), and performance of adaptive behaviors, on the other hand, were not.

47Opportunity was the most consistent predictor of residence-lifestyle Normalization. Physical integration could be predicted only for group homes, with community location, staff competence, and residential lifestyle Normalization accounting for 24% of the variance. Scores on social integration (activities outside the home in the company of at least one nondisabled friend) were so low for persons in GHs and FHs that there were no predictors. For persons in apartments, social integration was predicted by residence-lifestyle Normalization and a location accessible to downtown and community sites (Burchard et al., 1985).



Note. * Denotes a reliable difference between groups by statistical test; ns denotes no difference between groups (Burchard et al., 1992)



Note, Independent t tests were used to analyze group differences; ns denotes no difference between groups.
a Mean number of activities with handicapped peers = 3.70; mean number of activities with nonhandicapped peers = 0.41. b Approximately 13% of total activities were supervised by staff (Rosen & Burchard, 1990)


48Although the inclusion of participants living with their families provided an interesting comparison group for the examination of integration, Normalization, and adjustment issues, a normative comparison group was lacking. Given that the purpose of Normalization is to include persons with disabilities in community life, reduce differences, increase personal value, and promote lifestyles similar to those of nondisabled peers, it seemed appropriate to examine the lives of such community peers. A priori, there is no clear baseline against which to say that rates of physical or social integration, the types of and satisfaction with social networks, and so forth are similar to or different from those of community persons. How often, for example, do most people frequent banks, go shopping, or engage in social activities with friends? Some may work to avoid shopping trips or visits with neighbors, cherishing their time at home alone and undisturbed. Few studies have examined these issues.

49Julie Rosen (Rosen & Burchard, 1990) sought to answer some of these questions by examining the lifestyle, activities, and satisfaction of a normative group of adults and comparing them with a group of persons with disabilities who were living in supervised apartments. Because few adults normatively live in group-home kinds of settings (with the exception of college students and communes) and few adults live at home with their parents, Rosen chose to study a category of adults with disabilities for whom a normative comparison group could be constructed: persons living independently in the community, either alone or, if with another adult, without extensive supervision.

50Establishing a reasonable comparison group was not easy, because the modal living situation for adults is to live with a spouse, children, or significant other. This was the clearest lifestyle difference found among the participants in the longitudinal study: They were single, and none were living with a significant other. Nevertheless, by soliciting participants from among human-service workers, a normative comparison group was constituted, composed of 27 single adults between the ages of 23 and 35 who had no live-in significant other or children. The 27 comparison-group members were matched on the variables of age, gender, and community size with 27 persons living in SAs.

51The results of this comparison were quite enlightening (see Table 11.8). There were no differences in the rate or type of activities in which the two groups engaged over a 2-week period. The rate (9/week), location, purposes, and social milieux of the activities were similar. The only distinguishing characteristic lay in the friendships enjoyed by members of each group. For members of the group from SAs, three-fourths of their activities with friends were with individuals who also had disabilities. This was not the case for the normative comparison group. Also, some activities of the SA residents were done in the company of staff, whereas the normative sample had no comparable relationship in their networks.

52An examination of social-support networks revealed differences in the number of persons named (size) and in the types of relationships with network members. The comparison group named twice as many persons in their network. The proportion of network members who were relatives was similar in the two groups (less than 20% kin). However, while the remaining 80% of network members were friends for the comparison group, only 40% were friends for the individuals living in SAs. The remaining 40% of their network members were staff.

53There were no between-group differences, however, in the level of perceived support (high in both groups) or in the desire to increase contact with specific network members: Members of each group, on the average, wanted to see about 40% of their network more often. They felt the frequency of contact with the other 60% was sufficient. Lifestyle satisfaction and personal well-being were also similar and high for the two groups.

54This study showed that the rate of community access and social participation with friends, lifestyle satisfaction, well-being, and satisfaction with social support experienced by a group of young adults with mental retardation living in SAs were entirely similar to those of a comparison group of single young adults living in similar communities. The main difference found was that the “friends” of the individuals living in the SAs included staff (40%) and other individuals with disabilities (35%-40%). If social integration is defined as participating with nondisabled peer companions, then these results indicate that social integration was not being achieved, even by those individuals who were living most independently and with the greatest lifestyle Normalization in their communities. On the other hand, the self-reported satisfaction and well-being of this group were quite high and not different from those reported by the normative comparison group.


55The participants in the longitudinal study were selected to be as similar as possible on major personal characteristics that could significantly affect the relationship of environmental factors to program and personal outcomes. Because it is seldom possible to assign persons randomly to lifestyles and settings (Butterfield, 1987), participant selection and statistical procedures were used to control for confounding due to any systematic differences in cognitive level or behavioral challenges that may have been associated with type of residential setting.



Supervised apartment (n = 27)

Normative comparison (n = 27)


1 (3.7%)



14 (51.9%)


Handicapped peer

10 (37.0%)


Nonhandicapped peer

1 (3.7%)

27 (100%)


1 (3.7%)


Best friend

9 (33.3%)

19 (70.4%)*

*p <.05.

56The purpose of the longitudinal study was to examine whether individuals living in community settings were moving over time toward greater personal adjustment and lifestyle Normalization, or, on the contrary, toward more isolation in their community programs (as some had predicted). We also sought to investigate whether the environmental factors continued to be related to program outcomes and personal adjustment, as they had been in the cross-sectional analyses during the first year (Gordon et al., 1992).


57Throughout the 3-year study, persons living in supervised apartments continued to experience lifestyles that were closer than those of other participants to the program goals of Normalization: autonomy and independence in daily community living, and working, recreating, and living alongside and in the same manner as their community peers. These differences were found after differences due to cognitive functioning had been removed statistically. There was also considerable stability in indicators over time. Only residence-lifestyle Normalization and social integration changed. The former increased consistently over time, whereas the latter increased slightly and then returned to approximately the baseline rate in the third year. There was no indication within any setting that individuals’ lifestyle Normalization in the broad sense, including all of the variables, was deteriorating. If anything, it was stable or increasing (see Table 11.10). Examination of the personal adjustment and satisfaction of participants showed similar patterns (see Table 11.11). The adjustment and satisfaction of individuals within each of the settings remained fairly constant or else improved over time. Again, persons living in SAs were most similar to those living with their families on measures of adjustment and satisfaction and had more favorable scores on these dimensions than those living in GHs. There was more change over time on these variables than there was on Normalization indicators, and change was in a positive direction. Independent ratings of behavior adjustment showed that persons living with their families had the fewest identified problems and the GH residents the most. Over time, adjustment ratings improved in all settings, improving the most for persons in group homes. Residence satisfaction and personal well-being also increased over time. Work satisfaction was high, stable, and did not differ between settings.



Note. Results of 3 x 3 analysis of variance procedure across 3 years and three residence types, with level of mental retardation (LMR) used as a covariate to control for group differences on this variable. ns = no significant difference.
a Performance-of-adaptive-behavior measure was shortened in Year 3; scores were thus on a new scale and not directly comparable to scores in previous years. b Average number of weekly activities in the community. c Average number of weekly community activities with nonhandicapped peers. Careproviders could not report integration scores for 10 persons in supervised apartments because their activities were too numerous. d Community employment = paid work in nonsegregated settings on a part-time or full-time basis. These comparisons were made by chi-square analysis. *p <.05. **p <.01. ***p <.001



Note. Results of 3 x 3 analysis of variance procedures across 3 years and three residence types with level of mental retardation (LMR) used as a covariate to control for group differences on this variable. ns = no significant difference (Gordon et al., 1992).
a Range of rating of severity of maladaptive behaviors = 0-15. b Means with different superscripts (i.e., x,y,z) across residence types within each year are significantly different from one another, which creates the significant interaction.
*p <.05. **p <.01. ***p <.001

58Examination of environmental variables other than residence type, which in and of itself encompassed major organizational differences, showed some interesting patterns. Measures based upon Normalization principles operationalized the constructs of physical setting (condition, appearance, and setting of the residence); location of the residence (maximizing accessibility to community resources); provision of environmental opportunity (availability and “permission”); training of residents provided by staff; and staff competencies or, in the case of the family-home participants, parent Normalization attitudes. Scores on these measures are shown in Table 11.12.

59Supervised apartments, again, tended to receive the most favorable scores, except on environmental Normalization. Ratings of the apartments as homelike, having age-appropriate furnishings, being in a residential neighborhood, and so on, although generally high, were lower on the average than those of the other types of setting. This discrepancy may have been due to the relatively greater autonomy of the apartment residents to select, buy for, and maintain their residence. The furnishings were less middle-class and more impoverished than those in the corresponding family and group homes. This was almost surely affected by their financial status as well. Supervised apartments were not publicly financed, unlike the GHs, and were sometimes located in downtown areas rather than in residential neighborhoods. For these reasons, SAs received lower ratings on environmental Normalization. However, they received higher scores on other aspects of Normalization, and their residents had the highest levels of personal satisfaction with their lifestyles and autonomy.

60Measures of location favoring access to services and activities showed that SAs were the most favorably located and that their residents had the highest level of environmental opportunity (availability of resources and permission to use them). While persons in SAs began with almost equivalent training from staff, over time the training they received decreased significantly compared to that provided to persons living in group homes. At the same time, the ratings of staff competence of persons associated with SAs increased over time.

61These findings are not in the least anomalous. They clearly indicate that from the perspective of both staff and residents, the most interpersonally intensive and most closely supervised and monitored living settings (GHs) were the least personally desirable living (and for staff, working) environments. Good and more experienced staff appeared to be rewarded by becoming SA supervisors rather than continuing as live-in or shift GH staff. It is also notable that although GH residents continued to receive training from staff throughout the 3-year period, this was not reflected in increases in independent functioning, greater community access, or increased social integration. It thus appears that the necessity and opportunity afforded by more independent living, with staff support as needed, provided a greater impetus to skill acquisition than did training.

62Community integration is also affected by the constitution, continuity, and stability of an individual’s social network and support. Social networks, assessed on the basis of participants’ self-reports, showed surprising stability over the 3-year period. Constitution of networks (size and types of relationships reported) and degree of satisfaction remained stable over the 3-year period, with two exceptions. The number of nondisabled peer friends, although very few in number, did increase over the 3 years, except for individuals living in GH settings (see Table 11.13). In addition, the number of reciprocal relationships (again mostly none and averaging less than one) did increase slightly over the 3 years. Persons in FHs and SAs reported the largest number of nondisabled peers in their social networks (frequently none, one, or two), and persons in SAs and GHs reported the most friends (almost always other persons with disabilities) in their networks. Over 50% of the persons living in SAs and GHs reported having girl or boy friends, while relatively few of those living at home reported such relationships (Table 11.13). These results show that active inclusion in friendship networks was minimal, beyond family members, housemates, or staff.

63Social integration was not a fact of life for persons with mental retardation in any of the settings, including family homes. Many providers and others have reported great instability of community living by virtue of changes in staff, housemates, and residences (Bradley & Allard, 1992). Although there were staff and coresident changes for participants, particularly for those in GHs, this was not reflected by decreases in adjustment or well-being over time. These discontinuities, examined within participants’ social networks, showed considerable stability as well as instability. With regards to network constituents, 30%-45% were stable over 3 years (i.e., the individuals identified were the same). Networks were most stable for persons living with their families (see Table 11.14).



Note. Results are based on a 3 x 3 analysis of variance with level of mental retardation (LMR) used as a covariate to control for group differences on this variable. ns = no significant difference.
a Ratings of the physical appearance of the home, 0 (low) to 3 (high). b The number out of 20 generic services within walking distance of the home. c Careprovider rating of availability of and access to performance-of-adaptive-behavior items. Abbreviation of the scale in Year 3 invalidates direct comparison with previous years. d Percent of failed performance items for which home training is provided. e Different superscripts (i.e., x,y) indicate significantly different means by year, which explains the Time by ResType interaction. f Ratings of 40 competency statements by direct supervisor on scale from 1 (low) to 5 (high).
*p <.05. **p <.01. ***p <.001



Note. ns = no significant difference a These results are based upon a 3 x 3 analysis of variance of Time and Residence Type, with level of mental retardation (LMR) used as a covariate. b Few persons in group homes or apartments named kin in their social networks. c Numbers in parentheses indicate the number contributing to each score for each measure. d Only 7 persons in family homes named staff in their networks. e Only '20 persons named a nonhandicapped peer in their social network. f These data are based upon careprovider report.
(*)p <.10. *p <.05. **p <.01. ***p <.001



Note. hc = handicapped; nhc = nonhandicapped

64Staff were most stable for persons in SAs, and peers with disabilities were most stable for those living in GHs. The lack of true social integration with respect to establishing friendships with nondisabled peers was again evident in these results. Such relationships were rarely reported and, even when they existed, were very unstable (see Tables 11.14 and 11.15; also, Burchard & Hutchins-Fuhr, 1990).


Network characteristics

Group home (n = 38)

Supervised apartment (n = 37)

Family home (n = 33)

Any network members




Kin members




Staff members




Friends (hc)




Friends (nhc)




Note. hc = handicapped; nhc = nonhandicapped


65An analysis was conducted to investigate how objective environmental predictors affected three subjective, self-reported variables: personal satisfaction with living environment, satisfaction with work or day program, and personal well-being (Carpenter & Burchard, in preparation). Family-home data were examined separately, as the attitudinal measures of careprovider practices obtained in this group were not comparable to the competency measures obtained in the other groups. Hierarchical regression analysis was used to examine the relationship, among the 70 persons living in SAs or GHs, of four environmental predictors {type of residence, living-environment Normalization, frequency of family contact, and type of work, whether paid, community-integrated or segregated) to the criterion variables residence satisfaction and personal well-being. For the 32 persons living with their families, the environmental predictors for residence satisfaction were Normalization of the living environment and parent Normalization attitudes. Several other Normalization predictors were excluded; physical location and appearance were not related to any self-reported measures of personal adjustment, and staff-competency measures (although significantly related to satisfaction outcomes in concurrent correlational analyses) had been too inconsistently reported to avoid the loss of many subjects. To examine satisfaction with work, all participants were combined, with type of work used to predict work satisfaction. Because of the possibility that where a person lived was in part determined by personal characteristics, age, gender, behavioral problems during Year 1, and level of retardation were always entered first in the regressions. Because residence type was consistently related to many outcome measures, it was always entered next. Also, in the cross-time regressions, prior significant predictors were always entered before current-year predictors. These analyses were thus very conservative in assessing the contribution of factors such as residence-lifestyle Normalization, parent Normalization attitudes, and type of work on personal-adjustment outcomes.

66Normalization factors were significant predictors of personal lifestyle satisfaction for persons living in GHs and SAs, even after the prior entry of all other predictors. For persons in FHs, Normalization measures were significant predictors of both residence-lifestyle satisfaction and personal well-being. Work satisfaction for all participants was consistently related to being engaged in paid, integrated, community work rather than segregated day or work programs, a highly important aspect of a normalized lifestyle.


67There was only one personal characteristic that was consistently predictive of satisfaction measures, namely, behavioral adjustment, as rated by a careprovider. For persons in GHs and SAs, behavior ratings were significant concurrent predictors of residence satisfaction, well-being, and work satisfaction; for persons in FHs, of residence satisfaction and work satisfaction.

68Normalization factors were also predictive of satisfaction and well-being. For persons in GHs and SAs, type of residence (SAs) was always predictive of residence satisfaction (accounting for 8% of the variance). Beyond what behavioral adjustment and residence setting explained, however, an additional significant amount of variance was explained by living-environment Normalization (13%). The only environmental variable that consistently predicted personal well-being was the frequency of family contact.

69For persons living with their families, an environmental factor that predicted residence-lifestyle satisfaction, in addition to behavior ratings (10%-11% of the variance), was parental Normalization attitudes (11%-26%). For these individuals, the only factor, personal or environmental, that predicted well-being was residence-lifestyle Normalization. This Normalization measure had consistent predictive relationships with perceived personal well-being (14%, 29%, 35% of the variance, in Years 1, 2, and 3, respectively). Work satisfaction was predicted by behavior adjustment and type of work, an aspect of Normalization of lifestyle (6%-13%).


70Whether measured directly by the lifestyle Normalization interview, assessed by parent attitudes, or evaluated indirectly by organizational characteristics, including personal control and autonomy of the residence setting (residence type), Normalization was consistently related to the outcomes of self-reported satisfaction and adjustment. This underlines the importance of implementing Normalization principles for the well-being and personal satisfaction of adults with mental retardation. Beyond the inherent significance and credibility of Normalization principles, in light of fundamental considerations of equity, human and civil rights, personal respect and dignity, and personal protection by way of an enhanced image and reduced perception of differentness, these results also indicate that the promotion of normalized lifestyles is perceived as desirable by service recipients themselves. The importance of and justification for pursuing the values derived from Normalization philosophy thus find added validation from the recipients of the ideology in practice.


71What do the results of this series of studies tell us about Normalization philosophy, the success of implementation of Normalization ideas, and ways of better serving individuals in community living? First, we found that staff Normalization competence and lifestyle Normalization are important to consumers, providing more opportunities for exercising choice, having access to the community, initiating activities with friends, and experiencing individualization and independence in daily living. These differences were not simply a function of variations in cognitive, behavioral, or residential-setting factors. Lifestyle Normalization and integration opportunities were, in turn, related to greater self-reported satisfaction with the living environment and, for those persons living with their families, to enhanced well-being.

72Second, these studies unequivocally show that living environments, such as supervised apartments, that provide greater opportunities, autonomy, and independence, fewer housemates, and less restrictive supervision and control, achieve a closer approximation than other types of settings to the ideals of Normalization philosophy and the goals of social policy. By extrapolation from research reported in the literature, they surely surpass in this regard the large congregate settings that still exist in many states. Individuals living in SAs also expressed higher satisfaction and well-being than those in more congregate and more highly supervised settings. Their personal adjustment (which behaviorally began as no different from that of the GH residents) was no different from that of persons who had lived with their families most or all of their lives.

73What does this mean for social policy? Despite recent articles discounting the importance of the physical and structural (organizational) characteristics of living settings (Zigler et al., 1990), ours are not the only studies that clearly show that structural features such as size and organization, as well as careprovider philosophy and competence, have a very strong impact on the people living within them (Campbell & Bailey, 1984; Parker & Boles, 1990). Our research shows that individuals take on more responsibility, exercise more choice and independence, and express greater personal satisfaction when they live individually or with one or two friends in a supported (but not “managed”) situation, like most other adults. One needs only to think of the experience of most adults with group living, even in the absence of adult supervision and control. They find it very difficult to live together with three to five other adults for any length of time in one “family.” The least that can be said is that although such a choice is available to people, few exercise it for any length of time, and then usually only under the most extreme economic exigencies.

74These results also suggest, by extrapolation, that foster care is not an ideal setting to promote lifestyle Normalization or, perhaps, personal satisfaction. Such homes do not have the interpersonal history that families do, so unless providers have extraordinary Normalization competence, foster homes will probably be characterized by caretaking rather than life-sharing.

75Third, these results tell us that social integration, as envisioned by advocates and providers, is not a reality for adults with mental retardation, not even for those in supervised apartments or for those who have lived with their families in their home communities all of their lives. Few name as members of their social network any community peer friends, that is, persons who provide companionship, instrumental or emotional support, or friendship. At the same time, careproviders report that adults with retardation engage in very few (and mostly no) activities that include a community adult other than staff, kin, or other individuals with disabilities. While young people in supervised apartments lead lives that look no different from those of other single adults living in the community with regard to access, activation, and satisfaction, their networks are only half as large and contain few or no nondisabled peer friends, and half their friends are staff members. Although the most integrated and experiencing the most normalized lifestyles, these apartment residents are not achieving community acceptance and integration, in the sense of having developed informal support (Rosen & Burchard, 1990; Wolfensberger, 1992).

76Another major aspect of social integration and support is notably absent from the lives of persons with mental retardation, setting them apart and potentially depriving them of critical social support and personal satisfaction: They are rarely married or living with a partner, or raising children, which is by far the modal circumstance for adults in our society.

77Nevertheless, the participants themselves were generally highly satisfied with the social support they received and with their friendships, regardless of their setting. There were no differences between groups on these outcomes, and care was taken to avoid acquiescence in responding. Participants generally had peer friendships (with exceptions found mainly in family homes), often named a peer as a best friend, and relied on peers for some of their support, although most was obtained from formal support. This begs the question of choice, at least for those who were living in settings with more opportunities and more choice, with respect to developing friendships, which are usually based upon shared interests, shared activities, opportunities for contact, and reciprocity. It may be that the constitution of networks is a function of the latter determinants of friendship development, rather than reflecting a social-policy failure. Or it may be due to a lack of opportunity or a lack of skills in making friendships outside of the service system. It is clear, however, that investigating fruitful ways of developing and supporting friendships and informal supports is an important agenda item for social policy.

78Another surprising finding was the stability of the social-support networks over a 3-year time span. Vermont communities undoubtedly have greater stability than large urban cities, and these individuals did not move during that time. Nevertheless, about half the persons named were the same individuals in many networks across the 3 years.


79There were significant limitations to this research. It was completed within one small state that had a clearly articulated policy based on Normalization and a mandate to serve persons in settings no larger than six residents. This constituted a fairly homogeneous system. Participants in the longitudinal study were selected to be able to respond to simple interviews, which excluded individuals with severe disabilities. However, considering the results of our earlier staff-competency validation studies, which did include individuals with severe disabilities and a broader range of settings, we are of the opinion that had the adjustment and integration of persons with greater individual differences and living in a wider range of circumstances (more congregate sites and foster homes) been studied longitudinally, the results would have been even more pronounced with respect to the importance of careprovider attitudes and competencies and the impact of environmental Normalization. This, I believe, underlines the power and importance of efforts to employ Normalization principles in supporting people in the community. Even with a fairly “model” homogenous system and, in the later studies, a fairly homogeneous group of participants, relatively small differences in Normalization experiences and staff Normalization competencies among programs still had an important effect, in terms of autonomy, the independent performance of activities, individualization, physical and social integration, and personal satisfaction with residence and lifestyle.


80As Edgerton (1988) pointed out after following, for over 3 decades, persons with mental retardation living in the community, many individuals adapt successfully, from their own perspective, if given the opportunity and supports available to other community members. They may lead financially impoverished lives and have much smaller networks, compared to some standards, but they may not view their lives or support as impoverished. It seems important, for social policy, to provide living and working opportunities that afford as much individualization, choice, and personal control as possible and, for service systems, to have more respect and faith in individuals’ abilities to make and exercise choices, including lifestyle-related ones. At the same time, it is imperative to investigate ways of helping people obtain the opportunities and skills to develop and maintain friendships and to access support systems, both formal and informal, as needed throughout their lives.

81Finally, to quote Wolf Wolfensberger on the relationship of “research, empiricism, and the principle of Normalization” (1980): “How much ‘research’... should be conducted to support Normalization implications for attractive environments;... access to services; age-appropriate and culturally valued forms of personal appearance, labeling, activities,... decor; individualisation... programming; avoidance of crowding; competent and image-enhancing staff; warmth of interaction..” (p. 126). These are basic human values that should need little validation. Although it does indeed provide support and validation for the importance of Normalization values, even more to the point, the research presented above shows that environments are very important in furthering the goals of social policy. It shows which kinds of environments are organizationally better designed to promote Normalization values and which types of service providers foster the opportunities that environments provide. These factors, environmental organization and structure and staff competencies, contributed to the attainment of Normalization-derived social-policy goals, beyond any effect due to individual differences. Where one lives, whom one lives with, and how one lives are important factors in assisting persons with disabilities toward a lifestyle we all treasure. Our studies also point again to the limitations of the service system in establishing social integration for persons they are serving. This research should also serve as a caveat concerning the types of living arrangements that could militate against the provision of the best opportunities for lifestyle Normalization, namely, congregate, nonconsensual, other-controlled environments. It also points to the continued need for determining how to assist persons toward individualized, self-chosen personal integration as well as personal independence.

82We have come a long way in the last 25 years, since the introduction of Normalization values. Many places are still fighting the conditions that Normalization was introduced to combat. Even in places where living conditions have been established that maximize Normalization values and hence opportunities for people, changing financial conditions now threaten the progress made, as cheaper solutions are sought. And there is a continuing need to discover how to make true social integration into our communities a reality for more people.



Annual report to Congress on implementation of individuals with disabilities act (1991). Washington, DC: U.S. Government Printing Office.

Bradley, V. J., & Allard, M. A. (1992). The dynamics of change in residential services for people with developmental disabilities. In J. W. Jacobson, S. N. Burchard, & P. J. Carling (Eds.), Community living for people with developmental and psychiatric disabilities (pp. 284-302). Baltimore: John Hopkins University Press.

Braddock, D., & Fujiura, G. (1991). Politics, public policy, and the development of community services in the United States. American Journal of Mental Retardation, 95, 369-387.

Braddock, D., Hemp, R., Fujiura, G., Bachelder, L., & Mitchell, D. (1989). The third national study of public spending for mental retardation and developmental disabilities: Summary. Chicago: University of Illinois, Institute for Study of Developmental Disabilities.

Bruininks, R. H., Rotegard, L. L., Lakin, K. C., & Hill, B. K. (1987). Epidemiology of mental retardation and trends in residential services in the United States. In S. Landesman & P. Vietze (Eds.), Living environments and mental retardation (pp. 17-42). Washington, DC: American Association of Mental Retardation.

Burchard, S. N., Hasazi, J. E., Gordon, L. R., & Yoe, J. (1991). A comparison of lifestyle and adjustment in three community residential alternatives. Research in Developmental Disabilities, 12, 127-142.

Burchard, S. N., & Hutchins-Fuhr, M. (1990, December 7). Comparison of characteristics of social support and personal satisfaction over three years for adults living in the community. Paper presented at the annual meeting of the Association for Persons with Severe Handicaps (TASH), Chicago.

Burchard, S. N., Pine, J., & Gordon, L. R. (1984, August 28). Relationship of manager competence and program Normalization to client satisfaction in group homes. Paper presented at the annual meeting of the American Psychological Association, Toronto.

Burchard, S. N., Pine, J., & Gordon, L. R. (1990). Manager competence, program Normalization and client satisfaction in group homes. Education and Training in Mental Retardation, 25, 277-285.

Burchard, S. N., Pine, J., Gordon, L. R., Joffe, J. M., Widrick, G. C., & Gov, E. (1987). The relationship of manager competence to residential program quality in small community residences. In J. A. Mulick & R. Antonak (Eds.), Transitions in mental retardation (Vol. 2, pp. 47-69). Region X AAMD Monograph. Norwood, NJ: Ablex Press.

Burchard, S. N., Pine, J., Widrick, G. C., & Creedon, S. (1985, December 5). The relationship of entry-level competencies to job performance and program quality in small community residences. Paper presented at the annual meeting of the Association for Persons with Severe Handicaps (TASH), Boston.

Burchard, S. N., Rosen, J. W., Gordon, L. R., Yoe, J., Hasazi, J. E., & Simoneau, D. (1992). Evaluation of social support and lifestyle satisfaction in community living arrangements for persons with mental retardation. In J. W. Jacobson, S. N. Burchard, & P. J. Carling (Eds.), Community living for people with developmental and psychiatric disabilities (pp. 137-154). Baltimore: John Hopkins Press.

Butterfield, E. C. (1987). Why and how to study the influence of living arrangements. In S. Landesman & P. Vietze (Eds.), Living environments and mental retardation (pp. 43-60). Washington, DC: American Association of Mental Retardation.

Campbell, V. A. & Bailey, C. J. (1984). Comparison of methods for classifying community residential settings for mentally retarded individuals. American Journal of Mental Deficiency, 89, 44-49.

Carpenter, M., & Burchard, S. N. (in preparation). Environmental predictors of residence satisfaction, psychological well-being, and work satisfaction of adults with mental retardation: A longitudinal study. Burlington, VT: University of Vermont, Psychology Department.

Dupuy, H. J. (1978, October 17). Self-representations of general psychological wellbeing of American adults. Paper presented at the annual meeting of the American Public Health Association, Los Angeles.

Edgerton, R. B. (1988). Aging in the community: A matter of choice. American Journal of Mental Retardation, 92, 331-335.

Emerson, E. B. (1985). Evaluating the impact of deinstitutionalization on the lives of mentally retarded people. American Journal of Mental Deficiency, 90, 277-288.

Eyman, R. K., Demaine, C. G., & Lei, T. (1979). Relationship between community environments and resident changes in adaptive behavior: A path model. American Journal of Mental Deficiency, 83, 330-338.

Fiorelli, J. S., & Keating, D. J. (1979). Overview. In J. S. Fiorelli (Ed.), A curriculum model for preservice training of alternative living arrangement direct service personnel. Philadelphia: Temple University, Developmental Disabilities Center.

Gordon, L. R., Burchard, S. N., Hasazi, J. E., Yoe, J. T., Dietzel, L. C., & Simoneau, D. (1992). Stability and change in the life-style and adjustment of adults with mental retardation living in community residences. In J. W. Jacobson, S. N. Burchard, & P. J. Carling (Eds.), Community living for people with developmental and psychiatric disabilities (pp. 167-182). Baltimore: John Hopkins Press.

Hull, J. T., & Thompson, J. C. (1980). Predicting adaptive functioning of mentally retarded persons in community settings. American Journal of Mental Deficiency, 85, 253-261.

Hull, J. T., & Thompson, J. C. (1981). Factors contributing to Normalization in residential facilities for mentally retarded persons. Mental Retardation, 19, 69-73.

Landesman, S., & Butterfield, E. C. (1987). Normalization and deinstitutionalization of mentally retarded individuals. American Psychologist, 42, 809-816.

Nirje, B. (1970). Symposium on Normalization. The Normalization principle: Implications and comments. Journal of Mental Subnormality, 16(31), 62-70.

Parker, R., & Boles, S. (1990). Integration opportunities for residents with developmental disabilities: Differences among supported living sites and residents. Education and Training for the Mentally Retarded, 25, 76-82.

Pratt, M. W., Luszcz, M. A., & Brown, M. E. (1980). Measuring dimensions of quality of care in small community residences. American Journal of Mental Deficiency, 85, 188-194.

Rogers, C. R. (1969). The interpersonal relationship: The care of guidance. Harvard Review, 39, 475.

Rosen, J., & Burchard, S. N. (1990). Community activities and social support networks of adults in semi-independent living: A social comparison. Education and Training in Mental Retardation, 25, 193-204.

Seltzer, G. B. (1980). Residential satisfaction and community adjustment. Paper presented at the annual meeting of the American Association of Mental Deficiency, San Francisco.

Seltzer, G. B., & Seltzer, M. M. (1976). The community adjustment scale. Cambridge, MA: Educational Projects, Inc.

Seltzer, M. M. (1983). Nonexperimental field research methods. In J. L. Matson & J. A. Mulick, (Eds.), Handbook of mental retardation (pp. 557-570). New York: Pergamon.

Smull, M. W., & Bellamy, G. T. (1990). Community services for adults with disabilities: Policy challenges in the emerging support paradigm. In L. H. Meyer, C. A. Peck, & L.

Brown, (Eds.), Critical issues in the lives of people with severe disabilities (pp. 527-536). Baltimore: Brookes.

Thousand, J., Burchard, S. N., & Hasazi, J. E. (1986). Field-based generation and social validation of manager and staff competencies for small community residences. Applied Research in Mental Retardation, 7, 263-283.

Weinberg, R. B. (1984). Development of self-report for reliably measuring the social support system. Paper presented at the annual meeting of the American Psychological Association, Toronto.

Wolfensberger, W. (1972). The principle of Normalization in human services. Toronto: National Institute on Mental Retardation.

Wolfensberger, W. (1980). Research, empiricism, and the principle of Normalization. In R. J. Flynn & K. E. Nitsch (Eds.), Normalization, social integration, and community services (pp. 117-129). Baltimore: University Park Press.

Wolfensberger, W. (1992). A brief introduction to Social Role Valorization as a high-order concept for structuring human services (Rev. ed.). Syracuse, NY: Syracuse University, Training Institute for Human Service Planning.

Wolfensberger, W., & Glenn, L. (1975). Program analysis of service systems (PASS): A method for the quantitative evaluation of human services (3rd ed.). Toronto, ON: National Institute on Mental Retardation.

Wolfensberger, W., & Thomas, S. (1983). PASSING (Program analysis of service systems’ implementation of Normalization goals): Normalization criteria and ratings manual (2nd ed.). Downsview, ON: National Institute on Mental Retardation.

Zigler, E., Hodapp, R. M., & Edison, M. R. (1990). From theory to practice in the care and education of mentally retarded individuals. American Journal of Mental Retardation, 95, 1-12.


1 This research was supported in part by the Vermont State Department of Mental Health (DMH) and the National Institute of Disability and Rehabilitation, Research Grants Nos. 1-33-GH-40203 and 1-33-MH-50078. The author wishes to acknowledge the assistance of the participants, their families, state and local community mental retardation providers, and former DMH director Ronald Melzer, whose support, cooperation, and participation were essential. The author also wishes to acknowledge her many colleagues whose contributions to this research over the years were substantive and invaluable: Joseph Hasazi, Lawrence Gordon, Joan Pine, Jacqueline Thousand, James Yoe, Julie Rosen, Marc Carpenter, Gary Widrick, Justin Joffe, and Moira Hutchins-Fuhr.


Ph.D., Associate Professor of Psychology, University of Vermont, Burlington, VT, USA

© Les Presses de l’Université d’Ottawa | University of Ottawa Press, 1999

Conditions d’utilisation :

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.


Volume papier
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search