- Email: [email protected]
The community integration measure: Development and preliminary validation
429
ARTICLES
The Community Integration Measure: Development and Preliminary Validation Mary Ann McColl, Phi), Diane Davies, MSc, Peter Carlson, PhD, Jane Johnston, BA, BScN, Patricia Minnes, PhD ABSTRACT. McCol] MA, Davies D, Carlson P, Johnston J, Minnes P. The Community Integration Measure: development and preliminary validation. Arch Phys Med Rehabil 2001;82: 429-34. Objective: To present a new measure of commnnity integration, the Community Integration Measure (CIM), and to offer preliminary information about its psychometric properties. Design: Validation study. Setting: Community. Participants: Ninety-two participants placed in 3 subgroups (brain injury survivors, n = 41; significant others, n = 36; college students, n = 15). Main Outcome Measures: The distributional properties, factor structure, internal consistency reliability, content validity, discriminant validity, concurrent validity, and construct validity of the CIM. Results: All items correlated positively with each other and with the total score. Principal components factor analysis confirmed a 1-factor structure, which explained 44.1% of the variance. Inten]al consistency reliability, using Cronbach's alpha, was .87. Content validity was assured by the development procedure, correspondence with the theoretical model, and direct use of consumer language. Discriminant validity was supported by the CIM's ability to differentiate between subsamples. Criterion validity was supported by using correlations with the Community Integration Questionnaire. Construct validity was supported by correlations with the Interpersonal Support Evaluation List. Conclusion: The CIM offers a brieL easily administered measure of community integration that conforms to an empirically derived theoretical model and is psychometrically sound. Key Words: Brain injuries; Questionnaires; Rehabilitation; Social adjustment.
© 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation OMMUNITY INTEGRATION has been a focus for health service providers and researchers for the past 20 years, C since the World Health Organization highlighted community participation in its Health for All by the Year 2000. ~ And yet,
From the School of Rehabilitation Therapy (McColL Davies) and Department of Psychology (Minnes), Queen's University; and Regional Community Brain Injury Services (Carlson, Johnston), Kingston, Ont. Accepted in revised form February 7, 2000. Supported by the Ontario Ministry of Health. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Mary Ann McColl, PhD, Queen's University, School of Rehabilitation Therapy, Kingston, Ont K7L 3N6, Canada, e-mail: [email protected]. 0003-9993/01/8204-5804535.00/0 doi: 10.1053/apmr.2001.22195
20 years later, we still struggle with the definition of community and the meaning of community integration. 2 In rehabilitation, community integration has been conceptualized as the opposite of handicap, where handicap has been defined as the social disadvantage resulting from disability or impairment.3 5 Recent revision of the concept of handicap has led to the more universal notion of participation as a possibility for operationalizing integration. Participation is made up of 9 dimensions: personal maintenance; mobility; exchange of information; social relationships; home life and assistance to others; education; work; economic life; and community and civic life. 6 These dimensions of participation correspond closely with recent work on integration, primarily with developmentally disabled people. Halpern et al 7 identified 4 factors constituting community integration: occupation, residential environment, social support, and overall satisfaction. Bruininks et al 8 and McGrew and Bruininks 9 identified 3 other factors: leisure participation, family contact, and community assimilation and acceptance. This set of 7 factors was later confirmed by McGrew et al. 1° These investigations underline the multidimensional nature of community integration. Our own research lj used a qualitative approach to generate a definition of community integration. Between 1994 and 1996, we studied 18 people with moderate-to-severe acquired brain injuries for their first year in the community after inpatient rehabilitation. In monthly interviews, we discovered that they defined community integration as a function of 4 factors: assimilation (conformity, orientation, acceptance); social support (close and diffuse relationships); occupation (leisure, productivity); and independent living (personal independence, satisfaction with living arrangement) (fig 1). These factors are highly consistent with the research reported earlier. Community integration is arguably the ultimate aim of rehabilitation. Presumably, every person participating in rehabilitation hopes to be happily situated, productively occupied, and effectively supported in the community. However, confirming that community integration has been achieved at the end of rehabilitation has been hampered by a lack of consensus on definition and measurement of community integration. 1° MEASURES OF C O M M U N I T Y I N T E G R A T I O N The current standard for measuring community integration in the area of brain injury rehabilitation is the Community Integration Questionnaire (CIQ). Willer et aP .12 proposed a definition of community integration that took into account 3 dimensions: home and family, and social and vocational independence. They conceptualized community integration as the opposite of handicap, with emphasis on participation of the individual in his/her environment. Consistent with other research on community integration, they recognized that community integration cannot simply be observed, but must be understood in terms of consumer perceptions and the experience of integration? TM The CIQ is a 15-item measure designed for people with acquired brain injuries. It is divided into 3 subscales (home integration, social integration, productivity), Arch Phys Med Rehabil Vol 82, April 2001
430
COMMUNITY INTEGRATION MEASURE, MeColl
COMMUNITY INTEGRATION
/
ASSIMILATION • • •
SUPPORT
Acceptance
OCCUPATION •
Conformity Orientation
INDEPENDENT LIVING
•
Close
•
Productivity
Diffuse relationships
Independence Living situation
relationships •
•
•
Leisure
Fig 1. Theoretical model of community integration. Data from McColl et al. 11
which have been confirmed by using factor analyses. 15 The CIQ is widely used, and is valued for its quantitative properties and ease of administration. Scoring of the CIQ is based on a number of assumptions that may not apply in all situations. For example, scoring favors respondents with able-bodied friends over those with disabled friends; respondents who do things with friends rather than with family; and those who do household activities alone, rather than with others. Another option for measuring community integration is the Reintegration to Normal Living Index (RNLI). 16 The RNLI is made up of 11 items. The definition of community integration used for this measure was developed through consultations with consumers, family members, and health care professionals. The domain of the RNLI was derived by using a multistage process to identify domains and language for the concept. Items are scored with a 10-cm visual analog scale, anchored with the statements: "fully describes my situation" and "does not describe my situation." The RNLI is short, easy to administer, and widely used in rehabilitation research and practice. Pilot testing confirmed that it is conceptually and psychometrically sound. Rappaport et a117 developed another community integration measure for their study of 10-year outcomes after brain injury. They defined community integration according to 4 items dealing with: (1) the impact of physical impairment on living situation; (2) the impact of mental impairment on living situation; (3) the impact of mood on work; and (4) the impact of frustration tolerance on work. Rappaport presented normative data, but not psychometric. Another well known instrument to measure community integration is the Craig Handicap Assessment and Reporting Tool (CHART). 4 It is a 27-item scale that also defines community integration as the opposite of handicap. CHART items attempt to distinguish between levels of handicap, as outlined in the International Classification of Impairments, Disabilities, and Handicaps (ICIDH) (ie, ability to fulfill 6 community roles: orientation roles, physical independence, mobility, occupation, social integration, economic self-sufficiency). Most items have yes-no, forced choice, or other quantitative response formats, such as, "How many hours per week do you spend in home maintenance activities?" The definition of community integration used for the CHART and CIQ is the ICIDH 3 notion of handicap. When this notion was initially conceived, it was a useful and powerful way to understand the relation between disabled
Arch Phys Med Rehabil Vol 82, April 2001
individuals and their communities. However, many now consider the term unacceptable for 2 main reasons: (1) it places too much emphasis on the individual, and not enough on the community, as the locus of problems of community integration; and (2) it identifies people with disabilities as a special interest group, at the margins of society, rather than as full members of society entitled to all its rights and privileges. 18 A conceptual change is reflected in the revised ICFD, which is scheduled for final publication in 2001. 6 This reconceptualization of "disability" and "handicap" as "activity" and "participation" may require some rethinking of community integration measures. D E V E L O P M E N T OF T H E C O M M U N I T Y INTEGRATION MEASURE Although all the measures reviewed earlier have excellent reputations and properties, they do not use standard definition of community integration. This lack of consensus impelled us to undertake a 3-year study looking at intuitive definitions of community integration used by survivors of acquired brain injury in their assessments of their own integration. We identified 9 factors in that study, that mentioned previously, provided the basis for a measure of community integration--the Community Integration Measure (CIM).11 The CIM asks participants if they: (1) know the rules and how to follow them; (2) know their way around; (3) are accepted for who they are; (4) have people in the community with whom they feel close; (5) have relationships with different kinds of people in the community; (6) find things to do in their leisure time; (7) have something to do that makes them feel productive and worthwhile; (8) have some degree of independence; and (9) have a suitable place to live. The CIM is a 10-item, client-centered measure of community integration that uses the words of participants themselves, and makes no assumptions about the relative importance of particular activities or relationships. For example, unlike the CIQ, the CIM does not assume that independent participation is a sign of greater integration than is supported or mutual coparticipation. It can be administered either face-to-face, by telephone, or by the subject. It requires only a basic literacy level, and can be easily administered in 3 to 5 minutes. It has been used in practice with a broad range of clients with various disabilities; however, the present analyses focus on people with moderate-to-severe brain injuries. Presumably, this study population is more demanding in terms of both community inte-
COMMUNITY INTEGRATION MEASURE, McColl Table 1: Sample Characteristics and Subsample Comparisons Total Sample (n = 92) Age, mean ± SD (range) Gender Men (%) W o m e n (%)
College Students (n = 36)
Survivors {n - 41)
Family Members (n = 15)
32.1 _+ 13,8 35.4 ± 10.7 22.9 ± 4.1 55.2 -+ 13.3 (19-76) (20-62) (19-39) (26-76) 37 (40%) 55 (60%)
26 (58%) 15 (42%)
7 (19%) 27 (81%)
431
Table 2: Means -+ SDs and Distribution of CIM Scores n Total sample Survivors Coflege students Family members
Mean + SD
92 41 36 15
31.4 28.8 33.9 32
± + ± ±
7.1 7.7 4.4 8,4
Minimum
Maximum
9 9 21 20
40 40 40 40
4 (27%) 11 (73%)
gration and self-report measurement. The CIM results in a single summary score (between 10 and 50) that is the unweighted sum of the 10 items, each with 5 response options (see appendix). This study sought to offer preliminary information about the CIM and its psychometric properties, including its distributional properties, factor structure, internal consistency reliability, as well as its content, discriminant, concurrent, and construct validity. METHODS The study sample involved 92 participants, divided into 3 subsamples: (1) 41 participants in the original 3-year study, all of whom had moderate-to-severe acquired brain injuries ~~: (2) 36 community college students, recruited voluntarily; and (3) 15 family members, nominated by members of the acquired brain injury sample. Characteristics of the 3 samples are described in table 1. The following 3 measures were used in the validation study: the CIM, CIQ, and the Interpersonal Support Evaluation List (ISEL). The CIM, described earlier, is a brief, 10-item measure of community integration, developed on the basis of an empirically derived, explicit theoretical model. 1~ The CIQ, 15 described earlier, is a 15-item questionnaire with 3 possible response options for each question. It is well known in rehabilitation research and practice, particularly in the field of brain injury rehabilitation. The ISEL ~9 is a 40-item measure of social support that asks respondents about the availability of individuals to provide support. The ISEL has been used to measure social support among people with disabilities, including several items that address disability-specific issues (ie, transportation, assistance with activities of daily living, installation of household adaptations), ~° The adapted ISEL measures 3 types of support: informational (information, advice, guidance), emo-
tional (belonging, esteem support), and instrumental (practical, financial). This study complied with the standards of the Queen's University Research Ethics Bom'd. RESULTS Distributional Properties We examined item means and standard deviations (SDs), skewness and sensitivity, correlations among items, and correlations between items and the total score by using item analysis. All items were found to have means --_ SDs of between 2.86 ± .86 and 3.54 ± 1.62 (range, 1-5, with 1 = never, 5 = always). As regards sensitivity, though items tended to be skewed positively, only 1 item had more than 53% of participants choosing the top response option. (On item 8, "There are people I can say hello to . . . . " 72% of participants selected "always" as their response.) Table 2 shows the means and ranges for the total sample and for each of the subsamples. Notably, survivors had the lowest mean score, family members the greatest variance, and college students the least variance. All items correlated positively with each other and with the total score (table 3). Factor Structure Principal components factor analysis, using data from all 92 participants, resulted in a 1-factor solution that explained 44.1% of the variance in the 10 items of the CIM. The eigenvalues drop off dramatically after the extraction of the first factor (eigenvalues for the first and second factors, respectively: 4.40, 1.43) by using the scree-plot. Table 4 shows the factor loadings for the 10 items of the CIM, all of which are between .534 and ,797. Two-, 3-, and 4~factor structures were also explored, based on the underlying theoretical model discussed earlier and
Table 3: Correlations Between Items and With Total Score Item
1 2 3 4 5 6 7 8 9 10
Item 1
2
3
4
5
6
7
8
9
10
Total
1.00
.714 .634 .453 .659 .480 .416 .574 .545 .549 .579
1.00 .556 .368 .669 .301 .618 .469 .491 .358 .344
1.00 .397 .371 .561 .190 .441 .516 .313 .396
1.00 .404 .406 .027 .268 .173 .291 .381
1.00 .248 .495 .573 .314 .368 .430
1.00 .079 .172 .271 .362 .541
1.00 .371 .337 .308 .550
1.00 .476 .371 .336
1.00 .337 .339
1.00 ,550
Abbreviations: 1, I feel part of this c o m m u n i t y ; 2, I know m y w a y around; 3, I k n o w the rules; 4, I feel that I am accepted; 5, I can be independent; 6, I like w h e r e I'm living; 7, There are people I feel close to; 8, I k n o w people to say hello; 9, There are things I do for fun; 10, I have something useful to do.
Arch Phys Med Rehabil Vol 82, April 2001
432
COMMUNITY INTEGRATION MEASURE, McColl
Table 4: Factor Loadings of CIM Items on 1-Factor Solution (Principal Components Analysis) (n = 92) Item 1. I feel like part of this c o m m u n i t y , like I belong here. 4. I feel that I am accepted in this community, 2. I k n o w my w a y around in this community. 7. There are people that I feel close to in this community, 10. I have something to do in this c o m m u n i t y during the main part o f m y day that is useful and productive. 8. I k n o w a n u m b e r o f people in this c o m m u n i t y well enough to say hello and have them say hello back. 9. There are things that I can do in this c o m m u n i t y for fun in my free time. 5. I can be i n d e p e n d e n t in this community. 3, I k n o w the rules in this c o m m u n i t y and I can fit in with them, 6. I like w h e r e I'm living now.
,750 ,726 .687
Criterion validity was evaluated by using correlations of the CIM with the CIQ. 5 As mentioned, the CIQ is currently the standard for measuring community integration in brain injury rehabilitation. Our research found a correlation of .34 between the CIQ and the CIM (table 6). This suggests that there is some common variance between the 2 measures (12%), but that they do not duplicate each other's content. We believe the difference lies in the emphasis the CIQ places on independence.
.673
Construct Validity
Factor Loading ,797
.647
.637 .584 .653 .534
shown in figure 1. Only the 2-factor solution was readily interpretable, and produced eigenvalues greater than 1. The 2-factor solution explains 58.3% of variance in terms of 2 factors (table 5). After varimax rotation, the 2 factors each had 5 items loading on them, with loadings between .530 and .830. The 2 factors have been labeled: belonging and independent participation.
Internal Consistency Reliability Internal consistency reliability, using Cronbach's alpha, indicated a value of .87. Values above .80 are considered strong evidence that a measure is internally consistent; that is, that all items are measuring the same underlying construct. All items correlated positively with one another, and alpha value did not improve by excluding any single item. Subgroup alpha values were: survivors, .83: college students, .78; family members, .92.
Content Validity Content validity requires some assurance that the items on the measure cover the domain of the construct. In our case, content validity is assured by 3 factors associated with instrument development. First, the empirical, qualitative process leading to model development was well conceived and exhaustive11; second, the items on the CIM have a 1-to-1 relationship with concepts in the model (fig 1); and third, the items on the CIM were constructed by using the words of participants themselves, so to ensure authenticity and credibility.
Discriminant Validity Discriminant validity is supported by evidence that the measure distinguishes between groups that are expected to perform differently on the measure. In our study, we expected to attribute differences between acquired brain injury survivors and the other 2 groups to the disability. We used 1-way analysis of variance to compare CIM scores across the 3 subsamples. As expected, we found a significant difference between groups (F9J.2 : 5.5, p < .006), with the most marked contrast being between the acquired brain injury survivors and the college students (see table 2 for subgroup means). Arch Phys Med Rehabil Vol 82, April 2001
Criterion Validity
To show construct validity, one must show that CIM scores correlate in expected ways with theoretically related constructs. In our case, we expected that CIM scores would correlate positively with social support, occupation, and independent living--the 3 main factors that make up community integration, according to our model. 11 In the present study, we sought to confirm the first of these hypotheses by examining correlations between the CIM and the ISEL. ~9 We found a correlation of .42. showing that about 20% of the variance in the CIM is explained as a function of social support. This is consistent with the fact that social support items make up 2 of the 10 items on the CIM. The C1M also correlated with the informational and emotional subscales of the ISEL (table 6). DISCUSSION The study offers a new method for measuring community integration: the CIM. The CIM is based on an explicit theoretical model. It is client-centered, in that it assesses community integration from the client's perspective. It is accessible, in that it was constructed by using language derived from client interviews. It is brief, easy to administer (by interview or selfadministration), and easy to score. Finally, the CIM makes no assumptions about the relationship between independence and integration or about the relative value of different types of relationships.
Table 5: Factor Loadings of CIM Items on 2-Factor Solution (Principal Components Analysis With Varimax Rotation) (n = 92) Item Belonging like w h e r e I'm living now, feel like part of this c o m m u n i t y , like I belong here. feel that I am accepted in this community. There are people that I feel close to in this community. I k n o w a n u m b e r of people in this c o m m u n i t y well enough to say hello and have them say hello back. Independent Participation I can be i n d e p e n d e n t in this community, I have something to do in this c o m m u n i t y during the main part o f my day that is useful and productive. I know the rules in this c o m m u n i t y and I can fit in with them, I k n o w my w a y a r o u n d in this community. There are things that I can do in this c o m m u n i t y for fun in m y free time.
Factor 1
Factor 2
.830 .809 ,740
,302
.677 .475
.331
.830 .749
.671 .383
.656
.377
.530
433
COMMUNITY INTEGRATION MEASURE, McColl Table 6: Correlations Among the CIM, CIQ, and ISEL
CIM CIQ ISEL total ISEL instrumental ISEL informational ISEL emotional
CIM
CIQ
ISEL Total
1.00 .343 .425* .055 .380* .480*
1.00 .344 .264 .313 .287
1.00 .656* .856* .911'
ISEL I n st ru m e n t a l
1.00 .528* ,390"
ISEL Informational
ISEL Emotional
1.00
.640'
1.00
* p < .05.
* p < .01.
Psychometric findings to date offer assurances of the CIM's measurement properties. Item and scale analyses indicate that the construction and scoring of the measure are sound and reasonable. There is an underlying construct that can be measured with these 10 items that conforms to our definition of community integration. Further, enough shared variance exists between items that it makes sense to sum the 10 items to construct a single overall score representing community integration. The literature on community integration exhorts researchers to consider community integration a multidimensional construct that must be measured in depth and breadth, s.9 This study does not directly contradict that notion; in fact, the theoretical model on which the measure is built recognizes 4 categories of factors contributing to community integration (assimilation, occupation, relationships, living situation). However, our research does suggest that underlying that very complex construct is a unitary experience of belonging and participating. Thus, the CIM can be differentiated from other available instruments because it focuses on the underlying subjective experience of integration, rather than on the more objective or observable components of integration. The discriminant analysis has shown that the CIM can be used to measure community integration in the general population as well as in a rehabilitation population. Although the relative importance of particular items may change between different populations, the overall score of the 10 unweighted items remains viable and meaningful. Validity of the CIM is supported in our own work and independently. Nolte, 21 in a sample of 97 people with moderate-to-severe brain injuries, also found that CIM scores correlated to the CIQ (r = .26, p < .05) and to the 3 scales of the ISEL (instrumental support: r = .25, p < .05: informational support: r --- .26, p < .05; emotional support: r = .46, p < .001). In addition, Nolte found the CIM scores correlated to other model components, such as quality of life (r = .25, p < .05) and positive affect (Affect Balance Scale: r = .30, p < .05). Further studies of validity are planned to examine the relationship between the CIM and other model parameters, specifically, independent living and occupation. Our findings must be understood in the context of a relatively small sample for work of this nature. With 92 participants and a measure of 10 items, the standard of 10 subjects per variable in the factor analysis is not quite met, 22 though the sample base is close enough that estimates should be quite stable. Further, the present analysis is limited by the consideration of a single disability group and age--adults with acquired brain injuries. We have no evidence that the same properties could be expected with other age groups, particularly children, or with other disability groups. Future research is anticipated to address these issues. In addition, it would be necessary to
evaluate the extent to which the principle of universality is expressed in the CIM by testing it more extensively with general population samples. CONCLUSION The CIM offers clinicians a method for assessing the underlying experience of community integration and participation. It contributes to a full understanding of adjusUnent to community living after disability. The CIM offers a number of advantages--it is brief, easy to administer and score, and readily understandable by clients and families. On a more technical level, it is based on an empirically derived, theoretical model and has been shown to conform to that model and to show desirable psychometric properties.
Acknowledgments: We appreciate the assistance of our community partners--14 community residential programs for people with brain injuries in Ontario. In particular, we thank our local partner, Regional Community Brain Injury Services, in Kingston. References I. World Health Organization. Health for all by the year 2000. Geneva: WHO; 1980. 2. Jewkes R, Murcott A. Meanings of community. Soc Sci Med 1996;43:555-63. 3. World Health Organization. International classification of impairments, disabilities, and handicaps. Geneva: WHO; 1980. 4. Whiteneck GG, Charlifue SW, Gerhart KA, Overholser JD, Richardson GN. Quantifying handicap: a new measure of longterm rehabilitation outcomes. Arch Phys Med Rehabil 1992; 73:519-26. 5. Willer B, Rosenthal M, Kreutzer JS, Gordon WA, Rempel R. Assessment of community integration following rehabilitation for traumatic brain injury. J Head Trauma Rehabil 1993;8(2):75-87. 6. World Health Organization. International classification of functioning and disability-Beta-2 draft. Geneva: WHO: 1999. 7. Halpern A, Nave G, Close D, Nelson DJ. An empirical analysis of dimensions of community adjustments for adults with mental retardation in semi-independent living programs. Aust N Z J Dev Disabil 1986;12:147-57. 8. Brnininks R, Chen T, Lakin D, McGrew K. Components of personal competence and community integration for persons with mental retardation in small residential programs. Res Dev Disabil 1992;13:463-79. 9. McGrew K, Bruininks R. A multidimensional approach to measurement of community adjustment. In: Hayden M, Abery B, editors. Challenges for a service system in transition. Baltimore (MD): Brookes: 1994. p 65-79. 10. McGrew KS, Johnson DR, Brnininks RH. Factor analysis of community adjustment outcome measures for young adults with mild to severe disabilities. J Psychoeduc Assess 1994;12:55-66. 11. McColl MA, Carlson P, Johnston J, Minnes P, Shue K, Davies D, et al. Definition of community integration: perspectives of people with brain injuries. Brain Inj 1998;12:15-30. Arch Phys Med Rehabil Vol 82, April 2001
434
COMMUNITY INTEGRATION MEASURE, McColl
12. Willer B, Linn R, Allen K. Community integration and barriers to integration for individuals with brain injury. In: Finlayson MAJ, Garner S, editors. Brain injury rehabilitation: clinical considerations. Baltimore (MD): Williams & Wilkins; 1992. p 355-75. 13. Evans RW, Ruff RM. Outcome and value: a perspective on rehabilitation outcomes achieved in acquired brain injury. J Health Trauma Rehabil 1992;7:24-36. 14. Holosko MJ, Huege S. Perceived social adjustment and social support among a sample of head injured adults. Can J Rehabil 1989;2:145-54. 15. Sander M, Fuchs K, High W, Hall K, Kreutzer J, Rosenthal M. The Community Integration Questionnaire revisited: an assessment of factor structure and validity. Arch Phys Med Rehabil 1999;80:1303-8. 16. Wood-Dauphinee S, Opzoomer A, Williams JI, Marchand BB, Spitzer WO. Assessment of global function: the reintegration to normal living index. Arch Phys Med Rehabil 1988;69:583-90.
17. Rappaport M, Herrero-Backe C, Rappaport ML. Head injury outcome up to ten years later. Arch Phys Med Rehabil 1989;70: 885-92. 18. McColl MA, Bickenbach J. Introduction to disability. London: WB Saunders; 1998. 19. Cohen S, Mermelstein R, Kamarack T, Hoberman HM. Measuring the functional components of social support. In: Sarason IG, Sarason BR, editors. Social support: theory, research and applications. Boston (MA): Martinus Nijhoff; 1985. p 73-94. 20. McColl MA, Skinner HA. Assessing inter- and intra-personal resources for community living. Disabil Rehabil 1995;17:24-34. 21. Nolte ML. Disablement, context and quality of life following brain injury: measuring what matters [dissertation]. Kingston (Ont): Queen's University Dept of Psychology; 1999. 22. Tabachnick B, Fidell L. Using multivariate statistics. Northridge (CA): Harper-Collins; 1989.
APPENDIX: COMMUNITY INTEGRATION MEASURE For each of the following statements, please indicate whether you agree or disagree: 1. I feel like part of this community, like I belong here. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 2. I know my way around this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 3. I know the rules in this community and I can fit in with them. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 4. I feel that I am accepted in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 5. I can be independent in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 6. I like where I'm living now. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 7. There are people I feel close to in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 8. I know a number of people in this community well enough to say hello and have them say hello back. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 9. There are things that I can do in this community for fun in my free time. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 10. I have something to do in this community during that main part of my day that is useful and productive. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree Coding: 5, always agree; 4, sometimes agree; 3, neutral; 2, sometimes disagree; 1, always disagree.
Arch Phys Med Rehabil Vol 82, April 2001
[] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree
ARTICLES
The Community Integration Measure: Development and Preliminary Validation Mary Ann McColl, Phi), Diane Davies, MSc, Peter Carlson, PhD, Jane Johnston, BA, BScN, Patricia Minnes, PhD ABSTRACT. McCol] MA, Davies D, Carlson P, Johnston J, Minnes P. The Community Integration Measure: development and preliminary validation. Arch Phys Med Rehabil 2001;82: 429-34. Objective: To present a new measure of commnnity integration, the Community Integration Measure (CIM), and to offer preliminary information about its psychometric properties. Design: Validation study. Setting: Community. Participants: Ninety-two participants placed in 3 subgroups (brain injury survivors, n = 41; significant others, n = 36; college students, n = 15). Main Outcome Measures: The distributional properties, factor structure, internal consistency reliability, content validity, discriminant validity, concurrent validity, and construct validity of the CIM. Results: All items correlated positively with each other and with the total score. Principal components factor analysis confirmed a 1-factor structure, which explained 44.1% of the variance. Inten]al consistency reliability, using Cronbach's alpha, was .87. Content validity was assured by the development procedure, correspondence with the theoretical model, and direct use of consumer language. Discriminant validity was supported by the CIM's ability to differentiate between subsamples. Criterion validity was supported by using correlations with the Community Integration Questionnaire. Construct validity was supported by correlations with the Interpersonal Support Evaluation List. Conclusion: The CIM offers a brieL easily administered measure of community integration that conforms to an empirically derived theoretical model and is psychometrically sound. Key Words: Brain injuries; Questionnaires; Rehabilitation; Social adjustment.
© 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation OMMUNITY INTEGRATION has been a focus for health service providers and researchers for the past 20 years, C since the World Health Organization highlighted community participation in its Health for All by the Year 2000. ~ And yet,
From the School of Rehabilitation Therapy (McColL Davies) and Department of Psychology (Minnes), Queen's University; and Regional Community Brain Injury Services (Carlson, Johnston), Kingston, Ont. Accepted in revised form February 7, 2000. Supported by the Ontario Ministry of Health. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Mary Ann McColl, PhD, Queen's University, School of Rehabilitation Therapy, Kingston, Ont K7L 3N6, Canada, e-mail: [email protected]. 0003-9993/01/8204-5804535.00/0 doi: 10.1053/apmr.2001.22195
20 years later, we still struggle with the definition of community and the meaning of community integration. 2 In rehabilitation, community integration has been conceptualized as the opposite of handicap, where handicap has been defined as the social disadvantage resulting from disability or impairment.3 5 Recent revision of the concept of handicap has led to the more universal notion of participation as a possibility for operationalizing integration. Participation is made up of 9 dimensions: personal maintenance; mobility; exchange of information; social relationships; home life and assistance to others; education; work; economic life; and community and civic life. 6 These dimensions of participation correspond closely with recent work on integration, primarily with developmentally disabled people. Halpern et al 7 identified 4 factors constituting community integration: occupation, residential environment, social support, and overall satisfaction. Bruininks et al 8 and McGrew and Bruininks 9 identified 3 other factors: leisure participation, family contact, and community assimilation and acceptance. This set of 7 factors was later confirmed by McGrew et al. 1° These investigations underline the multidimensional nature of community integration. Our own research lj used a qualitative approach to generate a definition of community integration. Between 1994 and 1996, we studied 18 people with moderate-to-severe acquired brain injuries for their first year in the community after inpatient rehabilitation. In monthly interviews, we discovered that they defined community integration as a function of 4 factors: assimilation (conformity, orientation, acceptance); social support (close and diffuse relationships); occupation (leisure, productivity); and independent living (personal independence, satisfaction with living arrangement) (fig 1). These factors are highly consistent with the research reported earlier. Community integration is arguably the ultimate aim of rehabilitation. Presumably, every person participating in rehabilitation hopes to be happily situated, productively occupied, and effectively supported in the community. However, confirming that community integration has been achieved at the end of rehabilitation has been hampered by a lack of consensus on definition and measurement of community integration. 1° MEASURES OF C O M M U N I T Y I N T E G R A T I O N The current standard for measuring community integration in the area of brain injury rehabilitation is the Community Integration Questionnaire (CIQ). Willer et aP .12 proposed a definition of community integration that took into account 3 dimensions: home and family, and social and vocational independence. They conceptualized community integration as the opposite of handicap, with emphasis on participation of the individual in his/her environment. Consistent with other research on community integration, they recognized that community integration cannot simply be observed, but must be understood in terms of consumer perceptions and the experience of integration? TM The CIQ is a 15-item measure designed for people with acquired brain injuries. It is divided into 3 subscales (home integration, social integration, productivity), Arch Phys Med Rehabil Vol 82, April 2001
430
COMMUNITY INTEGRATION MEASURE, MeColl
COMMUNITY INTEGRATION
/
ASSIMILATION • • •
SUPPORT
Acceptance
OCCUPATION •
Conformity Orientation
INDEPENDENT LIVING
•
Close
•
Productivity
Diffuse relationships
Independence Living situation
relationships •
•
•
Leisure
Fig 1. Theoretical model of community integration. Data from McColl et al. 11
which have been confirmed by using factor analyses. 15 The CIQ is widely used, and is valued for its quantitative properties and ease of administration. Scoring of the CIQ is based on a number of assumptions that may not apply in all situations. For example, scoring favors respondents with able-bodied friends over those with disabled friends; respondents who do things with friends rather than with family; and those who do household activities alone, rather than with others. Another option for measuring community integration is the Reintegration to Normal Living Index (RNLI). 16 The RNLI is made up of 11 items. The definition of community integration used for this measure was developed through consultations with consumers, family members, and health care professionals. The domain of the RNLI was derived by using a multistage process to identify domains and language for the concept. Items are scored with a 10-cm visual analog scale, anchored with the statements: "fully describes my situation" and "does not describe my situation." The RNLI is short, easy to administer, and widely used in rehabilitation research and practice. Pilot testing confirmed that it is conceptually and psychometrically sound. Rappaport et a117 developed another community integration measure for their study of 10-year outcomes after brain injury. They defined community integration according to 4 items dealing with: (1) the impact of physical impairment on living situation; (2) the impact of mental impairment on living situation; (3) the impact of mood on work; and (4) the impact of frustration tolerance on work. Rappaport presented normative data, but not psychometric. Another well known instrument to measure community integration is the Craig Handicap Assessment and Reporting Tool (CHART). 4 It is a 27-item scale that also defines community integration as the opposite of handicap. CHART items attempt to distinguish between levels of handicap, as outlined in the International Classification of Impairments, Disabilities, and Handicaps (ICIDH) (ie, ability to fulfill 6 community roles: orientation roles, physical independence, mobility, occupation, social integration, economic self-sufficiency). Most items have yes-no, forced choice, or other quantitative response formats, such as, "How many hours per week do you spend in home maintenance activities?" The definition of community integration used for the CHART and CIQ is the ICIDH 3 notion of handicap. When this notion was initially conceived, it was a useful and powerful way to understand the relation between disabled
Arch Phys Med Rehabil Vol 82, April 2001
individuals and their communities. However, many now consider the term unacceptable for 2 main reasons: (1) it places too much emphasis on the individual, and not enough on the community, as the locus of problems of community integration; and (2) it identifies people with disabilities as a special interest group, at the margins of society, rather than as full members of society entitled to all its rights and privileges. 18 A conceptual change is reflected in the revised ICFD, which is scheduled for final publication in 2001. 6 This reconceptualization of "disability" and "handicap" as "activity" and "participation" may require some rethinking of community integration measures. D E V E L O P M E N T OF T H E C O M M U N I T Y INTEGRATION MEASURE Although all the measures reviewed earlier have excellent reputations and properties, they do not use standard definition of community integration. This lack of consensus impelled us to undertake a 3-year study looking at intuitive definitions of community integration used by survivors of acquired brain injury in their assessments of their own integration. We identified 9 factors in that study, that mentioned previously, provided the basis for a measure of community integration--the Community Integration Measure (CIM).11 The CIM asks participants if they: (1) know the rules and how to follow them; (2) know their way around; (3) are accepted for who they are; (4) have people in the community with whom they feel close; (5) have relationships with different kinds of people in the community; (6) find things to do in their leisure time; (7) have something to do that makes them feel productive and worthwhile; (8) have some degree of independence; and (9) have a suitable place to live. The CIM is a 10-item, client-centered measure of community integration that uses the words of participants themselves, and makes no assumptions about the relative importance of particular activities or relationships. For example, unlike the CIQ, the CIM does not assume that independent participation is a sign of greater integration than is supported or mutual coparticipation. It can be administered either face-to-face, by telephone, or by the subject. It requires only a basic literacy level, and can be easily administered in 3 to 5 minutes. It has been used in practice with a broad range of clients with various disabilities; however, the present analyses focus on people with moderate-to-severe brain injuries. Presumably, this study population is more demanding in terms of both community inte-
COMMUNITY INTEGRATION MEASURE, McColl Table 1: Sample Characteristics and Subsample Comparisons Total Sample (n = 92) Age, mean ± SD (range) Gender Men (%) W o m e n (%)
College Students (n = 36)
Survivors {n - 41)
Family Members (n = 15)
32.1 _+ 13,8 35.4 ± 10.7 22.9 ± 4.1 55.2 -+ 13.3 (19-76) (20-62) (19-39) (26-76) 37 (40%) 55 (60%)
26 (58%) 15 (42%)
7 (19%) 27 (81%)
431
Table 2: Means -+ SDs and Distribution of CIM Scores n Total sample Survivors Coflege students Family members
Mean + SD
92 41 36 15
31.4 28.8 33.9 32
± + ± ±
7.1 7.7 4.4 8,4
Minimum
Maximum
9 9 21 20
40 40 40 40
4 (27%) 11 (73%)
gration and self-report measurement. The CIM results in a single summary score (between 10 and 50) that is the unweighted sum of the 10 items, each with 5 response options (see appendix). This study sought to offer preliminary information about the CIM and its psychometric properties, including its distributional properties, factor structure, internal consistency reliability, as well as its content, discriminant, concurrent, and construct validity. METHODS The study sample involved 92 participants, divided into 3 subsamples: (1) 41 participants in the original 3-year study, all of whom had moderate-to-severe acquired brain injuries ~~: (2) 36 community college students, recruited voluntarily; and (3) 15 family members, nominated by members of the acquired brain injury sample. Characteristics of the 3 samples are described in table 1. The following 3 measures were used in the validation study: the CIM, CIQ, and the Interpersonal Support Evaluation List (ISEL). The CIM, described earlier, is a brief, 10-item measure of community integration, developed on the basis of an empirically derived, explicit theoretical model. 1~ The CIQ, 15 described earlier, is a 15-item questionnaire with 3 possible response options for each question. It is well known in rehabilitation research and practice, particularly in the field of brain injury rehabilitation. The ISEL ~9 is a 40-item measure of social support that asks respondents about the availability of individuals to provide support. The ISEL has been used to measure social support among people with disabilities, including several items that address disability-specific issues (ie, transportation, assistance with activities of daily living, installation of household adaptations), ~° The adapted ISEL measures 3 types of support: informational (information, advice, guidance), emo-
tional (belonging, esteem support), and instrumental (practical, financial). This study complied with the standards of the Queen's University Research Ethics Bom'd. RESULTS Distributional Properties We examined item means and standard deviations (SDs), skewness and sensitivity, correlations among items, and correlations between items and the total score by using item analysis. All items were found to have means --_ SDs of between 2.86 ± .86 and 3.54 ± 1.62 (range, 1-5, with 1 = never, 5 = always). As regards sensitivity, though items tended to be skewed positively, only 1 item had more than 53% of participants choosing the top response option. (On item 8, "There are people I can say hello to . . . . " 72% of participants selected "always" as their response.) Table 2 shows the means and ranges for the total sample and for each of the subsamples. Notably, survivors had the lowest mean score, family members the greatest variance, and college students the least variance. All items correlated positively with each other and with the total score (table 3). Factor Structure Principal components factor analysis, using data from all 92 participants, resulted in a 1-factor solution that explained 44.1% of the variance in the 10 items of the CIM. The eigenvalues drop off dramatically after the extraction of the first factor (eigenvalues for the first and second factors, respectively: 4.40, 1.43) by using the scree-plot. Table 4 shows the factor loadings for the 10 items of the CIM, all of which are between .534 and ,797. Two-, 3-, and 4~factor structures were also explored, based on the underlying theoretical model discussed earlier and
Table 3: Correlations Between Items and With Total Score Item
1 2 3 4 5 6 7 8 9 10
Item 1
2
3
4
5
6
7
8
9
10
Total
1.00
.714 .634 .453 .659 .480 .416 .574 .545 .549 .579
1.00 .556 .368 .669 .301 .618 .469 .491 .358 .344
1.00 .397 .371 .561 .190 .441 .516 .313 .396
1.00 .404 .406 .027 .268 .173 .291 .381
1.00 .248 .495 .573 .314 .368 .430
1.00 .079 .172 .271 .362 .541
1.00 .371 .337 .308 .550
1.00 .476 .371 .336
1.00 .337 .339
1.00 ,550
Abbreviations: 1, I feel part of this c o m m u n i t y ; 2, I know m y w a y around; 3, I k n o w the rules; 4, I feel that I am accepted; 5, I can be independent; 6, I like w h e r e I'm living; 7, There are people I feel close to; 8, I k n o w people to say hello; 9, There are things I do for fun; 10, I have something useful to do.
Arch Phys Med Rehabil Vol 82, April 2001
432
COMMUNITY INTEGRATION MEASURE, McColl
Table 4: Factor Loadings of CIM Items on 1-Factor Solution (Principal Components Analysis) (n = 92) Item 1. I feel like part of this c o m m u n i t y , like I belong here. 4. I feel that I am accepted in this community, 2. I k n o w my w a y around in this community. 7. There are people that I feel close to in this community, 10. I have something to do in this c o m m u n i t y during the main part o f m y day that is useful and productive. 8. I k n o w a n u m b e r o f people in this c o m m u n i t y well enough to say hello and have them say hello back. 9. There are things that I can do in this c o m m u n i t y for fun in my free time. 5. I can be i n d e p e n d e n t in this community. 3, I k n o w the rules in this c o m m u n i t y and I can fit in with them, 6. I like w h e r e I'm living now.
,750 ,726 .687
Criterion validity was evaluated by using correlations of the CIM with the CIQ. 5 As mentioned, the CIQ is currently the standard for measuring community integration in brain injury rehabilitation. Our research found a correlation of .34 between the CIQ and the CIM (table 6). This suggests that there is some common variance between the 2 measures (12%), but that they do not duplicate each other's content. We believe the difference lies in the emphasis the CIQ places on independence.
.673
Construct Validity
Factor Loading ,797
.647
.637 .584 .653 .534
shown in figure 1. Only the 2-factor solution was readily interpretable, and produced eigenvalues greater than 1. The 2-factor solution explains 58.3% of variance in terms of 2 factors (table 5). After varimax rotation, the 2 factors each had 5 items loading on them, with loadings between .530 and .830. The 2 factors have been labeled: belonging and independent participation.
Internal Consistency Reliability Internal consistency reliability, using Cronbach's alpha, indicated a value of .87. Values above .80 are considered strong evidence that a measure is internally consistent; that is, that all items are measuring the same underlying construct. All items correlated positively with one another, and alpha value did not improve by excluding any single item. Subgroup alpha values were: survivors, .83: college students, .78; family members, .92.
Content Validity Content validity requires some assurance that the items on the measure cover the domain of the construct. In our case, content validity is assured by 3 factors associated with instrument development. First, the empirical, qualitative process leading to model development was well conceived and exhaustive11; second, the items on the CIM have a 1-to-1 relationship with concepts in the model (fig 1); and third, the items on the CIM were constructed by using the words of participants themselves, so to ensure authenticity and credibility.
Discriminant Validity Discriminant validity is supported by evidence that the measure distinguishes between groups that are expected to perform differently on the measure. In our study, we expected to attribute differences between acquired brain injury survivors and the other 2 groups to the disability. We used 1-way analysis of variance to compare CIM scores across the 3 subsamples. As expected, we found a significant difference between groups (F9J.2 : 5.5, p < .006), with the most marked contrast being between the acquired brain injury survivors and the college students (see table 2 for subgroup means). Arch Phys Med Rehabil Vol 82, April 2001
Criterion Validity
To show construct validity, one must show that CIM scores correlate in expected ways with theoretically related constructs. In our case, we expected that CIM scores would correlate positively with social support, occupation, and independent living--the 3 main factors that make up community integration, according to our model. 11 In the present study, we sought to confirm the first of these hypotheses by examining correlations between the CIM and the ISEL. ~9 We found a correlation of .42. showing that about 20% of the variance in the CIM is explained as a function of social support. This is consistent with the fact that social support items make up 2 of the 10 items on the CIM. The C1M also correlated with the informational and emotional subscales of the ISEL (table 6). DISCUSSION The study offers a new method for measuring community integration: the CIM. The CIM is based on an explicit theoretical model. It is client-centered, in that it assesses community integration from the client's perspective. It is accessible, in that it was constructed by using language derived from client interviews. It is brief, easy to administer (by interview or selfadministration), and easy to score. Finally, the CIM makes no assumptions about the relationship between independence and integration or about the relative value of different types of relationships.
Table 5: Factor Loadings of CIM Items on 2-Factor Solution (Principal Components Analysis With Varimax Rotation) (n = 92) Item Belonging like w h e r e I'm living now, feel like part of this c o m m u n i t y , like I belong here. feel that I am accepted in this community. There are people that I feel close to in this community. I k n o w a n u m b e r of people in this c o m m u n i t y well enough to say hello and have them say hello back. Independent Participation I can be i n d e p e n d e n t in this community, I have something to do in this c o m m u n i t y during the main part o f my day that is useful and productive. I know the rules in this c o m m u n i t y and I can fit in with them, I k n o w my w a y a r o u n d in this community. There are things that I can do in this c o m m u n i t y for fun in m y free time.
Factor 1
Factor 2
.830 .809 ,740
,302
.677 .475
.331
.830 .749
.671 .383
.656
.377
.530
433
COMMUNITY INTEGRATION MEASURE, McColl Table 6: Correlations Among the CIM, CIQ, and ISEL
CIM CIQ ISEL total ISEL instrumental ISEL informational ISEL emotional
CIM
CIQ
ISEL Total
1.00 .343 .425* .055 .380* .480*
1.00 .344 .264 .313 .287
1.00 .656* .856* .911'
ISEL I n st ru m e n t a l
1.00 .528* ,390"
ISEL Informational
ISEL Emotional
1.00
.640'
1.00
* p < .05.
* p < .01.
Psychometric findings to date offer assurances of the CIM's measurement properties. Item and scale analyses indicate that the construction and scoring of the measure are sound and reasonable. There is an underlying construct that can be measured with these 10 items that conforms to our definition of community integration. Further, enough shared variance exists between items that it makes sense to sum the 10 items to construct a single overall score representing community integration. The literature on community integration exhorts researchers to consider community integration a multidimensional construct that must be measured in depth and breadth, s.9 This study does not directly contradict that notion; in fact, the theoretical model on which the measure is built recognizes 4 categories of factors contributing to community integration (assimilation, occupation, relationships, living situation). However, our research does suggest that underlying that very complex construct is a unitary experience of belonging and participating. Thus, the CIM can be differentiated from other available instruments because it focuses on the underlying subjective experience of integration, rather than on the more objective or observable components of integration. The discriminant analysis has shown that the CIM can be used to measure community integration in the general population as well as in a rehabilitation population. Although the relative importance of particular items may change between different populations, the overall score of the 10 unweighted items remains viable and meaningful. Validity of the CIM is supported in our own work and independently. Nolte, 21 in a sample of 97 people with moderate-to-severe brain injuries, also found that CIM scores correlated to the CIQ (r = .26, p < .05) and to the 3 scales of the ISEL (instrumental support: r = .25, p < .05: informational support: r --- .26, p < .05; emotional support: r = .46, p < .001). In addition, Nolte found the CIM scores correlated to other model components, such as quality of life (r = .25, p < .05) and positive affect (Affect Balance Scale: r = .30, p < .05). Further studies of validity are planned to examine the relationship between the CIM and other model parameters, specifically, independent living and occupation. Our findings must be understood in the context of a relatively small sample for work of this nature. With 92 participants and a measure of 10 items, the standard of 10 subjects per variable in the factor analysis is not quite met, 22 though the sample base is close enough that estimates should be quite stable. Further, the present analysis is limited by the consideration of a single disability group and age--adults with acquired brain injuries. We have no evidence that the same properties could be expected with other age groups, particularly children, or with other disability groups. Future research is anticipated to address these issues. In addition, it would be necessary to
evaluate the extent to which the principle of universality is expressed in the CIM by testing it more extensively with general population samples. CONCLUSION The CIM offers clinicians a method for assessing the underlying experience of community integration and participation. It contributes to a full understanding of adjusUnent to community living after disability. The CIM offers a number of advantages--it is brief, easy to administer and score, and readily understandable by clients and families. On a more technical level, it is based on an empirically derived, theoretical model and has been shown to conform to that model and to show desirable psychometric properties.
Acknowledgments: We appreciate the assistance of our community partners--14 community residential programs for people with brain injuries in Ontario. In particular, we thank our local partner, Regional Community Brain Injury Services, in Kingston. References I. World Health Organization. Health for all by the year 2000. Geneva: WHO; 1980. 2. Jewkes R, Murcott A. Meanings of community. Soc Sci Med 1996;43:555-63. 3. World Health Organization. International classification of impairments, disabilities, and handicaps. Geneva: WHO; 1980. 4. Whiteneck GG, Charlifue SW, Gerhart KA, Overholser JD, Richardson GN. Quantifying handicap: a new measure of longterm rehabilitation outcomes. Arch Phys Med Rehabil 1992; 73:519-26. 5. Willer B, Rosenthal M, Kreutzer JS, Gordon WA, Rempel R. Assessment of community integration following rehabilitation for traumatic brain injury. J Head Trauma Rehabil 1993;8(2):75-87. 6. World Health Organization. International classification of functioning and disability-Beta-2 draft. Geneva: WHO: 1999. 7. Halpern A, Nave G, Close D, Nelson DJ. An empirical analysis of dimensions of community adjustments for adults with mental retardation in semi-independent living programs. Aust N Z J Dev Disabil 1986;12:147-57. 8. Brnininks R, Chen T, Lakin D, McGrew K. Components of personal competence and community integration for persons with mental retardation in small residential programs. Res Dev Disabil 1992;13:463-79. 9. McGrew K, Bruininks R. A multidimensional approach to measurement of community adjustment. In: Hayden M, Abery B, editors. Challenges for a service system in transition. Baltimore (MD): Brookes: 1994. p 65-79. 10. McGrew KS, Johnson DR, Brnininks RH. Factor analysis of community adjustment outcome measures for young adults with mild to severe disabilities. J Psychoeduc Assess 1994;12:55-66. 11. McColl MA, Carlson P, Johnston J, Minnes P, Shue K, Davies D, et al. Definition of community integration: perspectives of people with brain injuries. Brain Inj 1998;12:15-30. Arch Phys Med Rehabil Vol 82, April 2001
434
COMMUNITY INTEGRATION MEASURE, McColl
12. Willer B, Linn R, Allen K. Community integration and barriers to integration for individuals with brain injury. In: Finlayson MAJ, Garner S, editors. Brain injury rehabilitation: clinical considerations. Baltimore (MD): Williams & Wilkins; 1992. p 355-75. 13. Evans RW, Ruff RM. Outcome and value: a perspective on rehabilitation outcomes achieved in acquired brain injury. J Health Trauma Rehabil 1992;7:24-36. 14. Holosko MJ, Huege S. Perceived social adjustment and social support among a sample of head injured adults. Can J Rehabil 1989;2:145-54. 15. Sander M, Fuchs K, High W, Hall K, Kreutzer J, Rosenthal M. The Community Integration Questionnaire revisited: an assessment of factor structure and validity. Arch Phys Med Rehabil 1999;80:1303-8. 16. Wood-Dauphinee S, Opzoomer A, Williams JI, Marchand BB, Spitzer WO. Assessment of global function: the reintegration to normal living index. Arch Phys Med Rehabil 1988;69:583-90.
17. Rappaport M, Herrero-Backe C, Rappaport ML. Head injury outcome up to ten years later. Arch Phys Med Rehabil 1989;70: 885-92. 18. McColl MA, Bickenbach J. Introduction to disability. London: WB Saunders; 1998. 19. Cohen S, Mermelstein R, Kamarack T, Hoberman HM. Measuring the functional components of social support. In: Sarason IG, Sarason BR, editors. Social support: theory, research and applications. Boston (MA): Martinus Nijhoff; 1985. p 73-94. 20. McColl MA, Skinner HA. Assessing inter- and intra-personal resources for community living. Disabil Rehabil 1995;17:24-34. 21. Nolte ML. Disablement, context and quality of life following brain injury: measuring what matters [dissertation]. Kingston (Ont): Queen's University Dept of Psychology; 1999. 22. Tabachnick B, Fidell L. Using multivariate statistics. Northridge (CA): Harper-Collins; 1989.
APPENDIX: COMMUNITY INTEGRATION MEASURE For each of the following statements, please indicate whether you agree or disagree: 1. I feel like part of this community, like I belong here. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 2. I know my way around this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 3. I know the rules in this community and I can fit in with them. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 4. I feel that I am accepted in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 5. I can be independent in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 6. I like where I'm living now. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 7. There are people I feel close to in this community. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 8. I know a number of people in this community well enough to say hello and have them say hello back. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 9. There are things that I can do in this community for fun in my free time. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree 10. I have something to do in this community during that main part of my day that is useful and productive. [] Always agree [] Sometimes agree [] Neutral [] Sometimes disagree Coding: 5, always agree; 4, sometimes agree; 3, neutral; 2, sometimes disagree; 1, always disagree.
Arch Phys Med Rehabil Vol 82, April 2001
[] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree [] Always disagree