INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.

University M icrofilms International
A Bell & Howell Inform ation C om pany
300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600

O rder N u m b er 9 4 3 1 2 5 3

C onfigurations o f fam ily su p p ort services in M ichigan:
O rgan ization al, c o n te x tu a l, and a ttitu d in a l influences
H azel, K elly Lee, P h .D .
Michigan State University, 1994

UMI

300 N. Zeeb Rd.
Ann Arbor, MI 48106

CO NFIG URA TIONS OF FAM ILY SU PPO RT SERVICES IN M ICHIGAN:
ORGANIZATIONAL, CONTEXTUAL, AND ATTITUDINAL INFLUENCES

By

Kelly L. Hazel

A DISSERTATION

Submilted to
Michigan Stale University
in partial fulfillment o f the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Departm ent of Psychology
1994

ABSTRACT

CONFIGURATIONS OF FAM ILY SUPPORT SERVICES IN M ICHIGAN:
ORGANIZATIONAL, CONTEXTUAL, AND ATTITUDINAL INFLUENCES

By

Kelly L. Hazel

Examined the implementation of family support services in M ichigan, and tested the
hypothesis that both characteristics of the organization and its context, as well as attitudes
and beliefs of agency personnel (stakeholders) would be predictive of implementation. A
mail survey, sent to 55 mental health agencies, assessed stakeholders' (N = 339) attitudes
toward family support, their perceptions o f forces which encourage or inhibit
implementation, their decision making status, and what family support services were
provided at their agency. Information regarding agency size, and service area population,
density, wealth and demand for services was obtained from secondary data sources. A
Michigan Department of Mental Health study provided data concerning the level of
administrative formalization of family support at each agency and corroborative information
regarding the num ber of services provided.
First-order correlations, corrected for attenuation, showed that organization and service
area characteristics (/• = .23 to .46), and attitudes of stakeholders (aggregated to the agency
level) toward family support (/• = .34 to .41) were significantly associated with the num ber
of services but not formalization. Only stakeholders' perceptions of forces which inhibit or
encourage family support were significantly associated with both number of services (r =
.60) and form alization (/• = .34 to .49).

Although the decision m aking status of the stakeholders was correlated with their
attitudes and perceptions (r = . 13 to .56), contrary to expectations, there was no evidence
that status moderated the effects o f individual characteristics on the criteria. Multiple
regression analyses revealed that attitudes/perceptions (R = .69; uniqueness = .27) were
more predictive of the num ber of services than were context characteristics (R = .48;
uniqueness = .08). Only perceptions were predictive of form alization (R = .69).
Causal models, based on reasoned action and field theories, were tested using ordinary
least squares path analysis. Consistent with expected mediation effects, attitudes toward
family support predicted preferences for expansion/im provem ent of services ([3 = .76),
which in turn predicted the num ber o f services ({3 = .21). Also, perceptions of com m unity
support predicted agency com m itm ent ({3 = .73), which in turn predicted the num ber of
services (/? = .43). Context characteristics were significant predictors of the num ber of
services ((3 = .42), but not adm inistrative formalization. Stakeholder perceptions of
com m unity (j0 = .62) and adm inistrative support for services (J3 = .36) predicted agency
com m itm ent, which in turn predicted formalization (/?= .50).

To the women in my life
who have made my success possible
Thank-you

ACKNOWLEDGMENTS

I would like to thank my com m ittee members, Dr. Frank J. Floyd (chair), Dr. Esther
O. Fergus, Dr. Sandra E. Herman, Dr. Ralph L. Levine, and Dr. W illiam S. Davidson,
III, for their help and support. I would especially like to acknowledge the contributions of
my dissertation com m ittee chairperson, Dr. Frank J. Floyd. His advice and consistent
encouragem ent were extremely helpful in the birth process, which took significantly longer
than I anticipated. Thank you, Frank, for empathizing with my frustrations and pushing
me to get beyond my insecurities so that I may achieve my life's goals. I would also like to
thank Dr. Sandra E. Herman for her steadfast support and belief in my abilities, and her
suggestion that I stake out a section of the Department of Mental Health study, of which
she w as the principal investigator, for my dissertation research. Thanks also to my
colleagues at the Michigan Department of Mental Health for seven years of collaborative
work experience in program and policy evaluation. This research was sponsored, in part,
by a grant from the Michigan Developmental Disabilities Council, whose foresight and
innovative policies have encouraged and nurtured the expansion of family support services
in M ichigan. Thank-you all.

v

TABLE OF CONTENTS
List of T a b le s ............................................................................................................................. ix
List of F ig u res............................................................................................................................ xi
Chapter 1.................................................................................................................................... 1
I n t r o d u c ti o n .....................................................................................................................1
Historical B ackground................................................................................................ 3
The Innovation: Family Support Services.............................................................. 7
Barriers to Implementing Family S u p p o rt.......................................................8
Evidence of the Implementation of Family Support S ervices............................. 10
Focus o f the Study: Implementation of Family Support in M ichigan................13
Need and Rationale for the S tu d y .............................................................................17
R e s e a rc h

H y p o th e s e s ................................................................................................ 20

The Criterion: Innovation C onfigurations.............................................................. 21
O rganization and C ontext Predictors o f Innovation C onfigurations..............23
Organization size................................................................................................... 23
C o n te x t

c h a r a c te r is tic s .................................................................................24

Individual Level Predictors of Innovation C onfigurations................................. 26
Stakeholders...........................................................................................................27
Attitudes toward the innovation......................................................................... 28
B eliefs regarding driving and inhibiting forces......................................31
Organizational status and decision making involvement as m oderator

33

Relative Importance o f Organization, Context, and Individual
Characteristics........................................................................................................36
Path m odel..............................................................................................................38
S um m ary....................................................................................................................... 39

TABLE OF CONTENTS (continued)
Chapter 2 .....................................................................................................................................41
M ethod..................................................................................................................................41
Organization and Design o f the S tu d y .................................................................... 41
Research Setting

................................................................................................42

Participants.................................................................................................................... 42
R esponse

r a t e ..................................................................................................43

Respondent characteristics.................................................................................. 45
P ro ced u res....................................................................................................................47
M easurement.................................................................................................................47
Criterion variab les................................................................................................48
Agency and service area descriptors................................................................. 50
A ttitudes, beliefs/perceptions and behavioral intentions........................ 51
Role and decision-making status........................................................................67
Chapter 3 .....................................................................................................................................69
R e su lts..................................................................................................................................69
H ypotheses 1 and 2: A gency and service area predictors............................ 69
H ypotheses 3 and 4: A ttitude and belief predictors..................................... 71
Behavioral intentions........................................................................................... 72
Altitudes, beliefs/perceptions and behavioral intentions related to
role and decision making sta tu s.........................................................................74
Hypothesis 5: Decision making and role status as m oderator.............................75
O verall predictors o f fam ily support configurations.................................... 78
Hypothesis 6 ......................................................................................................... 78
Summary of hypotheses tests....................................................................................80
Beyond the hypotheses: Exploratory path model o f predictors..........................83
Path model of predictors o f num ber of se rv ic e s ............................................84
Path m odel of predictors of fo rm alizatio n ............................................. 86
Chapter 4 .....................................................................................................................................90
D iscussion............................................................................................................................90

TABLE OF CONTENTS (continued)
Appendix A: Stakeholder Survey and Accompanying L etters......................................... 102
Appendix B: UCR1HS Approval L e tte r............................................................................... 117
A ppendix C: P relim inary A nalyses R esults

T ab les............................................... 118

B ibliography...............................................................................................................................136

viii

LIST OF TABLES
Table 1:

Fam ily Support activities across the N a tio n ....................................................12

Table 2:

Family Support Services in M ichigan.............................................................. 18

Table 3:

R esponse

Table 4:

Agency level response rates and comparisons based

r a t e .. ...............................................................................................43

on agency characteristics..................................................................................... 44
Table 5:

Participants' current position at a g e n c y ...........................................................45

Table 6:

P articip an ts' length o f tim e at a g e n cy ....................................................46

Table 7:

Participants' length of time in current position .............................................. 46

Table 8:

Agency and service area descriptive statistics................................................ 51

Table 9:

Final principal axis factoring of attitude items after varimax rotation........ 53

Table 10:

Final principal axis factoring of belief/perception items after
varimax rotation.....................................................................................................55

Table 11:

Parent involvement in services planning attitude s c a le .................................58

Table 12:

Agency comm itm ent to hiring and training staff for family support
services behavioral intention sc a le .................................................................... 58

Table 13:

Expansion and improvement of family support services behavioral
intention s c a le ........................................................................................................59

Table 14:

Family and community support for family support services belief scale. .59

Table 15:

Agency capacity to provide family support services belief sc a le ................60

Table 16:

Positive attitude toward fam ily support services scale.......................... 60

Table 17:

N egative attitude toward fam ily support services scale......................... 61

Table 18:

Perception of staff support for family support services scale................ 61

Table 19:

Perception of administrative support for family support services scale .. .62

Table 20:

Attitude items dropped from scales.................................................................. 63

Table 21:

Belief/perception items dropped from s c a le s................................................. 64

Table 22:

Attitude, belief/perception and behavior intention scale statistics.............. 65

Table 23:

Altitude, belief/perception and behavior intention inter-scale
correlations corrected for attenuation................................................................ 65

Table 24:

Participants' perceptions of their role in agency decision m aking.......... 68

Table 25:

Percent of agencies where respondents indicated provision of services...70

LIST OF TABLES (continued)

Table 26:

Zero-order correlations between agency and service area characteristics
and the criteria........................................................................................................ 71

Table 27:

Corrected zero-order correlations between attitude, belief/perception
and behavioral intention scales and the criteria................................................73

Table 28:

Corrected zero-order correlations between decision m aking status
and role status, and attitudes, beliefs/perceptions and behavioral
i n te n ti o n s ............................................................................................................. 75

Table 29:

Status as moderator: results o f regression analyses for hypothesis 5 ...... 77

Table 30:

Predictors of num ber o f family support services. Results of stepwise
multiple regression analyses............................................................................... 79

Table 31:

Predictors of administrative formalization. Results of stepwise multiple
regression analyses...............................................................................................81

Table 32:

Review of hypotheses re su lts.............................................................................82

Table C - 1: Services identified as supportive of families o f people with
developmental disabilities.....................................................................................118
Table C-2: Descriptive statistics of attitude item s................................................................. 120
Table C-3: Descriptive statistics of perceptions o f support for services from others. .125
Table C-4: Descriptive statistics of belief/perception items regarding forces which
hinder or facilitate the implementation of family support se rv ic e s.............. 126
Table C-5: Principal axis factoring of attitude items after varim ax rotation............... 130
Table C-6: Principal axis factoring of belief/perception items after varimax rotation.. 132
Table C-7: Altitude factors IV and V and belief/perception factor V internal
reliability..................................................................................................................134
Table C-8: Correlation between criteria and items and factors not included in
final predictor scales.............................................................................................. 135

x

LIST OF FIGURES
Figure 1: General hypothesized path model of predictors o f family support
configurations........................................................................................................ 39
Figure 2:

Hypothesized path model of predictors of num ber o f different serv ices. .84

Figure 3:

Exam ined path m odel o f predictors of num ber of different services........ 85

Figure 4:

Hypothesized path model of predictors of administrative formalization . .87

Figure 5:

Examined path model of predictors of administrative formalization ......... 88

CH APTER 1
In tro d u ctio n

Implementation is the installation of new policies or practices in public and private
organizations (Yin, 1982). Research focusing on implementation has attempted to
document the results of policy changes and determine factors that affect the outcomes o f the
implementation process (Hasenfeld & Brock, 1991; Ingram & M ann, 1980; M ajone &
W ildavsky, 1978; Mazmanian & Sabatier, 1981; M cLaughlin, 1987; Sabatier, 1987a;
Scheircr, 1981). A common theme in the literature is that the implementation of innovative
policies is often frustrated by individual and organizational resistance and, consequently,
organizations do not always implement policies as intended (Alexander, 1985; Bardach,
1977; Blakely, M ayer, Gottschalk, et ah, 1987; Elmore, 1979, 1985; Fairw eather &
Tornatzky, 1977; Hasenfeld & Brock, 1991; Johnson & O'Connor, 1979; Kirsl & Jung,
1980; M cLaughlin, 1987; Sabatier, 1987b). Research has shown that variations in policy
wording and content, technical requirements of proposed changes, control and How of
resources, incentives, organizational structures, and bureaucratic politics have produced
wide variations in the implementation of proposed changes (Altewell & Gerstein, 1979;
Downs, 1976; Edwards, 1980; M azmanian & Sabatier, 1981, 1983; Van M eter & Van
Horn, 1975).
Theory and research has stressed the role of individuals responsible for implementing
changes both during the policy's framing (Sabatier, 1987a) and its implementation at the
local level (Browning, Marshall & Tabb, 1981; M cLaughlin, 1987). It has been argued
that a human service agency's 'ideological bent' has important implications for the type and

diversity of services it im plem ents (Frank & Davidson, 1983). In effect, it is not the
organizations that innovate or implement change, it's the individuals in the organizations
(M cLaughlin, 1987). Research on policy implementation (Berman & M cLaughlin, 1978;
Browning, M arshall & Tabb, 1981; M cLaughlin, 1987; Sabatier, 1987a; Scheirer, 1981)
and on innovation diffusion (Backer, Liberman & Kuehnel, 1986; Fairweather, Sanders &
Tornatzky, 1974; G laser, Abelson & Garrison, 1983; Rogers, 1983) has shown that the
results o f implementation are to a large extent dependent upon the skills, attitudes, and
behaviors of the individuals responsible at various levels in the implementing system.
Recently, innovative ways have been proposed for providing public support to families
caring at hom e for their family m em ber who has a developmental disability. Research has
documented a wide diversity in local agency responses to these proposed changes in public
support services for fam ilies (Agosta, Jennings & Bradley, 1985; Castellani, Downey,
Tausig & Bird, 1986; Herman & Hazel, 1987, 1991; Herman, Thom pson, Linklater &
Hazel, 1992; Ireys, Hauck & Perrin, 1985; Knoll, Covert, Osuch, et al., 1990; W eiss,
1989). Variations in local responses to family support have been related to resources,
incentives, and size and administrative structure of implementing organizations (Castellani
et ah, 1986; Herman & Hazel, 1991; Ireys et ah, 1985). Yet, none of the studies have
system atically examined the attitudes and beliefs of the individuals responsible for
implem enting family support programs at the local level. Nor, has previous research
evaluated the extent to which these individual characteristics explain variations in the
implem entation of family support services.
This research was geared toward filling that gap. As part o f a larger project evaluating
the implem entation of family support services in Michigan (Herman, Thompson el ah,
1992), attitudes and beliefs of local agency personnel were examined in order to determine
the extent to which these factors explain observed variations in family support services. In
general, this study evaluated the hypothesis that organizational performance is determined

3
by structural characteristics o f the organization and the context in which it operates, and by
characteristics o f organization personnel.
H isto rica l B ack grou n d
Traditionally, services lor persons with developmental disabilities have focused on the
habilitation of the individual with handicaps. Acceptance of the principles of normalization
and the right of people with developmental disabilities to live and work in the least
restrictive environment prompted changes in the service delivery system from institutional
based eare to comm unity and family based care (Landesman & Butterfield, 1987;
Sigelm an, Roeder & Sigelman, 1981). However, the reduction of institutional placements
and increased discharges to the community which were fostered by policies on
deinstitutionalization were not balanced by a comparable expansion of community
residential settings and support services (Moroney, 1986). In effect, the implem entation of
these policy changes challenged families to marshal the stamina, resources, and will to raise
and nurture a family m em ber with a handicap (Agosta & Bradley, 1985; M oroney, 1979,
1986; Seltzer & Krauss, 1984).
Families caring for members who are developmentally disabled confront challenges and
bear burdens unknown to other families. Among the challenges that parents of children
with disabilities face are the shock of the initial diagnosis and the consequent search for
understanding and information, the exhausting nature of specialized care and training,
family tensions aggravated by the fatigue of specialized care, and worries about the well­
being o f siblings and the future well-being of the child with a disability (Agosta & Bradley,
1985; Blacher, 1984; Seligman, 1983; Singer & Irvin, 1989). Other challenges include
persistent financial concerns related to the specialized care of the child with disabilities,
unpredictable crises, and the family's consequent involvement with an often unresponsive,
inefficient, uncoordinated, and dehumanized service system (Agosta & Bradley, 1985;
Blacher, 1984; Bruininks & Kranlz, 1979; Gallagher & Vietze, 1986; Singer & Irvin,
1989; Slater & W ikler, 1986). These factors may produce stresses with which many

4
families find it difficult to copc. Research suggests that families with members who have a
developmental disability, compared with other families, are likely to experience emotional
and physical stress, social isolation, reduced autonomy, and extraordinary financial, time,
and energy dem ands (Beckm an, 1983; Blacher, 1984; Breslau & Prabucki, 1987;
Cavanagh & Ashm an, 1985; Chelwynd, 1985; Gallagher & Vietze, 1986; Gallagher,
Beckm an & Cross, 1983; Kazak, 1987; M cKinney & Peterson, 1987; Quine & Pahl, 1985;
Seligm an, 1983; Sherman & Cocozza, 1984; W ikler, 1986).
Many factors determine the problems individual families may experience and the degree
to which they are able to adjust to their situation. These factors include the seriousness of
the fam ily m em ber's disability, the family's social and economic resources, and their
capaeity for coping with adversity (Blacher, 1984; Gallagher & Vietze, 1986; Singer &
Irvin, 1989). Research has also shown that formal and informal supports available to
families can increase their capacity to cope with the various demands and stresses (Agosta,
1989; Cohen, Agosta, Cohen & W arren, 1989; Dunsl, Trivette & Deal, 1988; Gallagher,
Beckm an & Cross, 1983; Perlman & Giele, 1983; Rowitz, 1985; Sherman & Cocozza,
1984; Singer & Irvin, 1989; Sum mers, Behr & Turnbull, 1989;Tausig, 1988). Social and
comm unity supports can help normalize the family environm ent and help to reduce the
stress experienced by families, thereby increasing their eapacity to cope. The support of
relatives, friends, service providers, and the comm unity can further help by enabling and
em powering families in their care giving roles (Dunst, Trivette & Deal, 1988; Dunst,
Trivette, Gordon & Plelcher, 1989).
For example, respite care (the provision o f periodic, tem porary relief of care giving by
sitters, in-hom e care workers, or a care agency) is seen as an essential part of the overall
support that families may need to keep their child with a disability at home (Cernoch, 1989;
Salisbury & Inlagliata, 1986). Research has found that respite care often plays a crucial
role in reducing family stress, improving the capacity of the family to care for their member
with a disability, improving the quality of family life, and preventing long term out-of­

5
hom e placem ents (Botuek & W insberg, 1991; Cavanagh & Ashm an, 1985; Factor, Perry
& Freem an, 1990; G rant & M cGrath, 1990; Halpern, 1985; Joyce, Singer & Isralowitz,
1983; Lulzer & Brubaker, 1988; M arc & M acDonald, 1988; Rimmerman, 1989; Upshur,
1982). By providing relief from the care giving situation, respite care helps reduce fatigue,
stress, and provide time for the family to take care of personal and family needs, such as
recreation and personal growth activities, that would otherwise be difficult, if not
im possible to do. Further, research has also found that the absence of comm unity-based
support services for families, such as respite care, is a significant contributor to a family's
decision to seek an out-of-home placement for their member with a disability (Bromley &
Blacher, 1989; Cole & M eyer, 1989; Sherman, 1988; Sherman & Cocozza, 1984).
However, even though public policy and service practices have led to an increased
prevalence o f family-based cane, families have received only sporadic public support
(Krauss, 1986). In many areas of the country, rather than providing the necessary
resources to support families in their care giving roles, public resources have been used for
services, such as foster care and comm unity group homes, that serve as substitutes for
fam ily-based care.
Part of the problem lies in the fact that the deinstitutionalization movement has stressed
the developm ent of comm unity based services to support individuals returning to the
community. The emphasis on individualized services has m eant that typically only the
child or adult with disabilities is the client of services and professional support. W hen
family m em bers are involved in services, they, as care givers, have been viewed as
resources to the person with disabilities, or as resources to the professionals by serving as
m em bers of the care giving team. In some cases, families have been viewed and treated as
the source o f the problem, one o f the reasons why the pathology exists. Rarely have
fam ilies been seen as needing supports and resources themselves (M oroney, 1979).
Recently, service professionals have begun to recognize the role of the family and the
fundamental importance of healthy family functioning in providing the individual with a

disability the best opportunities for personal growth and independence. Further, by
seeking information and control, and challenging authority in order to secure services to
m eet the needs o f their family m em ber with a disability, "parental entrepreneurs" (Darling,
1988) have advocated not only for the rights of their children but also for the rights of the
family. In effect, support services for parents are increasingly being championed and
implemented through legislative and policy changes. Yet, parents are frequently unaware
of services that are potentially available to them (Ayer, 1984; Ineichen, 1986; Justice,
O 'C onnor & W arren, 1971; Rubin & Quinn-Curran, 1984; W inton, 1986). Further, those
services that are available have often been unable to adequately match family needs.
Services have lacked individualization and flexibility, and have left parents powerless to
m odify the services that were offered (Agosta & Bradley, 1985; Herman, Thom pson et al„
1992; Krauss, 1986).
Families caring for members with disabilities at home affirm that support services
should be provided in a way that responds to the dynamic aspects of the family system
(e.g., changes over the life span, multiple roles of family m em bers), and responds to the
family as a unit rather than just the individual with disabilities. For them, services need to
be flexible and family-focused in order to match their family's needs throughout its life­
cycle (Byrne & Cunningham, 1985; Gallagher & Gallagher, 1985; Herman, Hazel &
M arccnko, 1989; Singer & Irvin, 1989; Turnbull, Summers & Brotherson, 1986). A
growing num ber of family advocates have articulated a need for programs where the locus
o f control rests less with the state and professionals and more with the family; where
"empowerment" (Katz, 1984; Rappaport, 1987) of families is a recognized and accepted
goal (Dunst, Trivette & Deal, 1988; Herman, Hazel & Marcenko, 1989; Herman,
Thom pson et al., 1992). In order to em pow er families, service program s should
encourage families' active participation in planning the service system, and render them
control over designing and selecting the services they receive (Agosta, Jennings & Bradley,
1985; Herm an, Thom pson et al., 1992).

7
The Innovation : Fam ily S u p port Services
Fam ily support, a relatively new movement in the broader social services arena, is a
m odel of service provision with the goal of providing comm unity resources for parents in
order to enhance their capacity in their child-rearing roles. By building strengths in families
rather than 'curing deficiencies', family support creates settings in which parents are
em pow ered to act on their own and their family's behalf (Kagan, Powell, W eissbourd &
Zigler, 1987; W eiss, 1988, 1989; W eissbourd & Kagan, 1989; Zigler & Black, 1989).
Family support includes programs and policies which span the human services
network. Relevant community-based programs include early intervention and parent
education services in schools (Halpern, 1986; Hinckley & Ellis, 1985; M eisels, 1989;
Roberts, W asik, Casto & Ramey, 1991; W iegerink & Comfort, 1987), home health aide
services in public health agencies (Ireys & Eichler, 1988; Ireys et al., 1985; Soyka, 1976),
and respite care services provided by mental health and developmental disability agencies
(Cohen, 1982; Grant & M cGrath, 1990; Salisbury & Intagliata, 1986; Starkey & Sarlie,
1989; Upshur, 1982). The diverse practices under the umbrella of "family support" are
linked by common principles that stress the importance of prevention, an ecological
approach to service delivery, and the universal value of social support (W eissbourd &
Kagan, 1989).
The present study focused on family support initiatives within the developmental
disabilities service system. Here, family support services foster family based care by
providing necessary social, educational, physical, and financial supports to sustain family
structure, maintain healthy family functioning, and reduce daily stress for families earing at
home for their member(s) with a developmental disability (Agosta & Bradley, 1985;
Krauss, 1986). The primary goal is to enhance the care giving capacity of fam ilies, thereby
preventing or delaying out-of-home placement of the family m em ber with a disability
(K rauss, 1986).

8
In contrast to child and individual oriented service provision, family support services
focus on the family as the unit to he supported and maintained. Consistent with family
system s theory (Turnbull, Brotherson & Sommers, 1985), family support services should
be constructed and implemented so as to allow for variation in family structure and
functions over the life cycle of the family. A family-centered approach reflects the premise
that the child is part of a family system, and that effective change for the child cannot be
achieved without concern for the family (Cohen et al., 1989).
Further, based on the ecological premise that factors outside the family affect the
family's capacity to nurture and rear its children (Bronfenbrenner, 1986), family support
services recognize the interdependence of the family and its community (Agosta, 1989;
W eiss, 1989). The ecological orientation requires that services be sensitive to the cultural
and social traditions o f the family and its community. It recognizes that building strengths
with one family will have a "ripple effect that strengthens a comm unity's collective
capacity" (W eissbourd & Kagan, 1989; p. 22). Ecological based approaches to family
support stress the principle that "natural" sources of support, including neighbors,
extended family members, friends, and community associations, can be more effective and
responsive to family needs than most government sponsored programs and professional
services (Taylor, Knoll, L e h r& W alker, 1989). Thus, professional support services
should, whenever possible, work to support existing social networks for families,
strengthen natural resources, and help build connections within existing community
resources (Center on Human Policy, 1987). Yet, in recognition of the isolation of many
families, professional or agency operated support services need to be available when
natural sources of support cannot meet the needs of families.
B arriers to Im plem enting Fam ily Support. Although the goals and underlying
service philosophy of family support are growing in acceptance, there is a wide diversity in
the extent to which support services have been implemented (Knoll et al., 1990). Agosta
and associates (Agosta, Bradley, Rugg, el al., 1985) identified three types of attitudinal

9
barriers which have constrained efforts to implem ent family support services. First, there
has been a lack of professional consensus as to the role of the family in providing care and
support. Some professionals discount the family's ability and capacity to make sound
decisions and to care adequately for their family member; often leading to recommendations
for out-of-hom e placements (Darling, 1983; M oroney, 1979). Other professionals do view
the family as competent care givers and see the positive benefits the family can provide.
They relate to family members as teachers, trainers and case managers for the individual
with disabilities, but fail to see family m em bers as needing supports themselves. This
often results in further expectations placed on the family by professionals to fulfill required
roles, with little increase in expectations placed on the service system and professionals to
support the family's efforts (Rubin & Q uinn-Curran, 1983).
Second, attitudes favoring fam ily-based care assum e a 'traditional' family system; a
family member, usually the mother, will remain at home to care for the person with a
disability. This assumption is inconsistent with current social and economic trends
(Agosta, Bradley, Rugg, et al., 1985). W ith high divorce rales and births out of wedlock,
the fastest growing family form is the single-parent family headed by a woman (Schroeder,
1989). Thus, family-based care is difficult for many 'fam ilies' to carry-out without
additional supports. Public support services have not kept pace with these changes and
families that do not fit the traditional mold have been penalized (Schroeder, 1989).
Third, Agosta and associates (Agosta, Bradley, Rugg, et al., 1985) point out that
society has yet to reach a consensus on the public's role in private family affairs. Much
public opinion holds that the State should assume responsibility to provide the necessary
resources for the care of the individual only after the family is no longer able to care, or
decides to relinquish their responsibilities by placing their family m em ber in an institution
or other out-of-home care situation. These altitudes are reflected in fiscal patterns which
focus most of the public service dollars on out-of-hom e care.

10
Im plem entation o f com prehensive support services for families would require a
significant re-allocation o f existing funds. However, because m any policy m akers believe
that current fiscal resources are scarce, and because there are significant pressures to
maintain current allocation patterns, they are reluctant to support further implementation of
family support program s. In contrast, those supporting fam ily-based care argue that the
family should be provided with adequate resources to live a life that is as close to normal as
possible. Support services that help the family to care for their m em ber at home are more
cost efficient in the long run than out-of-hom e placements (Agosta, Bradley, Rugg et al.,
1985).
E v id en ce o f th e Im p lem en tation o f F a m ily Sup p ort Services
It is clear that, due to various ideological and attitudinal barriers surrounding the
interplay of family versus public support, the establishm ent of an effective state wide
family support program can be a complex political and administrative task. Despite these
barriers, available evidence indicates that public officials are increasingly meeting the
challenge and are slowly beginning to clarify and implement family support initiatives. In
1984, 22 states had developed family support services (Agosta, Jennings & Bradley,
1985). A more recent national evaluation of stale and local efforts indicated that 41 states
had developed programs with a specific focus on supporting families who are caring for a
child with a developm ental disability (Knoll e ta l., 1990). Yet, there was a wide diversity
in the extent to which family support was firmly established in each state. Only twenty
stales had legislation which mandated some type of family support initiative, and at the
other extrem e, 10 other states were only running pilot projects.
There also existed a wide diversity in methods and models o f providing family support.
The num ber o f different types of support initiatives available in any one slate ranged from 0
to 27. These initiatives were broadly categorized into three activities: services, service
coordination, and financial assistance. Table 1 shows the different types of initiatives,
within each category, that were provided by family support programs. W ithin the services

11
category were initiatives which were provided to assist and enable the family to care for
their m em ber with a disability such as respite care, behavioral m anagem ent training,
attendant care services, and transportation. Other services were provided to help support
the family in dealing with stresses, such as family counseling, support groups, and stress
m anagement training; while others were focused on comm unity integration and
em pow erm ent such as parent to parent networking, and advocacy. Services that focused
on enabling the individual with disabilities to access his/her environm ent and become more
integrated into the community, such as adaptive equipment, home renovations, special
therapies, and recreation programs, were also seen as helpful for the family.
The second category of initiatives, lamily-hased case management, aims to empower
the family by connecting them with a person in the system who can help them advocate and
access the services in their community (Dunst & Trivelte, 1981). W orking in partnership
with the family, a family advocate or case worker can also help families coordinate the
multiple services, entitlements, and insurance programs with which they are involved.
The third form of family support is financial assistance programs. These programs
provide funds to the family so that the family can purchase needed services and supports
available from both private and public service programs. Financial assistance programs
vary in the amount of cash provided, eligibility standards, means of disbursem ent, and the
degree of control and flexibility a family has in determining how the funds are used
(Agosta, 1989). The m ost em powering programs are those, such as the subsidy program
in M ichigan (Arneaud & Herman, 1989; Herman, 1986, 1991; M eyers & M arcenko, 1989;
Parrot & Herman, 1987, 1988), which render to the family total control over the spending
decisions of the financial assistance provided.
Knoll and associates (1990) noted other differences in the way family support had been
implemented. Not only did the number of services vary, but the com binations of services
that had been implemented varied. Comparing across states, four general combinations of
family support initiatives were found: respite care only, other support services, financial

12
Table 1: Fam ily Support Activities Across the Nation.

SERVICES:
Respite and Child Care

In-home Assistance

Respite and sitter services

Hom emaker and chore services

Day care services

Attendant care services

After school care services

Home health care services

Environmental Adaptations

Recreation

Adaptive equipment

Sum mer camp programs

Home modifications

Community recreational programs

Transportation vehicle modifications

Com panion programs

Counseling and Peer Support

Extra-ordinary and Ordinary Needs

Family counseling

Transportation

Fam ily support groups

Dietary and clothing assistance

Peer support groups

Health and dental insurance

Parent to parent networking

Rent and utility assistance

Parent Education and Training

Systemic Assistance

Behavioral management training

Information and referral

Stress m anagement training

Advocacy

Parental skills training

Legal assistance

Traditional Developmental Disability Services
Speech, physical and occupational therapies
Individual counseling
Medical and dental care
Skills training
Evaluation and Assessment

CA SE M ANAGEM ENT and SERVICES COORDINATION

FINA NCIA L ASSISTANCE
Discretionary cash subsidy
Allowances and lines o f credit
Reimbursements and vouchers
Adapted from Knoll et al. (1990)

13
assistance only, and a combination of financial assistance and support services. Other
differences in implementation found across the Nation included variations in: structural and
administrative patterns, such as centralization of services and state versus local control over
resources; eligibility for services, based on the age of the individual with disabilities and the
severity of their disability; and the extent to which family support was consistently and
equitably available throughout the state.
In general, Knoll and his associates (1990) concluded that the family support efforts
they discovered were "few and small". The majority of states were only starting to explore
family support and, as such, had not yet made a full com m itm ent to supporting families.
Because, family support in m ost states had been sold to policy makers under the rubric of
'prevention' and cost effectiveness, the assistance that was provided was usually enough to
avoid costly out-of-home placements for some families, but not enough to establish a
family- and community-centered service system state-wide. The actual fiscal commitments
to family support were minute portions of the overall budgets allocated for services for
persons with developmental disabilities. The majority of funds were still largely absorbed
by substitute care programs. The basic values of family support, that of developmental
disability services as community-based and family centered, were only beginning to be
recognized in a few states.
F ocus o f the Study: Im p lem en tation o f F am ily S u p port in M ichigan
Michigan has consistently been identified as a national leader in family support services
(Agosta, 1989; C enter on Hum an Policy, 1987; Knoll et al., 1990; Krauss, 1986; Slater &
W ikler, 1986; Taylor et al., 1989). As early as 1978, Michigan was im plem enting family
support services. M ichigan's family support efforts are the outcom e of several forces
including effective parental advocacy, bureaucratic innovativeness, and legislative action.
First, in response to parental advocacy, funds were appropriated from the legislature
through the Department of Mental Health for a pilot project in Lansing. The family support
effort was subsequently expanded in the next fiscal year to include five additional

14
demonstration sites funded by the Services Research Division of the Department of Mental
Health (Herman, 1983; M owbray & Herman, 1991). Three of these demonstration sites
were able to successfully implem ent family support services and were continued through
1982 with support funds from M ichigan's federally m andated and funded Governor's
Council on Developmental Disabilities (DD Council). An evaluation (Herman, 1983)
indicated that throughout the implementation of the demonstration projects, none of the
families who participated placed their children in settings outside of their home (e.g.,
institution, foster care, or group home). Further, families viewed the services as a positive
sustaining force in their lives.
The positive results of the demonstration sites helped spur efforts to expand family
support across the state. In 1983, through advocacy of parent groups and key policy
makers in both the legislature and state bureaucracy, the Department of Mental Health
received funding authorization from the legislature to expand family support services. The
three successful demonstration sites along with the initial pilot site were recommended for
operational status. Further, the rem ainder of the State's 55 county-based comm unity
mental health boards were encouraged to adopt at least two components of family support
as part of their ongoing services: client services management and respite care. Between
1983 and 1986, the State provided $4.65 million to the boards to establish or expand
family support services (Herman & Hazel, 1991).
In response to the proposed expansion, the DD Council funded a project to disseminate
information regarding family support services in order to encourage the adoption and
expansion of family support services at the boards. The project was implemented by the
M ichigan Association of Community Mental Health Boards. The Association provided
descriptive pamphlets and other forms of assistance to the boards in order to aid them in
form ulating expansion requests, and planning and implementing family support services.
This project also organized monthly meetings of board staff who were assigned the task of
im plem enting and providing family support services. These meetings provided a forum for

15
information exchange and support and the beginnings of a network of informal technical
assistance that could be drawn upon when implementation difficulties arose. This group
was later formalized into a slate-wide "Family Support Coordinator's Council" and has
been instrumental in advocating for and developing new family support initiatives (e.g.
M edicaid policy changes).
Further impetus for family support services came from legislative authorization of the
Family Support Subsidy Program. Established as part of the M ichigan Mental Flealth code
(P.A. 249, 1983), families with children under age 18 who are severely m entally impaired,
severely multiply impaired or autistic impaired, and who have a yearly taxable income less
than $60,(XK), are eligible to receive a uniform monthly stipend to help them care for their
child (Michigan Department o f Mental Health [MDMH], 1986). The subsidy program,
administered by the Department of Mental Health, began operation in July 1984 and by
August 1988 had over 3300 families enrolled (Arneaud & Herman, 1989). Each family
received $256.74 per month per child with a disability.
In June 1984, in recognition of the growing demand for and implem entation o f family
support services across the stale, the Department of Mental Health amended its
administrative guidelines to establish policy and standards for the provision of family
support services (MDM H, 1984). Service and administrative components o f family
support were defined through the combined efforts of the DD Council and the Department
of Mental Health. Some of the components are similar to national trends (see Table 1)
while others are unique. The components include: consum er input through formal parent
advisory committees, case m anagement and advocacy, respite and sitter services (including
family friend respite services), information and referral services, parent training and
education, in- and out-of-hom e support services (e.g., chore, m edical, nursing, special
therapies), family and sibling counseling services, family peer support groups,
accom modative home improvements and adaptive equipment, crisis intervention placement,

16
individualized family service plans, and a cash subsidy program (M ichigan Developmental
Disabilities Council [MDDCJ, 1988; M DM H, 1988).
Family support services have continued to evolve in M ichigan. A comparison o f state­
wide evaluations of family support services pre- (Herman, 1984) and post-dissem ination
(Herman & Hazel, 1987) indicated an overall increase in service delivery across the state.
By 1985, all 55 boards had al least som e family support services available, and over 90%
of the boards provided case m anagement and respite care services (Herman & Hazel, 1987,
1991). However, despite this growth, the m ainstay of m ost program s was a group of
service components characteristic of traditional outpatient services (e.g., case management,
assessments, counseling, and psychiatric consultations) developed to support
deinstitutionalization and community placement, bolstered by some respite care (Herman &
Hazel, 1987). Few agencies had implemented programs with distinct family orientations.
Support services such as in-home nursing, parent or family support groups, parent training
and education, and home renovations were not widely available. It was concluded that
although resources remained a significant factor in implementing family support services,
this factor did not completely explain the continuing lack of family oriented programs.
Resource intensive services, such as case management, respite care, and assessments were
widely available. Yet, less intensive com ponents such as support groups, parent advisory
eomm ittees, and parent training classes were not widely available. The authors argued that
w hat was needed to increase the availability of family support was a shift in the philosophy
o f service providers to view services for families as supportive, rather than as treatment
(Herm an & Hazel, 1987).
Although no new mental health funds were specifically targeted for family support
services since the initial expansion funding in 1984-1985, several factors spurred the
continued adoption and implementation of family support across the slate. These factors
included changes in Department of Mental Health policy which reflected the growing
dem and for family support, the emergence of a state-wide Family Support Coordinator's

17
Council, and advocacy, evaluation, and demonstration project activities of the DD Council.
Results of a recent five year follow-up evaluation of the availability and accessibility of
family support services (Table 2) indicated some expansion in and refocusing o f service
components along with the development of programmatic administrative structures
(Herman, Thom pson, et al., 1992). As of 1990, all but one board reported having respite
care services available to families (Herman, Hazel, Thom pson & Linklater, 1992a), and 38
boards (69%) had implemented an identifiable family support program which included a
wide range o f service components under the auspices of a family oriented program
adm inistrative unit (Herman, Thom pson, et al., 1992). Further, 22 boards (40% ) had
added a parent advisory function to the agency's decision making process regarding family
support services (Herman, Hazel, Thom pson & Linklater, 1992b).
Nevertheless, despite continued growth, it is clear from this information that family
support is still not a fully implemented service philosophy. Sim ilar to variations across the
nation in family support service implementation, variations exist in local responses to
policy and advocacy incentives to implement family support in Michigan. Although family
support philosophy seems to have increased in acceptance with the implementation of
family oriented program structures and more parent advisory committees, the overall lack
of family oriented, low resource intensive service components remains.
N eed and R ationale for the Study
W hat factors account for the variability in the implementation of family support
services? Evaluation research regarding state and national implementation efforts points to
the role that resources, incentives, and various organizational characteristics play in local
configurations of family support services. Ireys and associates (1985), studying the
variation in slates' implem entation of 'Crippled Children's Program s', found that the
percentage of clients served was positively correlated to per capita expenditures (/• = .40, p
< .02). Relatively wealthier programs served more children. Further, regression analyses

Table 2: Family Support Services in Michigan.

# Agencies Providing
S erv ice C o m p o n en t

( N - 55)

Family case management

45

Family friend respite care

47

In-home respite (not family friend)

30

Out-of-home respite (not family friend)

41

In-home parent behavioral m anagement training

26

Parent education classes

20

Parent-to-parent networking

17

Newsletter

16

Parent or family support group

20

Family counseling

36

In-hom e nursing

16

In-home special therapies

9

In-home behavioral m anagement trainer

22

Adaptive equipment

16

Home renovations

9

Recreation or campership program

21

After school or day care program

4

A d m in istrative S tru ctu re
Program Structure

38

Parent Advisory Committee

22

Line Budget for Respite Care

31

From Herm an, Thom pson, et al. (1992)

19
suggested that programs located in wealthier stales were likely to provide m ore coverages
to a w ider range of disabling conditions (coefficients not reported).
Castellani and associates (1986), studying the availability and accessibility of family
support services in New York Stale, found that the greatest availability of family support
services was reported by government agencies. Sim ilar to Ireys and associates, larger
agencies (measured by budget size and num ber of clients) provided more services,
although not more types of services. Further, overall availability was related to population
density and county wealth; wealthy, urban counties reported more family support service
programs available. Further, a wider range of support services were available in the less
wealthy and rural counties. Flowever, these results were based on percentage
comparisons, no probability statistical tests were used to evaluate the observed differences.
Herman and Hazel (1991) evaluated the change in availability o f family support
services following a dissemination effort accompanied by a large influx of incentive
funding to Michigan's public mental health system. They discovered a general increase in
family support services across the state. In regard to the effect of funding on service
availability, agencies that had received lower levels of funding reported a significant
reduction in the variety of services provided. Agencies that received average to large
amounts o f funding had either increased the number of different services or had remained
stable. However, no statistically significant correlation was found between funding and
change in services implementation.
W hat is clear from a review of the scant research evidence regarding family support
services is that variations in implementation have been most often related to structural
aspects of the organization and community. However, methodological problem s exist in
that few researchers reported correlation or regression coefficients or used probability
testing to show that observed differences were statistically significant. Research has also
neglected an examination of the characteristics of the individuals involved in implementing
the services, such as their attitudes and beliefs. Research regarding the im plem entation of

20
family support services needs to take into account not only the structural and administrative
variables related to program characteristics, but also the role that the implementors play in
the design and delivery of service programs. In particular, research should focus on the
question o f what relationships exist between implementors' beliefs and altitudes and the
character of family support service programs they implement.
M ichigan has been the focus of a rather intense longitudinal evaluation effort focused
on the im plem entation o f family support services (Herman, 1983, 1984; Herman & Hazel,
1987, 1991; Herman, Hazel & M arcenko, 1989; Herman, M arcenko & Hazel, 1991;
Herm an, Thom pson, et al., 1992). Thus, focusing on M ichigan's fam ily support service
system provided an opportunity to integrate historical information with current evaluative
data, thus providing ample contextual information for understanding the relative importance
of individual and organizational characteristics for program implementation.

R esea rch H y p o th eses

The principle hypothesis explored by this research is that organizational performance is
determined by structural characteristics of the organization and its environm ent, and
characteristics of the individuals that make up the organization. In regard to family
support, variations in local responses to policy initiatives have been primarily related to
various structural characteristics of agencies and the context in which they operate. This
research proposed a new line o f investigation which included both structural characteristics
of the organization and its context, along with attitudes and beliefs of service agency
personnel.
The strategy used in this research was to examine a num ber of bivariate relationships
that have been found in previous research to be potential predictors of innovation
implementation, and then to re-examine these relationships within a multivariate
fram ework. Few phenomena are products of a single cause, including implementation.
Evaluating the effect of one variable without consideration for the effects other variables

21
may have on implementation, such as would be the ease using bivariate analyses, creates
bias in the research results. M ultivariate analysis, on the other hand, allows for the effect
of a particular variable to be made more certain, for the possibility of distorting influences
from the other variables can be accounted for (Lewis-Beek, 1980).
The following discussion presents definitions o f the criterion and predictor variables
and the specific hypotheses explored in this research. Also presented is a review of
relevant literature regarding the hypotheses. The chapter ends with the explication and
discussion of a proposed exploratory path model based on Lewin's (1951) field theory and
Ajzen and Fishbein's (1980) theory of reasoned action.
T he C riterion : Inn ovation C on figurations
Family support is a relatively new and innovative service philosophy within the human
services arena, thus the introduction of family support to the developmental disabilities
service system has necessitated a significant redirection in service philosophy, types of
services offered, and methods for providing those services. The observed variety of
approaches to family support are similar to the variety o f program adaptations that have
been found by researchers evaluating the adoption and implementation of innovations in
organizations (Blakely et al., 1987; G laser et al., 1983; Rice & Rogers, 1980; Rogers,
1983; Roitman, 1984; Scheirer, 1981; Van de Ven & Rogers, 1988). Evaluators have
discovered that during the course of implementation, a multitude of variables interact to
change not only the organization and its people, but also the innovative policy or program
that is being implemented. The term "innovation configurations" (Hall & Loucks, 1978)
was introduced to describe the variety of ways innovations are reinvented or adapted to fit
the needs of the implementing organization. According to Hall and Loucks (1978),
"innovation configurations are the operational patterns of the innovation that result from
selection and use of different innovation com ponent variations" (p. 9).
Research has shown that there is a tendency for adopters of an innovation to only
implement selected aspects of the innovation rather than the entire innovation (Hall &

22
Loucks, 1978; Rogers, 1983). Rogers (1983) suggested that characteristics of the social
innovation, such as whether it involves a single entity or a cluster of technologies, can
influence the extent to which it may be adapted. A narrowly defined and well integrated
innovation, in which all com ponents are highly interdependent, and which has achieved a
stable and comm only shared conceptualization, is least likely to be adapted (Rice & Rogers,
1980). In contrast, an innovation cluster which is less structured and defined, i.e. "loosely
bundled", allows for more flexibility in designing locally acceptable versions of the
innovation. Adopters can mix and match components of the innovation cluster to match
their needs and interests. Research has shown that modifications which fit the innovation
into preexisting patterns o f organizational behavior, can make it seem to be a less radical
change than it potentially is, and therefore more acceptable to the local system (Rogers,
1983).
Regarding family support, a num ber of services have been identified as components of
the innovation (see Table 1). These components are most characteristic of a 'loosely
bundled' technology cluster, and therefore family support in practice is highly adaptable to
local conditions. In M ichigan, Herman (1983) found that the initial family support
dem onstration projects represented varied configurations o f services. Even though case
m anagement and respite care had been implemented across all projects, the evaluation
identified four distinct models of service delivery that were based primarily on preexisting
organizational and comm unity service practices (Mowbray & Herman, 1991).
Information gathered by the five year follow-up study indicated that the number of
service com ponents (identified in Table 2) that were implemented at any one board ranged
from 0 to 15 (Herm an, Thom pson et al., 1992). Adm inistrative practices also varied. As
was mentioned previously, 69% of the boards had implemented a family support program
structure. Yin (1981) used the term 'routinization' to describe how innovations become
part o f an organization's standard practice, and thus are less likely to suffer cuts during
times o f financial exigency. Yin suggested that events such as the establishment of

23
appropriate organizational status for the program, hiring and training staff for the program,
and identification of local funding, among others, were needed to sustain any
organizational practice over time. As a relatively new service philosophy within the mental
health system, the extent to which family support has become routinized through various
administrative practices, such as the identification of a program structure and line item
budget for services, may be indicative of continued utilization.
For this research, the resultant character of family support, i.e., the criterion, was
operationalized by two specific configurations. The first criterion described the range of
different services provided. The other described the degree o f administrative formalization
of the services (similar to routinization), such as whether or not the agency had
implemented a program structure from which family support services were administered,
whether there was a line-ilem budget for respite care services, and whether the agency had
implem ented a parent advisory council for family support services.
O rgan ization and C on text P redictors o f In n ovation C on figu ration s
A variety of factors have been proposed and therefore are potentially applicable to the
prediction of innovation configurations. In general, the literature suggests that
organizational adoption and implementation of innovations is influenced by characteristics
of the organization, characteristics o f the context in which the organization operates, and
characteristics of individual people in the organization (Kimberly & Evanisko, 1981).
O rganization size. Various authors studying the diffusion of social and
technological innovations have argued that characteristics of an organization's structure are
related to innovation and program change. Innovation diffusion research has related an
organization's size, centralization, complexity, and formalization to its willingness to
innovate (Davis, 1982; Hage & Dewar, 1973; Havelock, 1971; Glaser et al., 1983; Moch
& M orse, 1977; Rogers, 1983; Scheirer, 1981; Zaltm an, Duncan & Holbek, 1973). The
size of the organization has m ost consistently been used a predictor of an organization's
willingness to innovate or adopt new programs (Glaser et al., 1983; Kimberly & Evanisko,

24
1981; Rogers, 1983; M och & M orse, 1977; Nord & Tucker, 1987). In general, larger
organizations often have access to more resources and the capacity to shift internal
resources in order to support new programmatic initiatives. The amount of organizational
'slack' in resources has been found to be positively related to innovation adoption (Rogers,
1983). Since large organizations tend to have more slack, they are in a better position to be
m ore innovative than smaller ones.
The size of an agency's budget or its expenditures have often been used in research to
operationalize the concept o f organizational size. As discussed previously, research
regarding family support services has found that expenditures were related to service
im plem entation (Caslellani et al., 1986; Ireys el al., 1985). However, the relationship
between size (as measured by budget size) and the breadth of services provided is not clear.
Ireys and associates (1985) found that larger agencies (based on program per capita
expenditures) provided a broader range of services (/■ = .19, n.s.), however, the
relationship was not statistically significant. In contrast, Caslellani and associates (1986)
found that size was not related to the number of services provided. The num ber of services
provided was more related to population density and service area wealth. Once again, this
finding was based on percentage comparisons, the differences were not tested for statistical
significance.
In M ichigan, funding patterns for mental health services vary across the organizations
responsible for implementing family support services. Some agencies have larger budgets
than others, allowing for larger staff sizes and more resources to support service provision
and administration. In this research, the agency budget was used as a measure of the
overall size of the agency. It was hypothesized that agencies with larger budgets would
have implemented more services and more formal administrative structures.
C o n te x t c h a ra c te ris tic s . The characteristics of the com m unity served by an
organization can affect its ability to implement innovative policies and procedures (Glaser et
al., 1983; Havelock, 1971). The comm unity can provide pressures and demands that

25
either encourage change in the organization or constrain innovative practices. However,
the research to this point is not clear as to the relationship between context characteristics
and implementation configurations. The importance of the context in which an organization
operates for its innovative practices has been acknowledged conceptually, but rarely
examined empirically.
Baldridge and Burnham (1975) examined the relationship between environmental
factors, such as population density (/• = .30), racial mix (/• = .25), urbanization (/• = .37),
wealth (/• = .06), and governm ent expenditures (/• = -.26), and innovative behavior of
educational organizations (statistical significance estimates not reported). They found that
organizations with heterogeneous environments (i.e., urbanized, racially mixed, dense
population) were more likely to adopt innovations than organizations with relatively stable,
hom ogeneous environm ents (r = .27). In comparisons of organization size/com plexity (R
= .49), and environm ental change (i.e., migration, growth, change in wealth and racial
mix) (R = .57), environmental variables explained 31% of the variance in innovativeness
(R = .56). However, when exam ined sim ultaneously with organizational variables (i.e.,
size and complexity), environmental variables (R 2 change = .08) explained only 8% more
of the variance already explained by the organizational variables (R 2 change = .24).
Kimberly and Evanisko (1981) studied the influence of individual, organization, and
context factors on hospital adoption of technological and administrative innovations. They
found that context variables, such as competition and population size, were predictive of
technological innovations (R = .55), but not of administrative innovations (R = .32).
W hen individual, context and organizational variables were considered simultaneously,
context variables were not predictive of innovativeness.
Context characteristics have also been related to the implementation o f family support
services (Caslellani et al., 1986; Ireys et al., 1985). Agencies in wealthier and more dense
service areas provided more services, while those in less dense and wealthy areas provided

26
a broader range o f services (Castellani et al., 1986). Once again, com parisons were not
tested for statistical significance.
In M ichigan, agency service areas vary in regards to population size and density,
economic vitality, and demand for family support services. These context characteristics
can potentially create a demand for and support the implementation of family support
services. Since the research literature is unclear as to the relative importance of
environmental factors in the determination of innovation configurations, demand for
services, service area population size and density, and service area wealth were evaluated in
order to determine the relative explanatory power of these context characteristics to family
support configurations. It was hypothesized that agencies with larger and more dense
populations, higher demand for services, and wealthier service areas would have
implemented more services and more formal administrative structures.
In d ivid u al L evel P red ictors o f In n ovation C o n figu ration s
Organizational decision making is in large part a function of individual decision
making, and thus, it is useful to look at how individuals within an organization evaluate the
innovation (Zaltman et al., 1973). Individuals responsible for implementing change act not
only from internal and external organizational incentives but also from professional and
personal motivation (M cLaughlin, 1987). Several researchers have argued that the
ideology, interests, and agendas of local actors have a major impact on implementation
(Berman & M cLaughlin, 1978; Browning et al., 1981; M cLaughlin, 1987; Zaltman et al.,
1973). Innovations are more acceptable, and thus, more likely to be implemented if they
appear to be com patible with the user's previously established values, norms, procedures
and resources (Glaser et al., 1983).
However, innovative practices are frequently in conflict with existing attitudes, custom s
and beliefs of agency personnel. This situation often discourages adoption and
implem entation of innovative programs. For an innovation to be accepted and thus
implemented, it must be assim ilated within the professional ideology and practice o f the

27
adopting organization's personnel. Innovations which run counter to the ideological
orientation of agency personnel require a change in their attitudes and beliefs before the
innovation can be fully implemented. Therefore, characteristics o f people, their social roles
and altitudes toward the innovation, may strongly affect the implementation process and
resultant innovation configurations.
Specific to family support, Knoll and colleagues (1990) and Agosta and colleagues
(Agosta, Bradley, Rugg, et al., 1985) argued that diversity in family support initiatives
may correspond to the extent to which service providers subscribe to values associated with
family-centered and family em pow erm ent oriented service provision. Taylor and his
associates (1989) argued that the degree to which service providers are comm itted to the
core values of a policy initiative largely determines the quality of the services available to
families. Yet, none of the empirical studies have systematically examined the beliefs and
attitudes of the individuals responsible for implementing family support programs at the
local level.
Stakeholders. The present study tested the general hypothesis that the character of
family support services at the agency level is influenced by the attitudes and beliefs of the
stakeholders in the implementation process at the local level. Stakeholders have been
defined as those "whose participation and cooperation is needed to operationalize a policy"
(Hascnfeld and Brock, 1991, p. 468). The stakeholders in the im plem entation o f family
support services are those who are in a direct decision making line regarding those
services, from the top o f the organizational leadership to the 'front line' workers. These
stakeholders include individuals directly responsible for the operation of family support
services, such as the family support and respite care coordinators, and those people
involved in the management of developmental disabilities services in general. Also
included in the stakeholder category are those people responsible for the decision making of
the organization as a whole, such as the agency directors and chairperson of the board.

28
A ttitudes toward the innovation. An individual's attitudes regarding a change
program influence that individual's willingness to carry out the necessary implementation
tasks (Scheirer, 1981). How individuals within the organization interpret their knowledge
and understanding of family support can greatly influence implementation. Backer and
associates (1986) noted that in mental health organizations, emotional reactions of
participants to a change program were critical to successful adoption and implementation.
Other researchers have noted that attiludinal resistance to the implementation of an
innovation may he so great that it is never fully or properly implemented (Fairweather et
al., 1974; G laser et al., 1983; Rogers, 1983).
Zaltman and associates (1973) argued that attitudes affect individuals' exposure to new
information, their perceptions of the innovation and change, and their actions toward the
change. Attitudes provide a cognitive fram ework from which a person interprets new
information, screens out unwanted information, and evaluates potential consequences of
the change, thereby helping the individual to determine which actions to take.
Consequently, attitudes held by organizational personnel can either huild and maintain
support for the status quo and thus lead to resistance to change, or can encourage
innovation and change.
Research and theory o f Ajzen and Fishbein (1980) suggests that intentions to perform a
particular behavior (i.e., to implem ent family support services) are a function o f two
constructs: attitudes toward the behavior and the subjective norm for that event. Their
theory of 'reasoned action' hypothesizes that a person forms an attitude toward the
behavior (i.e., the implementation of family support services) by subjectively evaluating the
consequences that are likely to occur if the behavior is carried out. If those consequences
are determined to he salient and positive, the likelihood of the person carrying out the
behavior increases. In general, their theory suggests that a person who believes that
implementing family support services will lead to mostly positive outcomes will hold a
favorable attitude toward family support, while a person who believes that implementation

29
will lead to mostly negative outcom es will hold unfavorable attitudes. As a result, attitudes
held by stakeholders determine their intentions, and thereby their behaviors, to implement
family support services.
The second determinant o f the behavioral intention is the person's evaluations of the
subjective norm for the behavior; in this case, their beliefs regarding the social pressures to
im plem ent or not implement family support services. I will discuss this factor, along with
the role of behavioral intentions, in the next section regarding beliefs about driving and
inhibiting forces.
Other research and theory has stressed the fact that people have a limited capacity for
absorbing change and therefore may resist change that seems to occur too fast (Legge,
1984). The degree to which agencies have implemented an innovative program at any
particular point in time is dependent on the amount of time needed for that organization and
its personnel to become com fortable with the necessary changes. Some organizations and
individuals are more readily accepting of change than others and, therefore, are more likely
to implem ent the changes at a faster pace. Thus, the resultant variability in family support
configurations, especially agencies which have implemented very few services, can also be
seen as a resistance response to change.
Sabatier and Mazmanian (1981) noted "the greater the amount o f behavioral change, the
more problematic successful implementation" (p. 9). Innovative policies which require a
rapid change in basic operations will meet more resistance than ones which can be
gradually implemented over time.

In agreement, Rogers (1983) noted that innovations

which can be viewed as "demonstrations" or "trials" are more likely to be implemented than
innovations which are seen as perm anent, unmodifiable changes. Innovations which can
be implem ented in small doses, thus allowing for new skills and procedures to be learned
and implem ented gradually, are less threatening.
Thus, implementation of a new policy or service delivery process, such as family
support, can be marked by the em ergence of reactions to the change in the form of

30
resistance. The extent of the resistance will be influenced by the amount of perceived
change required of the organization and its personnel and the degree to which personnel are
accepting of the changes (Legge, 1984). Attitudes of agency personnel toward changes
required to implement family support services may, therefore, have an impact on the
resultant configurations of services implemented at the agency. For example, the
implementation o f a parent advisory council requires changes in attitudes regarding the role
of parents in the service delivery process and also changes in behavior patterns when
interacting with parents. Parents can no longer be seen as passive recipients of services or
treated with indifference, but must be accepted as active partners in service delivery and
incorporated into the decision making process. Also, the creation of a program structure
may require a reorganization of the agency which could potentially shift the power structure
away from more traditional developmental disability services. Consequently, openness to
change by those involved in implementing new policies is critical to the success of family
support.
Little research, if any, has directly examined the relationship between attitudes and the
resultant implementation of innovative programs. Theories and research regarding attitudes
and reactions to change suggest that attitudes regarding the innovation and associated
changes may play an important part in the determination of implementation practices.
Therefore, attitudes of stakeholders regarding family support and associated changes, such
as parent involvement in decision making, were evaluated in order to determine their
relative importance to the explanation of family support services configurations. It was
hypothesized that agencies in which personnel expressed favorable attitudes toward family
support philosophy (e.g. family centered and driven services) and the implementation of
family support at their agency would have implemented more family support services and
more formal administrative structures.

31
B eliefs regarding d rivin g and inhibiting forces. Structural characteristics of
the organization and its environm ent are not static forces. W ithin a dynamic ecological
context these forces can directly inhibit or enhance the implementation of new programs by
providing the necessary resources and push to support the implementation of innovative
program s, or not. However, measures of organization and context elem ents that hinder or
constrain innovation implementation are not always readily available from secondary data
sources or easily obtained by direct m easurem ent. Even so, as Kurt Lewin's field theory
suggests (Lewin, 1951), behavior is a function, not of the objective environm ent, but the
subjective. A person's perception of their environm ent primarily directs their behavior.
Therefore, in order to understand how forces in the environm ent impact family support
services implementation behavior, it is important to understand the perceptions or beliefs
agency personnel hold in regard to those forces.
According to Lewin's field theory, agency structure and context forces not only have
direct impacts on implementation but also have indirect effects based on what agency
personnel believe to be the nature of the push and support for the implementation of the
innovative program. If personnel do not perceive a need for or believe that there are
resources to support the new program, it is unlikely that the organization will carry out the
necessary tasks to implem ent the program. For example, resources may be available to
develop the new program, but unless people in the organization believe a need exists they
may not search for and access those resources. In contrast, if people are aware of
resources and recognize that a need exists, they are more likely to implement the new
program .
Therefore, a potentially valuable means of measuring the effects of structural and
context forces is to ask people in the organization what they perceive to be the forces which
have either inhibited or encouraged the implementation of the innovative program in
question. In the five year follow-up study of family support services in M ichigan, service
providers and administrators identified a num ber of factors which they believed to inhibit or

32
encourage the growth o f family support services (Herman, Thom pson et al., 1992).
Identified factors included organization structure and context issues, such as funding,
demand, com m unity resources, need, and agency resources, and also characteristics of
agency personnel, such as their attitudes and perceived support from administration.
However, the open-ended question format did not allow for an exam ination of the extent to
which particular factors explained the variability in family support configurations.
Com parability across the agencies of the impact of any particular factor was not possible.
As noted earlier, Ajzen and Fishbein's (1980) theory of 'reasoned action' suggests that
a person's intentions to implem ent family support services are in part a function o f the
person's perception of the social pressures (i.e., the subjective norm) to im plem ent or not
implement family support services. During the implementation process, individuals seek
reinforcement for their actions by observing the attitudes and behaviors of their clients,
colleagues and supervisors (Rogers, 1983; Zaltman et al., 1973). This process of
legitimation helps the individual to determine the appropriateness of their actions, lending
support and motivation for further actions. For example, a person who perceives that
clients, colleagues and supervisors think family support services should be implem ented
will feel social pressure to do so. Conversely, a person who perceives that clients,
colleagues and supervisors think family support services should not be implem ented will
have a subjective norm that puls pressure on him/her to avoid implementing family support
services. Thus, a person's beliefs regarding the degree of support held by other agency
personnel for the implementation of an innovative program determine, in part, his/her
response to the situation. W ithout a belief that their actions will be supported, agency
personnel may waver and thus not implement family support to its fullest capacity.
Backer and colleagues (1986), in reviewing successful innovation dissem ination
efforts, noted that organizational support for the innovation was critical for successful
adoption and implementation. During the follow-up evaluation of family support services
in Michigan, family support coordinators identified support from agency administration as

a key factor that cither hindered or enhanced the implementation of family support services
(Herman, Thom pson et al., 1992). Although beliefs of agency personnel may be
predictive o f the outcom es of the implementation process, very few, if any, researchers
have systematically evaluated this possibility. Therefore, beliefs regarding the forces that
have inhibited or enhanced agency implementation o f family support services, such as
agency capacity, state funding, family and com m unity support, and support from
colleagues and supervisors, were explored in order to determine the relative importance of
these variables to the prediction o f family support configurations. In general, it was
hypothesized that agencies in which stakeholders' believed there to be sufficient capacity,
high levels of family and com m unity support, and support from staff and administration
would have implemented more family support services and more formal administrative
structures.
Further, as suggested by reasoned action theory, it was hypothesized that behavioral
intentions would be predictive o f family support configurations, and would m oderate the
relationship between attitudes/beliefs and the criteria. Thus, stakeholders' perceptions of
their agency's willingness to hire and train staff for family support and their preferences for
expansion and improvement of family support were measured. It was hypothesized that
agencies in which personnel favored the expansion and improvement of family support,
and in which staff were hired and trained for family support would have implemented more
services and more formal administrative structures. The effectiveness of behavioral
intentions as a m ediator between the relationship of attitudes and beliefs with the criteria
was explored in the path model, which I present in the last section of this chapter.
Q rgan ization ai statu s and d ecision m aking in volvem en t as m oderator.
One of the most frequently cited generalizations regarding resistance to innovation is that it
occurs when those affected by a change perceive it as threatening (Glaser el al., 1983).
Legge (1984) suggested, based on an exchange-theory approach, that individuals calculate
the balance of costs and benefits to themselves, and to the organization, and decide whether

34
to accept or resist the proposed change. Individuals who view the innovation as providing
more benefits to themselves and the organization will be more supportive of the
implementation process; while those who view the innovation as less than beneficial may
choose to resist, either passively or actively, efforts to implem ent the innovation. People
who have benefited the m ost from an existing order are unlikely to welcome a major
change. Fear of loss of status, prestige and power is often cited as a m ajor reason for
resistance to change (Glaser et al., 1983).
Power holders in the organization may resist required changes of an innovative policy
because of their "vested interest" in the existing order. Innovations which are perceived to
devalue a person's acquired knowledge and skills are threatening and therefore, less likely
to be fully implem ented (Glaser et al., 1983). Fairw ealher and his colleagues (1974), in
disseminating the Lodge program for people with mental illness, noted that innovations
which involved social status and role changes were less likely to be adopted or fully
implemented. This "top-down" (Sabatier, 1987b) perspective suggests that innovations
which threaten the established power hierarchy are not likely to be implemented without the
power hierarchy's acceptance and cooperation. Hasenleld and Brock (1991) noted that
when the group that controls the key resources in the organization is in conflict with the
proposed changes, the agency is more likely to undertake symbolic rather than substantive
implementation. The organization may comply with the formal requirements of the policy
but with minimal comm itm ent of organizational resources. Backer and associates (1986) in
their study of the dissemination and adoption of three innovative psychosocial
interventions, concurred that organizational support for the innovation, particularly from
top management, was critical for successful implementation.
However, from the "bottom-up" perspective, social innovation diffusion researchers
(Fairwealher et al., 1974; Berman & McLaughlin, 1977, 1978) and program
implementation analysts (Scheirer, 1981) have noted that participation o f direct service
providers in the decision making surrounding the implementation of innovative programs is

35
essential to effective implementation. The fundamental flaw in top-down models o f policy
implementation is that they assume that the framers of the policy decision are the key actors
and that others arc basically impediments (Sabatier, 1987b). Initiatives of local
implementing officials and other community resources are neglected in policy decisions,
often resulting in less than effective program implementation. Hjern and colleagues (as
reviewed by Sabatier, 1987b) in evaluating policy areas involving m any public and private
organizations, concluded that program success was more dependent upon the skills of the
individuals in local agencies than upon the efforts of central government officials.
This sam e 'bottom up' perspective can be applied to individual organizations.
Although organizational research has traditionally assumed organizations to be heavily
influenced by their leaders, Hagc and Dewar (1973) found that the beliefs of individuals
who reported high levels of participation in strategic decisions were more predictive of
innovative performance than the beliefs ol'lhe organization's formal leadership.
Fairw ealher and colleagues (1974), in disseminating the Lodge program for people with
mental illness, discovered that high social-organizational status of the person who is
contacted by the change agent was relatively unrelated to whether or not actual change
occurred. In actuality, contact with lower status agency personnel was related to greater
implem entation. In other words, the person at the top of the authority structure in an
organization does not necessarily have omnipotent power in determining program
innovation and implementation. Beliefs of all those in decision making roles are more
im portant in determining an organization's performance than the formal leadership alone.
Therefore, both the formal status of stakeholders and their perceived level of
involvement in decision-making regarding the implementation and provision of family
support services was determined. Analyses were conducted to determine the extent to
which attitudes, beliefs and intentions were correlated with stakeholders' status and levels
o f decision making involvement. Further, analyses were conducted in order to determine
w hether the predictive value o f attitudes, beliefs and intentions to observed variations in

36
family support configurations was moderated by status and levels of decision making. As
would be hypothesized by the 'lop down' perspective, a significant m oderator effect would
be indicative of the higher importance of the attitudes and beliefs of administrators and
highly involved decision makers to the implementation o f family support services.
R elative Im p ortance o f O rgan ization , C on text and Individ ual C h aracteristics
The arguments and literature thus far presented suggest that organizational performance
is determined by characteristics of the organization and it's environm ent, and characteristics
of people in the organization, their altitudes and beliefs and their relative status in the
organization. A number of variables within each of these categories (organization, context,
and individual factors) have been empirically evaluated and found to be related to
innovation adoption and/or implementation. There is little evidence, however, which
suggests which variables are most relevant in the prediction of implementation
configurations. Few studies have included variables within all three categories.
Baldridge and Burnham's (1975) research regarding organizational innovation included
measures of all three categories. However, although individual characteristics, such as
age, sex, cosmopolitanism and education were measured, they were not included in
comparison analyses. Also, administrative positions and roles, which were argued to have
an impact on perceptions of innovativeness, were not evaluated in regards to actual
implementation practices. The methodology and analysis strategy did not allow for a
determination of the relative importance of the three categories, as only organizational and
environmental characteristics were used as predictors of innovation. Also, individual
attitudes and beliefs regarding change or the innovation were not evaluated.
Kimberly & Evanisko (1981) also evaluated the influence of individual, organizational
and environmental factors on adoption of innovations. Their research concluded that
organizational level variables were the best predictors of innovation (correlation coefficients
ranged from . 15 to .39). Their research also suggested that different innovation
configurations, i.e., technological and administrative, were influenced by different sets of

37
variables. However, similar to Baldridge and Burnham (1975), the individual level
variables evaluated did not include altitudes and beliefs of organizational personnel in
regal'd to change or the particular innovations studied.
In contrast, Hage and Dewar's (1973) research, which focused on individual and
organizational variables, did include individual evaluations of change. Their research
suggested that individual values (partial correlation coefficients ranged from .70 to .75)
were more predictive of innovation implementation than organizational structure (partial
correlation coefficients ranged from .32 to .58). They concluded that the values of decision
makers "represent a guiding force in the organization, which can change its direction, set
policy, and introduce change irrespective of the structural constraints built into the
operations of the organization" (p. 286).
In order to add to the literature regarding the relative role o f individual versus setting
influences in the implementation process, the final hypothesis explored by this research
wits that beliefs and attitudes o f stakeholders would be more predictive o f family support
configurations than agency and service area characteristics. However, this debate does not
take into account the possible joint influence of individual and setting influences on
implementation. Research and theory regarding the adoption and implementation of
innovative services has suggested that both context and individual characteristics are
predictive of implementation. As discussed previously, Lewin's (1951) field theory
suggests that behavior, in this case implementation, is a function of both the person and the
environm ent, and, more specifically, the person's perceptions of driving and inhibiting
forces within the environm ent. Arising out of field theory, Ajzen and Fishbein's (1980)
theory of reasoned action suggests that behavior is a function o f attitudes toward the
behavior and the subjective norm for that event (e.g., perceptions of the social pressures
for the behavior) which lead to intentions, when then lead to praetice. In other words, an
individual's altitudes and perceptions of the amount of support in the environm ent for the
behavior informs their decisions regarding whether or not to carry out a particular behavior,

38
which in turn determines their actions. These two theories help to explain the ecological
context in which the implementation of innovations in mental health policy and service
delivery occur. W hen we com bine these two theories we can view implementation as a
function of three primary forces: driving forces external to the organization, organizational
personnel's attitudes toward the innovation, and their perceptions of the internal and
external norms for implementation. As suggested by reasoned action theory, the impaet of
the individual attitudes and perceptions/beliefs on innovation implementation will be
mediated by expressed intentions regarding implementation.
P a th m odel. In effect, a path model depicting the im plem entation of family support
services would have three prim ary paths: agency personnel's perceptions of support from
comm unity and peers, their attitudes toward family support, and agency/service area
demand characteristics. Further, it is expected that attitudes and beliefs, on average will be
correlated in agencies which have more fully implemented family support, since the norm
for the behavior will be positive, which in turn helps to facilitate positive attitudes, and vice
versa. Likewise, actual dem and characteristics and perceptions of demand will be
correlated. As shown in Figure 1, the path model which was developed and exam ined in
this research depicted family support configurations (i.e., the num ber o f services and
administrative formalization) as a function of agency and context characteristics and agency
personnel's altitudes toward family support and their beliefs/perceptions of the subjective
norm. Further, the relationship between family support configurations and attitudes and
beliefs was hypothesized to be mediated by expressed behavioral intentions regarding the
implementation of family support services. The model also depicts expected correlations
between attitudes and beliefs, and agency/context characteristics and perceptions regarding
the demand for services.

39

Perception/belief
regarding norm
Behavioral
Intention
Attitude toward
Family Support

Fam ily Support
Configuration

Agency/Context
Characteristics

Figure 1
General hypothesized path model of predictors of family support configurations.

Su m m ary
To summarize, family support services are a relatively new and innovative service
philosophy and practice within the human services arena. Docum ented evidence regarding
the implementation of family support initiatives within developmental disabilities services
has illustrated the wide diversity in service provider practices. Variations in local responses
to family support policy have been related to various internal and external organizational
characteristics. Little research to date has evaluated the impact of individual characteristics
on implementation. By studying the implementation o f family support services in
Michigan, this research evaluated the hypothesis that both structural characteristics of the
organization and its context, and characteristics of organizational personnel are predictive of
organizational performance. Specifically, it was hypothesized that larger agency budgets
and greater service area population size, density, wealth, and demand would be positively
related to more services and greater formalization. Also, positive altitudes toward family

40
support philosophy and implementation, and perceptions o f stakeholders of sufficient
agency capacity, and high levels of family and community support, and support from staff
and administration would be positively related to higher levels o f implementation (i.e.,
num ber of services and formalization). Organizational status of stakeholders and their
relative levels o f participation in decision making regarding family support services were
hypothesized to moderate the predictive value of stakeholders' attitudes and beliefs to
observed variations in implementation, such that the attitudes and beliefs of stakeholders
with a high level of administrative status and decision making involvement would be
predictive o f more services and greater formalization. It w as also hypothesized that
stakeholder characteristics would be more predictive of configurations than agency and
service area characteristics. And, finally, a theoretically driven path model was developed
to describe the hypothesized relationship between innovative behavior and structural
characteristics of the organization and its context along with characteristics of organizational
personnel.

CHAPTER 2
M eth od

O rgan ization and D esign o f the Study
The primary purpose of this study was to evaluate the general hypothesis that
organizational performance is determined hy structural characteristics of the organization
and the context in which it operates, and hy characteristics of organizational personnel. In
doing so, this research explored the relative importance of stakeholder attitudes toward
family support implementation and their beliefs regarding driving and inhibiting forces, and
characteristics of agencies and service areas as predictors of observed variations in the
implem entation family support services in M ichigan. The design of this study was crosssectional and multivariate. The hypotheses were evaluated using correlational and multiple
regression analyses.
The criterion variables were two configurations of family support services: the number
of different family support services implemented and the num ber of administrative
components implemented (i.e., formalization). Data were collected from a five year
follow-up to a longitudinal evaluation of the implementation of family support services in
M ichigan's community mental health system (Herman, Thom pson et al., 1992) and from a
mail survey conducted specifically for this study.
Explanatory variables hypothesized to be predictive of the criteria included individual,
organization and service area characteristics. Individual level data consisted of measures,
developed for this study, of stakeholders' attitudes toward family support philosophy and
implementation, beliefs/perceptions regarding driving and inhibiting forces, their expressed
intentions related to the implementation of family support, and their perceived role in

41

42
agency decision making. Organization level data included a measure o f the organization's
size based on the agency's total authorized budget. Service area data were obtained from
secondary data sources and included measures of the area's population size, density,
wealth, and potential demand for services.
R esearch S ettin g
M ichigan's com m unity mental health system is comprised of 55 county-based mental
health agencies. Each agency is responsible for serving the comm unity in their service area
which can range from 1 to 4 o f M ichigan's 83 counties. Thirty-nine o f the agencies serve
single counties.
Fam ily support services are those services which are provided to support families who
are caring at home for a family member with a developmental disability. Although the state
has issued policy guidelines regarding family support services (MDM H, 1988), each
agency is ultimately responsible for determining which of the family support services, if
any, they provide and how those services are implemented.
P a r tic ip a n ts
The participants were the stakeholders in family support services at the 55 community
mental health agencies. As staled in the introduction, stakeholders were defined as those
staff and administrative personnel who were in a direct decision making line regarding
family support services at the agency. They included individuals directly responsible for
the delivery and operation of family support services, such as the family support and
respite care coordinators, and family support case managers; and those people involved in
the m anagem ent o f developmental disabilities services in general (e.g., the developmental
disabilities services director and/or life consultation services director). Also included in the
stakeholder category were people responsible for decision making for the agency as a
whole, such as the executive administration and board comm ittee members.
Directed phone interviewing with family support services personnel, guided by
preliminary information gathered by the live year follow-up study, was used to generate a

43
pool ol' potential participants for the study. At least one person at each board was identified
for each of the following stakeholder categories: family support services personnel,
developmental disabilities services administration, agency administration, and board
chairperson. In larger agencies, the family support services personnel category may have
included more than one individual, such as a family support services coordinator, a respite
care coordinator, and/or a family support case manager. The developmental disabilities
services administration and agency administration categories also may have included more
than one individual depending upon the hierarchical nature and line of decision making at
the agency. The resulting pool of participants included 442 individuals: 237
adm inistrators, and 205 staff.
R esponse R ate. O f the 442 surveys sent out, 347 (79%) were returned, 8 of which
were returned unanswered or were otherwise not usable. As shown in Table 3, a higher
percentage o f staff than administrators returned the survey. O f the administrators,
developm ental disabilities services directors had the highest percentage response, followed
by agency executive level administrators. Board chairpersons had the lowest response rate.

Table 3: Response rate.

Recruited

Responded

N

%

Board chair or member

55

51

Executive level administration

78

64

Services level administration

104

77

Total administration

237

67

Total services level staff

205

87

Total overall

442

79

Stakeholders

44
Response rales also varied across agencies (see Table 4). Overall, 78% of the agencies
had m ore than 60% of their potential participants respond to the survey. Com parisons
(two-tailed M csts) between agencies with 60% or more and less response rates were made
to determine if there was any potential bias associated with the variability in response rates.
Agency response rales were com pared based on agency size (budget) and demand for
services (subsidy cases), and service area size (population and density) and wealth (average
personal incom e and governm ent spending per capita). As shown in Table 4, analyses
failed to detect any significant bias associated with agency characteristics.

Table 4: Agency level response rates and comparisons based on agency characteristics.

Agency Level Response Rates
Percentage of

Percent of
Agencies (N=55)

Participants Responded
0-59 %

22

6 0 -7 9 %

31

8 0 -1 0 0 %

47

Stakeholder response rate at agency was:

Characteristics

60% or less

More than 60%

A t 13
Mean (SD)

A - 42
Mean (SD)

t

2-tail
probability

Budget ($ mil.)

5.2 (4.2)

13.3 (38)

-.76

n.s.

D em and (subsidy cases)

45.5 (65)

77.4 (136)

-.81

n.s.

Service area population (per 1000)

117 (188)

185 (357)

-.65

n.s.

Service area density

193(394)

247 (552)

-.33

n.s.

13.6 (2)

14(3)

-.51

n.s.

1831 (378)

1642 (316)

1.80

n.s.

Average personal income ($ 10(H))
Per capita governm ent spending ($)

45
R e sp o n d e n t c h a ra c te ris tic s . As shown in T able 5, 29.7% o f the participants
indicated that they were directly involved in family support services provision or
administration. A minority of participants (20.1%) identified them selves as services staff,
but, in their view, were not directly involved with "family support". M ost of the
participants (50.1%) identified themselves as program or agency administration (board
m em bers included).

Table 5: Participants' current position at agency.

Participants
Position

Board committee chairperson or member

(/V=3 39)

8.4%

Community mental health agency or affiliate administration

19.5%

Service Program administration

22.2%

Family Support Services administration

6.3%

Family Support Services staff

23.4%

Services staff (not family support)

20.1 %

Most (80%) o f the participants had been employed with the agency for more than three
years (see Table 6), and most (61 %) had been in their current position within the agency
for more than three years (see Table 7). Slightly more than half of the participants (51%,
N= 174) indicated that they had a caseload of clients or families for which they were
responsible. O f these, 77% indicated that their caseloads consisted entirely o f individuals

46
Table 6: Participants' length o f time at agency.

Participants
{N=339)
Length of time at agency

%

Less than one year

5

1 to 2 years

14

3 to 5 years

25

6 to 10 years

24

M ore than 10 years

31

Table 7: Participants' length o f time in current position.

Participants
(W=339)
Length of time in position

%

Less than one year

10

1 to 2 years

27

3 to 5 years

33

6 to 10 years

17

M ore than 10 years

11

47
with developmental disabilities or their families. A small percentage of participants (13%)
indicated that members o f their immediate family had a developmental disability.
P roced ures
A mail survey was developed to measure altitudes, beliefs, and other characteristics of
family support stakeholders (see Appendix A). The survey, a cover letter introducing the
study, and support letters from M DM H were sent to identified stakeholders in March 1992.
A follow-up reminder letter was sent one week later. A second rem inder letter was sent
two weeks following the initial survey to those participants who had not returned their
survey. A third follow-up letter and a second copy of the survey were sent approxim ately
four weeks following the initial survey mailing to non-respondents.
Code numbers were used to keep track o f survey returns. Only the study director had
access to the coding system which was destroyed at the completion o f the data collection
phase. In accordance with the American Psychological Association's ethical guidelines
(APA, 1992) regarding the involvem ent o f hum ans in social science research, participants
were assured in the survey cover letter that their answers were confidential. The cover
letter also explained that their participation was strictly voluntary. The study was approved
by the University's Com m ittee on Research Involving Hum an Subjects (UCRIH S) prior to
data collection (sec letter in Appendix B).
M easu rem en t
In the survey, participants were asked to indicate what services and activities their
agency provided to support families (see question 1 of survey in Appendix A). Participants
were also asked to answer questions about themselves, such as the length of time they had
worked for the organization, their current position, w hether or not they had a caseload of
families for which they were responsible, and if they had a family m em ber with a
developmental disability (questions 10 through 14).
The majority of the questions on the survey m easured attitudes, beliefs/perceptions and
behavioral intentions of participants. Items for the belief/perception questions were

48
generated by a review o f responses to open-ended questions regarding the growth
promoting and inhibiting factors o f family support services from the five year follow-up
study (Herm an, Thom pson, et al., 1992). Experts in the field and the literature regarding
family support were consulted for service principles and philosophical themes for the
attitude measurements. In general, question items assessed live atlilude/bclief/intention
domains: beliefs/perceptions regarding the forces which inhibit or encourage the
implem entation of family support services (question 4), perceptions of others' acceptance
o f family support (question 3), preferences for change, such as expansion and
im provem ent of services (question 2 and 4), attitudes toward family support (question 2),
and perceived degree of involvement in agency decision making regarding family support
services (questions 6 through 9). All attitude/belief/intention questions were written in a
Likert-type form at (Likert, 1932) with five response categories. This allowed for the items
to be further combined into summ ated scales (Ghiselli, Campbell & Zedeck, 1981).
C riterion variables. Tw o criterion variables were used. The first was the num ber
of administrative com ponents implemented which provided a measure of the degree of
formalization of family support services at the agency. Data for this variable were obtained
directly from results o f the five year follow-up study (Herman, Thom pson, et al., 1992).
This 'baseline' survey was com pleted by the person at each board who was most familiar
with the board's im plem entation of family support services. In m ost cases, the respondent
was the family support coordinator.
Three administrative components were used to measure the extent to which the agencies
had formalized/routinized family support services: the presence of a "formally recognized
family support program", the presence of a "single line item budget" for respite care, and
the existence of a "formalized comm ittee which served as a link between the agency and
families of persons with a developmental disability" (i.e., a parent advisory committee).
Agencies were given one point for each component, for a range of scores from 0 to 3.

49
The second criterion was a measure o f the num ber of family support services the
agencies had implemented. Data for this criterion were obtained from the attitude survey.
Specifically, in the first open-ended question o f the survey, participants were asked to
indicate "what services and activities their board/agency currently provided to support
families who are caring at home for their member(s) who have a developmental disability?"
Answers were coded into 45 categories of service provision. See Table C - 1 in Appendix C
for results o f this question.
Often, staff and administrative personnel who were not directly involved in family
support services identified non-family support services. Therefore, many of the services
listed had not been previously listed as 'family support' in Departm ent policy or by
advocates of family support. In order to maintain content validity, only those services
which had been previously identified as family support and thus included in evaluations of
family support services were utilized in the 'services' criterion variable. Services included
in the measure were case management, respite care, family friend respite, parent education,
family support newsletter, family support groups, family networking, family counseling,
in-hom e nursing, behavior m anagement training, specialized therapies, adaptive equipm ent,
home renovations, recreation or campership programs, day care, and shared parenting. A
service was counted if at least one family support services staff or administrative person
identified it as provided at their agency. Agencies were given one point for each service
identified, for a range in scores from 0 to 16.
In order to assess the validity o f the services criterion measure, results were com pared
with information provided by the baseline survey conducted one year earlier. It must be
noted that these reports differed not only in time of assessment, but also in the procedures
used to collect the data. The attitude survey used a single question open-ended format,
while the baseline survey used a elosed-ended, multiple question form at to obtain yes or no
answers on specific services provided by the agencies. Responses to the present attitude
survey were coded, when possible, to be comparable to categories established by the earlier

50
baseline survey. The results from each survey were summ ed as a frequency count of the
num ber of family support service components, from 0 to 16, implemented at each of the
agencies. The correlation between the scores was .64 (p < .(Mil). Given the differences in
data collection and liming, the remarkable similarity in responses supports the validity of
the criterion measure used for this study.
A gency a n d service a re a d e s c rip to rs . Data pertaining to the characteristics of the
agency and its service area (i.e., context) were obtained from several sources. First,
m easures o f agency size were based on the agencies' overall budget. Budget allocation
information for fiscal year 1989-1990 was obtained from the Michigan Department of
Mental Health for each of the agencies.
Second, the num ber o f families receiving the family support subsidy was used to
determine the potential demand for family support services at each o f the agencies. The
subsidy enrollm ent data for fiscal year '89-90 were obtained from the Department of Mental
Health's Family Support Subsidy office. These data are highly reliable and are an exact
count of the numbers of families who received the family support subsidy. At a minimum,
these are the families who are most eligible for family support services. Families who do
not qualify for the subsidy, such as families with adult members who are disabled or
whose m em ber has a non-qualifying school assessm ent (other than SXI, SMI, Al), are
also eligible for family support services. However, accurate and comparable data regarding
these families are not available (e.g., school district service areas do not coincide with
mental health agency service areas). Consequently, the subsidy data were utilized as a
proxy measure of the demand for family support services, with a realization that the
demand is most likely much greater. The amount o f underestimation should be similar
across agency service areas.
Third, data pertaining to service area population size, density and wealth were obtained
from both the United Slates' Departm ent of Commerce, Bureau of the Census, and from
the Michigan Department of Commerce. All context descriptive data were based on county

51
statistics aggregated to the agency service area level. Population and square mileage data
were obtained from the 1990 census figures. The average personal income of county
residents in 1988 and governm ent spending per capita in 1987 were obtained from the
M ichigan Department o f Com m erce's (1990) compilation o f county descriptive statistics.
These data were used to determine the average wealth and governm ent resources of the
service area counties. As these data are primarily based on U.S. Census Bureau statistics
(according to citations noted in the data book), the reliability of the data is satisfactory (as
suggested by Stewart, 1984). Table 8 displays descriptive statistics regarding agency and
service area characteristics obtained from secondary data sources.
A ttitu d es, b e lie fs /p e rc e p tio n s and behavioral in ten tion s. S ta k e h o ld e r
beliefs/perceptions regarding the forces which inhibit or encourage the implementation of
family support services, others' acceptance of family support, their behavioral intentions,
and their attitudes toward family support philosophy and implementation were measured by
the mail survey. In order to reduce the overall num ber of variables, responses to 27
attitude and 31 belief/perception items (see Table C-2 through C-4 in Appendix C for item

Table 8: Agency and service area descriptive statistics.
Distributions
Characteristics

Mean

SD

low

high

Agency budget ($ mil.)

11.33

33.38

4.82

248.68

70

123

2

805

Service area population (per 1000)

169

325

5.76

2111

Area population density

234

516

6.37

3433

13,930

2663

9196

24,448

1686

338

971

2485

Dem and (subsidy cases)

Area avg. personal income ($)
Area government spending per capita ($)

52
descriptive statistics) were aggregated into a smaller number of scales utilizing standard
scale developm ent techniques (as suggested by DeVellis, 1991). First, all items were
submitted to an exploratory factor analysis (i.e., principal axis factoring with varimax
rotation) in order to identify common factors underlying the individual items. The results
yielded 13 factors with eigenvalues greater than 1 which accounted for 61 % of the
variance. These results also indicated that the attitude items formed 6 factors and the
belief/perception items formed 7 separate factors. Therefore, in further analyses the attitude
items were submitted independently of the belief/perception items.
Varimax rotated factor loadings from principal axis factoring were used to identify
potential items for each scale. Items which loaded .4 or higher on each factor were
grouped together. Items which did not load highly on a single factor were deleted and the
factor analysis was repeated. In the course of two analysis runs, 4 attitude items and 4
belief/perception items were deleted for these reasons (as indicated in Tables C-5 and C-6
in Appendix C). Tables 9 and 10 display the final results of the exploratory factor
analyses. The five attitude factors accounted for 57% of the variance prior to rotation. All
items loaded singly on one factor except for two items in the first altitude factor that also
loaded highly on the second factor. The six belief/perception factors accounted for 61% of
the variance prior to rotation. However, since no items loaded highly on the sixth
belief/perception factor, this factor was disregarded.
Confirm atory factor analytic techniques (as suggested by Hunter, 1980) were utilized to
further determine the factor structure of the two sets of items. Item-factor and inter-item
correlation analyses were utilized to determine the best item to factor inclusion (item
hom ogeneity) and, thereby, scale content validity (as suggested by Ghiselli, Cam pbell &
Zedeck, 1981). Examination of inter-item and item-factor correlations suggested that the
two attitude items which had double loadings in the exploratory analysis were highly
intercorrelated with and thus, better placed with two items of the second factor (items
identified with an A in Table 9) to form a sixth attitude factor (see Table 11). Further, two

53
Table 9: Final principal axis factoring of attitude items after varimax rotation.
Factors
I

II

III

IV

V

Eigenvalue (before rotation)

7.3

1.9

1.4

1.3

1.2

Percent o f variance (after rotation)

29.5

6 .0

3.5

3.1

2.6

JA
M
Al

-.25

-.14

.05

.13

-.31

-.05

.12

.20

-.17

-.09

.07

.10

.52

-.03

-.28

.24

.32

.52

-.26

-.39

.11

.10

-A9
-A6

.45

.12

-.09 -.03

Al

.14

-.17

.04

Al

-.37

-.27

.25

.26

-.19

.57

.08

.02 -.03

-.32

.50

.21

-.19 -.12

Supporting families should be top priority al agency

-.23

.29

-.29 -.10

Services more effective when family centered and driven

-.16

.12

-.12 -.07

Personally comm itted to FSS

-.20

AH
M
Al

.38

.08 -.19

AFamilies not capable care givers (R)

-.19

.46

.05

-.26 -.18

Like to help others implement FSS

-.09

.40

.24

-.09 -.08

Factor statistics

Items (abbreviated)
D on't think there is a need for FSS
Not convinced FSS is effective or appropriate
Like to see agency discontinue FSS
Families have help from others, mental health
should not be using resources too
Not strong enough need to warrant further
developm ent o f FSS
AParents views are important to services planning
AParents should be actively involved in services planning
Family support lakes up too much staff time
better devoted to other services
AParenLs are knowledgeable and competent resources
Fam ily support concepts should be agency's way of
doing business

54
Table 9 (cont'd).
Factors
Item s (abbreviated)_______________________

I

II

III

IV

V

Agency should redirect resources to provide more FSS

.08

.07 .75

-.17

.01

FSS should be greater percentage of agency budget

.23

.19 .66

-.13

.04

.08

.27 .61

-.18

.00

W ould like to see agency expand or improve FSS

.36

.28 .46

.16

.09

Focus on person with disability, not family

.11

-.12 -.19

j52 -.07

Services for individuals priority over services for family

.09

-.12 -.07

^49

FSS require too much change

.09

-.21 .09

-.03 M .

.15

-.03 -.02

.10 -.48

Staff resources should be redirected,
m ore em phasis on FSS

.15

A gency should find out more about FSS before
going further

(R) identifies items which were reverse scored for factor analysis
A identifies items which form ed a sixth factor in confirmatory analyses.

55
Table 10: Final principal axis factoring of belief/perception items after varimax rotation.

Factors
Factor statistics
Eigenvalue (before rotation)
Percent of variance (after rotation)

I

II

III

IV

V

VI

7.7

2.6

2.2

1.6

1.3

1.0

26.9 7.8

6.0

4.0

2.7

1.8

Items (abbreviated)

JJ

.22

.07 -.06

.21

JA

.03

.06

.22

.06 -.13

.02

.21

.09 -.12

.03

.03

.1 1 .30

.1 1 .05

.08 .09

.08 -.01

.06 .12

People at agency are not interested in providing FSS (R)

JA .11
JA .24
JA .33
J 1 .39
AA .19

-.03

.13

.19 .02

Providing FSS is a top budget priority

.58 -.00

.14

.31

.09 -.12

People at agency lack clear understanding of FSS (R)

A5

.1 1

.03

.19

.27 -.12

Self is supportive of FSS

.11

•74

.06

.1 1 -.12 .06

Agency's developmental disabilities staff is supportive

.22

•74

.06 -.02

FSS staff is supportive

.13

.73

-.03

.1 1 .05 -.09

Professional peers are supportive

.14

.68

-.07

.14

.16 .06

Staff supervised are supportive

.18

•64

.05

.14

.17 .22

Developmental disabilities administration is supportive

M

.59

.16 -.15

.09 -.07

Immediate supervisor is supportive

.28

AJ

.07

.21 -.09 .29

Agency lacks funding capacity (R)

.04

.11

Al

.09

.01 -.03

Agency has sufficient capacity

.07 -.02

.65 -.07

.09 .10

Agency lacks sufficient staff (R)

.12

.59 -.03

.06 -.11

A4

.02 .09

Board committee members are supportive

.24

Board committee members committed to
develop/expand FSS
Agency's administration is comm itted to
develop/expand FSS
Board chairperson is supportive
Agency's executive director is supportive
Agency's lop administration is supportive

.02

.19 -.17

Agency not able to keep up with demand,
families wail (R)

-.04 -.02

-.38

56
Table 10 (cont'd).
Factors
Items (abbreviated)

I

II

III

IV

V

Community agencies refer families to agency
.17

.09

-.07

.52 --.07

.05

.15

-.08

M

.13

Families active in advocating for FSS

.21

.01

.09

M

.1 4 ■

Fam ilies don't w ant FSS (R)

.01

.38

-.01

M

.21

for services
Families reluctant to take advantage of services
available (R)

Altitudes and practices in com m unity support individual
services, not family services (R)
System does not recognize family as client (R)

-.16 -.14 -.03 -.13 -AL-.27 -.00 -.15

.05 -.44

Lack of comm unity interest in supporting individuals and
families (R)

-.12 -.13 -.05 -.17 - A 4

(R) identifies items which were reverse scored for factor analysis.

57
of the belief/perception items which were originally deleted were found to be highly inter­
correlated. These two items were thus combined to form one of two behavioral intention
factors (see Table 12). The other behavioral intention factor was formed by items in factor
III o f the attitude item analysis (see Table 13). And last, after examining inter-item and
item-factor correlations, one item which was originally deleted from the fourth
belief/perception factor was returned (see item 5 in Table 14). Confirm atory analysis
indicated that all other item-laetor and inter-item correlations were consistent with expected
patterns (high item-factor and inter-item correlations, and low item to other factor
correlations), thus no other changes were made to the factors obtained during the
exploratory stage.
The resulting 12 factors (4 attitude, 5 belief/perception, 3 intentions) were then
submitted to internal consistency reliability analysis using Cronbach's alpha (Cronbach,
1951). Three factors (1 each of attitude, belief/perception, and intention) failed to achieve
reliability estimates greater than .60 and were thus deleted (see Table C-7 in Appendix C
for results of internal reliability and confirmatory analyses of deleted factors). Tables 11
through 19 present inter-item correlations and item-total correlations (i.e., results of
confirm atory analyses) for each of the nine remaining scales. Table 20 and 21 present
summ aries of the reasons for dropping items from the factor analyses and resultant scales
(see Table C-8 for correlations between these items/factors and the criterion variables).
Scale scores were computed as the average rating of the items in the factor, and ranged
from 1 to 5. Table 22 presents means, standard deviations and internal reliability for each
of the scales. Three attitude scales measured participants' altitudes toward family support
services and philosophical themes. Specifically, the support for parent involvem ent in
service/agency decision making scale (see Table 11) measured attitudes toward the core
philosophy of family support of parents as com petent resources and parent em pow erm ent
in decision making. Items in the positive attitudes toward family support scale (see Table
16) measured personal commitment to implementing family support along with altitudes

58
Table 11: Parent involvement in services planning attitude scale.

Inter-item correlations
Items (abbreviated)

1

2

Corrected

3

Item-total r

1. Parents views are im portant to services
planning

.52

2. Parents should be actively involved in
services planning

.55

.48

3. Parents are knowledgeable and competent
resources and should be involved
4. (R) Fam ilies not capable care givers

.39

.34

.29

.29

.48
.38

.42

(R) indicates items in this scale which were reverse scored.

Table 12: Agency com m itm ent to hiring and training sta ll lor family support services (FSS)
behavioral intention scale.

Inter-item
correlation

Corrected

1

Item-total r

Items (abbreviated)
1. Agency provided FSS staff training opportunities
2. Agency hired staff with FSS expertise

.55
.55

.55

59
Table 13: Expansion and improvement of family support services (FSS) behavioral
intention scale.
Inter-item correlations
1

Items (abbreviated!

2

3

Corrected
Item-total r

1. Agency should redirect resources to
provide more FSS

.65

2. FSS should be greater percentage of
.58

agency budget

.65

3. Staff resources should be redirected to
put m ore em phasis on FSS

.59

.42

.31

.52

.57

4. W ould like to see agency expand or
improve FSS

.32

.46

Table 14: Fam ily and com m unity support for family support services (FSS) belief scale.

Inter-item correlations
Items (abbreviated!

1

2

3

Corrected
4

Itern-total r

1. Families active in advocating for
expansion/developm ent of FSS

.34

2. (R) Families reluctant to take
advantage of services available

.17

.42

3. Com m unity agencies refer families
to agency for services
4. (R) Fam ilies don't w ant FSS

.21

.38

.22

.42

.27

.31

.20

.14

.36
.46

5. Sufficient resources in community
to support agency's efforts to
provide FSS

(R) indicates items in this scale which were reverse scored.

.3 0

.35

60
Table 15: Agency capacity to provide family support services (FSS) belief scale.
Inter-item correlations
Item s (abbreviated)

1

2

Collected

3

Item-total r

1. Agency has sufficient capacity

.53

2. (R) Agency lacks funding capacity

.46

3. (R) Agency lacks sufficient staff

.36

.44

.32

.18

.49
.43

4. (R) Agency not able to keep up with
demand, families have to wait

.18

.28

(R) indicates items in this scale which were reverse scored.

Table 16: Positive attitude toward family support services (FSS) scale.

Inter-item correlations
Items (abbreviated)

J

2

3

4

Corrected
Item-total r

1. Fam ily support concepts should be
agency's way of doing business

.56

2. Supporting families should be top
priority at agency

Al

.55

3. Like to help others im plem ent FSS

.35

.36

4. Personally com m itted to FSS

.39

.38

.33

.39

.35

.21

.43
.50

5. Services more effective when family
centered and driven

.33

.46

61
Table 17: Negative attitude toward family support services (FSS) scale.
Inter-item correlations
1

Items (abbreviated!

2

3

4

Corrected

5

Item-total

1. D on't think there is a need for FSS

.69

2. Not convinced FSS is effective or
appropriate
3. Like to see agency discontinue FSS

.63

.66

.50 .43

.54

4. Families have help from others, mental
health should not be using resources loo

.45 .49

.39

.6 0

.52 .48

.44 .45

.62

.49 .48

.35 .52 .40

.61

5. Do not see strong enough need to warrant
further developm ent of FSS
6. Fam ily support lakes up too much staff
time better devoted to other services

Table 18: Perception of staff support for family support services (FSS) scale.

Inter-item correlations
Items (abbreviated)

1

2

3

4

Corrected
5

6

1. Self is supportive of FSS
2. FSS staff if supportive

Item-total r
.67

.61

.63

3. Agency's developmental
disabilities staff is supportive

.54

.59

4. Professional peers are supportive

.46

.46

.62

5. Staff supervised are supportive

.47

.48

.46

.46

.47

.46

.62

.44

.40

.38

.23

.29

.39

.35

.72
.65
.59

6. Developmental disabilities
administration is supportive
7. Im m ediate supervisor is supportive

.59
.28

.42

62
Table 19: Perception of administrative support for family support services (FSS) scale.
Inter-item correlations
Items (abbreviated)______________

1 2

3

4

5

6

Corrected
7

8 Item-total r

1. Board com m ittee mem bers are supportive

.78

2. Board comm ittee members committed
to develop/expand FSS
3. Board chairperson is supportive

.58

.71

.84 .55

.76

.55 .68 .58

.72

.68 .52 .68 .52

.74

4. Agency's administration is com m itted
to develop/expand FSS
5. Agency's executive director
is supportive
6. Agency's top adm inistration
is supportive

.73 .52 .71 .55 .83

.76

.57 .46 .58 .52 .56 .55

.66

7. (R) People at agency not interested
in providing FSS

8. Providing FSS is a top budget priority .46 .62 .44 .61 .42 .45 .51

.62

9. (R) People at agency lack
clear understanding of FSS

.39 .44 .39 .39 .40 .37 .39 .37

(R) indicates items in this scale which were reverse scored.

.49

63
Table 20: Attitude items dropped from scales.

Attitude item

Reason for dropping

1 feel that programs and services for people

Internal reliability (alpha) of attitude factor

with disabilities should take priority over

IV which contained item was .55, below

support services for families

criteria of .60 for scale construction.

Family support services require too much

Internal reliability (alpha) of attitude factor

change in the way my board/agency

V which contained item was .54, below

provides services to our clients

criteria of .60 for scale construction.

There are needs in our community that are

Formed separate factor in initial factor

much more pressing which should be

analysis, did not load strongly on other

addressed first, before we develop family

factors, deleted from final factor analysis.

support services any further.
I feel that the focus of services should be on Internal reliability (alpha) of attitude factor
the person with disabilities, not the family

IV which contained item was .55, below
criteria of .60 for scale construction.

Many activities and services related to

Item failed to load highly on any factors in

family support are not within the domain of

initial factor analysis, deleted from final

mental health responsibility, other service

factor analysis.

providers should be providing these
supports
I think my agency should find out more

Internal reliability (alpha) of attitude factor

about family support services before we go

V which contained item was .54, below

any further

criteria of .60 for scale construction.

I would like to know what other agencies

Item failed to load highly on any factors in

are doing in regard to family support

initial factor analysis, deleted from final

services

factor analysis.

Since most families with children with

Item failed to load highly on any factors in

disabilities are supported by the school

initial factor analysis, deleted from final

system , it is not necessary for mental health

factor analysis.

to also provide services for them.

64
Table 21: Belief/perception items dropped from scales.

Belief/perception item

Reason for dropping

State Department funding allocations have

Item failed to load highly on any factors in

helped my agency's ability to provide

initial factor analysis, deleted from final

family support services.

factor analysis.

There is a general lack of community

Internal reliability (alpha) of belief factor V

interest and involvement in supporting

which contained item was .47, below

individuals with disabilities and their

criteria o f .60 for scale construction.

families
The system does not recognize the family as Internal reliability (alpha) of belief factor V
the client, only the individual with

which contained item was .47, below

disabilities, which hinders our efforts to

criteria o f .60 for scale construction.

implement support services for families
The attitudes and practices of m ost service

Internal reliability (alpha) of belief factor V

providers in our comm unity are that support which contained item was .47, below
services should be provided only to the
individual with disabilities, they do not
involve or support the family

criteria of .60 for scale construction.

65
Table 22: Attitude, belief/perception and behavior intention scale statistics.

Mean
# Items

Alpha

Mean(SD)

Negative attitudes toward FSS

6

.84

1.60(.48)

.47

Parent involvement in planning

4

.70

4.21 (.55)

.37

Positive attitudes toward FSS

5

.74

4.03(.58)

.36

Expansion and improvement of FSS

4

.77

3.76(.66)

.46

Agency commitment

2

.71

3.63(.96)

.55

Administrative support

9

.91

3.69C.72)

.54

Staff support

7

.85

4.52(.52)

.45

Agency capacity

4

.65

2.26(.85)

.32

Family/community support

5

.64

3.45(.66)

.26

Inter-item r

Attitude scales

Behavioral intention scales

Belief/perception scales

Table 23: Attitude, belief/perception and behavior intention inter-scale correlations corrected
for attenuation.
Correeled zero -order correlations
Attitude scales:

1

2

3

4

5

6

7

8

1. Negative attitudes to FSS
2. Parent involvement

-.78

3. Positive attitudes to FSS

-.76

.85

4. Expansion and improvement

-.58

.53

.69

5. Agency commitment

-.28

.20

.40

-.20

6. Adm inistrative support

-.14

.10

.21

-.24

.62

7. Staff support

-.54

.36

.43

.19

.40

.55

.03

-.04

-.10

-.30

.26

.14

.11

-.31

.12

.42

.10

.65

.47

.34

Behavioral intention scales:

Belief/perception scales:

8. Agency capacity
9. Fam ily/com munity support

W=339

-.02

66
favoring core family support concepts such as family centered and driven services. Items
in the negative attitudes toward family support scale (see Table 17) focused on the extent to
which participants disapproved of family support implementation, possibly because they
did not see a need for services, or were not convinced that family support was effective, or
that resources would be better used for other services. The means o f the attitude scales
suggested that the distribution of scores was som ew hat skewed to the positive end of the
scale.
Tw o m easures of behavioral intentions were developed, one associated with attitude
items and one associated with belief/perception items. The support for expansion and
im provem ent of family support services scale (see Table 13) measured preferences for
redirecting resources to provide more family support services. The second behavioral
intention scale measured perceptions of activities which have been carried out by the agency
which support the implem entation of family support, such as hiring and training staff for
family support services (see Table 12).
The four belief/perception scales measured stakeholders' awareness of the internal and
external forces which drive or inhibit family support services implementation.
Beliefs/perceptions regarding external forces were measured by the family and community
support for family support services scale (see Table 14). Belief/perceptions regarding
internal forces were measured by three scales, one which dealt with capacity issues (see
Table 15), and two which dealt with the subjective norm in the agency, i.e., support from
staff (see Table 18) and support from administration (see Table 19). Similar to the attitude
scales, the distribution of one o f the belief/perception scales, staff support, was somewhat
skew ed toward the highly supportive end of the scale.
Table 23 presents zero-order correlations corrected for attenuation between the nine
scales. In general, corrected correlations among the scales ranged from .02 to a high of
.78. The attitude scales were more highly inter-correlated than the belief scales. The two
behavioral intention scales were negatively correlated.

67
Role and decision m aking statu s. Survey participants identified their role within
the agency (question 11 of the survey). Responses to this question were condensed into an
index o f role status which included three levels: staff (50%), direct services administration
(22%), and agency level administration (28%). This variable was treated as an ordinal
scale m easuring participants' role status from lowest (staff = 1) to highest (agency
administration = 3).
Participants were also asked a series of four questions concerning their perceived
involvem ent in decision m aking at the agency. Tw o questions concerned the participants'
perceptions of their impact on, and frequency o f participation in "overall planning and
implem entation of services" at their agency. The other two questions explored their
perceived impact on, and frequency of participation in decisions regarding "services and
supports to individuals with disabilities and their families" at the agency. Table 24 displays
the results concerning respondents' reported role in decision making at the agency.
The four impact and frequency items were highly inter-correlated and, thus, were
com bined into one 'Decision M aking Status' scale score. The items were summed and the
mean calculated for the scale score. The alpha for this scale was .88, the mean inter-item
correlation was .64, and the corrected item-total correlations ranged from .59 to .80. The
overall scale mean was 3.25 (SD = 1.14).

68
Table 24: Participants' perceptions of their role in agency decision making.

% Response

Impact on:

Very little

Little

Some

Much

Very much

2.99 (1.47)

25

14

21

20

21

3.34 (1.29)

12

13

26

26

23

Mean (SD)

overall services
planning
DD services planning

Never Seldom

Frequency of
participation in:

Sometimes

Often

2_______ 3________ 4

Always

Mean (SD)

1

3.09 (1.37)

16

23

17

25

20

3.59 (1.16)

7

12

19

39

23

5

overall services
planning
DD services planning

N = 339

CH APTER 3
R e s u lts

The percent o f agencies where at least one respondent reported family support services
is shown in Table 25. Sim ilar to previous findings (Herman, Thom pson et al., 1992),
respite care and case m anagement were provided by almost all agencies, and daycare and
shared parenting were least often provided. The num ber of services provided at the
agencies ranged between 1 and 14, the mean and mode was 7 with a standard deviation of
3.
In regard to the level of formalization of services, 20% of the agencies had implemented
all three indicators (i.e., program structure, respite budget, parent advisory council), 38%
had implem ented two indicators, 29% had implemented one indicator, and 13% had not
implemented any of the indicators of administrative formalization. The average level of
form alization was 1.66 with a standard deviation o f .95.
H yp oth eses 1 and 2: A gency and Service A rea P red ictors
The first hypothesis stated that organizational size would be positively correlated with
higher levels of family support services implementation. Pearson correlational analyses
were used to lest this hypothesis with agency budget used as an indicator of agency size.
As can be seen in Table 26, agency budget was significantly positively correlated with the
num ber of family support services provided, but not with the degree of formalization of
family support within the agency. Thus, larger budgets were related to more services, but
not necessarily with a higher level of formalization. It should be noted that the num ber of
different services was positively correlated with formalization (/• = .36, p < .01),

69

70
Table 25: Percent of agencies where respondents indicated provision of services.

Percent o f agencies
Service________________
Case management
Respite Care

______ (N= 551
96
100

Family Friend Respite

53

Parent Education

55

Fam ily Support G roups

38

Parent-parent Network

35

Newsletter

26

Family Counseling

67

In-hom e N ursing

24

Behavior M anagement Training

69

Special Therapies

22

Adaptive Equipment

51

Home Renovations

33

Recreation or Campership

44

Daycare

9

Shared Parenting

2

suggesting that agencies with more services had implemented more formal administrative
structures.
The second hypothesis stated that service area characteristics (i.e., population, demand,
density, average personal incom e, and average governm ent spending per capita) would be
positively correlated with higher levels of family support services implementation. Pearson
correlational analyses were used to test this hypothesis. These results are presented in
Table 26. As predicted, three service area characteristics (i.e., population size, demand,
and density) were significantly positively correlated with the num ber o f family support
services provided. Larger service area populations, higher levels o f population density.

71
Table 26: Zero-order correlations between agency and service area characteristics and the
criteria.

Zero-order Correlations
Variables_______________

1

2

3

4

5

6

7

1. Agency budget
2. Dem and (subsidy enrollment)

.90**

3. Service Area Population

.89** .98**

4. Service Area Density

.90** .96**

.98**

5. Area Avg. Personal Income

.24*

.48**

.53** .48**

6. Service Area Governm ent

.29*

.28*

.26*

.23*

.13

Spending Per capita
7. # F am ily S u p p ort S ervices
8. D egree o f F orm alization

.37* * .46* * .39* * .36* * .23*
.10 .19
.1 5
.09
.03

.06
.07 .36* *

N=55

* p < .05, ** p < .01, one-tail.

higher demand lor services were associated with agencies implementing more family
support components. Average personal income and average governm ent spending per
capita were not significantly correlated with number o f services. Further, none of the
service area characteristics were significantly correlated with the degree of formalization.
Therefore, hypotheses 1 and 2 were upheld for the services criterion, but not for the
administrative formalization criterion. Service area characteristics were significantly
correlated with the number of services, but not formalization.
H ypoth eses 3 and 4: A ttitu d e and B elief P redictors
The third and fourth hypotheses stated that positive attitudes toward family support
philosophy and implem entation, and perceptions of stakeholders of sufficient agency
capacity, and high levels of family and community support, and support from staff and
administration would be positively correlated with family support services configurations.

72
For these analyses, the seale scores of all stakeholders of an agency were aggregated and
the mean computed to establish the agency scale score. Pearson correlational analyses were
used to determine correlations between the aggregated altitude, belief/perception and
behavioral intention scales and the criteria. Correlation coefficients were then corrected for
attenuation and statistical significance determined based on 95% confidence interval
estimates. As shown in Table 27, two attitude measures (negative attitudes toward FSS
and positive attitudes toward family support) were significantly correlated (i.e., confidence
interval did not include 0) in the expected direction with the number of family support
services implemented. Higher levels of positive attitudes toward services, and lower levels
o f negative attitudes toward services were related to greater numbers of services. No
attitude measures were significantly correlated with administrative formalization. Thus, the
hypothesis that attitudes would be positively correlated with family support configurations
was upheld for the services criterion, but not for the formalization criterion.
Regarding beliefs/perceptions, the results presented in Table 27 indicated one
belief/perception measure, family and community support, was significantly correlated in
the expected direction with the number of family support services provided. This measure,
together with perceptions of administrative support for services, was also correlated
positively with degree of formalization. Thus, beliefs/perceptions were positively
correlated with both of the family support configurations, upholding the third and fourth
hypotheses.
B ehavioral intentions. As shown in Table 27, preference for expansion and
im provem ent of family support services was positively correlated with num ber of services.
Further, perception of agency com m itm ent was significantly correlated, in the expected
direction, with both criteria. Perceptions of the agency's willingness to hire and train staff
for family support was related to both a greater number o f services provided and more
formal administrative structures.

73
Table 27: Corrected zero-order correlations between attitude, belief/perception, and
behavioral intention scales and the criteria.

Corrected zero-order Correlations
(± 95% confidence interval)
Attitude scales

1

2

3

4

5

6

7

8

1. Negative attitudes to FSS
2. Parent involvement

-.72
(.26)

3. Positive attitudes to FSS

-.74
.93
(.22) (.20)

Behavioral intention scales
4. Expansion/im provem ent

.54
.76
-.70
(.22) (.30) (.24)

5. Agency commitment

-.61
.69
.45
.05
(-28) (-34) (.28) (.38)

Belief/perception scales
6. Administrative support

.04
-.11
.09 -.44
.55
(.30) (.36) (.34) (-28) (.28)

7. Staff support

-.64
.45
.47
.11
.68
.53
(.24) (-32) (.30) (-32) (.26) (.26)

8. Agency capacity

.04 -.08
-.04
.10
.32 -.16
.11
(.38) (.42) (.40) (-40) (-38) (-34) (.38)

9. Fam ily/com munity support

-.57
.16
.71
.48
.30
.42 -.22
.73
(.30) (.38) (.30) (.34) (.30) (-34) (.32) (.40)

Criterion
.32 .44 .21 .23 -.19 .60
#F am iIy S u p p ort S erv ices - . 3 4 . 3 0 .4 1
(.28) (.32) (.28) (.30) (.28) (.28) (.28) (.32) (.26)
D egree o f F orm alization

N = 55

-. 2 0 . 1 2 .21 -.11 .5 0 . 3 4 .2 4 . 1 6 . 4 9
(.28) (.32) (.30) (-30) (-26) (.26) (.28) (-32) (.30)

!

74
A ttitu d es, b eliefs/p ercep tio n s and b eh avioral in ten tion s related to role
and decision m aking status. Pearson correlations w ere calculated betw een altitudes,
beliefs/perceptions and behavioral intentions and the status variables. Correlations were
then corrected for attenuation and statistical significance was determined based on 95%
probability confidence interval estimates. As shown in Table 28, several of the scales were
significantly correlated (i.e., confidence intervals did not contain 0) with decision making
status and role status. Individuals who indicated greater involvement in decision making
were more likely to report positive attitudes toward family support services and parent
involvem ent in services planning, perceptions of greater agency commitment, greater
capacity, greater administrative and staff support for services, and greater family advocacy
and com m unity support for services. Decision making was negatively correlated with
intentions toward expansion and improvement in family support services, suggesting that
people who were more involved in agency decision making were less likely to affirm
desires to expand or improve services. This was corroborated by a negative correlation
between role status and expansion intentions, suggesting that people who were in higher
levels in the agency expressed less positive intentions toward expansion and improvement
o f services.
Role status was also positively correlated with negative attitudes, suggesting that higher
levels o f administration expressed more negative attitudes toward family support.
However, keep in mind that attitude scales were som ew hat skewed in the positive direction
and therefore relatively negative attitudes may still in fact be positive in an absolute sense.
Role status was also positively correlated with perceptions of greater agency comm itm ent
and adm inistrative support. Thus, higher level administrators, although less favorable
toward family support philosophy and expansion, expressed perceptions of greater agency
com m itm ent to implementing family support services and greater administrative support for
services. On the other hand, the sam e correlations indicate that staff were less likely to

75
Table 28: Corrected zero-order correlations between decision making status and role status,
and attitudes, beliefs/perceptions and behavioral intentions.

Corrected Zero-order Correlations
(± 95% confidence interval)
Decision Making

Role

1. Negative attitudes to FSS

-.06
(.12)

.17
(.10)

2. Parent involvem ent in planning

.13
(.12)

-.11
(.12)

3. Positive attitudes to FSS

.21
(.12)

-.03
(.12)

4. Expansion/im provem ent of FSS

-.24
(-12)

-.42
(.12)

5. Agency comm itm ent

.56
(.12)

.27
(.12)

6. Adm inistrative support

.53
(.08)

.34
(-10)

7. S taff support

.22
(.12)

.01
(-10)

8. Agency capacity

.19
(-14)

.01
(.12)

9. Fam ily and com m unity support

.25
(.14)

.11
(-12)

Attitude scales

Behavioral intention scales

Belief/perception scales

believe that their agency had hired and trained staff for family support, and were less likely
to indicate that administrators were supportive, possibly recognizing administrative
reluctance to expand services.
H yp oth esis 5: D ecision M aking and Role Statu s as M oderator
The fifth hypothesis stated that decision-making status and role status of stakeholders
would m oderate the prediction of family support configurations from stakeholder attitudes,

76
beliefs/perceptions and behavioral intentions. A series of multiple regression analyses were
conducted in order to test this hypothesis. All analyses for this hypothesis were conducted
at the individual level o f analysis (as compared with previous analyses which were at the
agency level). Pearson correlations, corrected for attenuation, were used to determine the
relationship between decision-m aking status and role status. As would be expected, a
higher level of decision making status was highly correlated with a higher administrative
role in the agency (corrected r = .70). Decision making status and role status were, thus,
standardized and the mean of the two scores was used as a single index of status. The
status scores were then standardized for the regression analyses. Further, all scales and
criteria were standardized.
A series of hierarchical multiple regression analyses were used to lest the hypothesis
that status moderated the effect of attitudes, beliefs/perceptions and behavioral intentions on
the criteria. At the first step in each regression, a variable which identified the person's
agency was entered in order to control for any potential effects of sim ilar agency
membership. Then the main effects of the attitude, belief/perception or behavioral intention
score and the status score were assessed by entering these variables into the equation.
Next, interaction terms were calculated by computing the cross-products of the status score
with each of the attitude, belief/perception, and behavioral intention scores. These crossproduct terms were entered into the regression equations at the third step. Cross-products
with significant partial correlations with the criteria, al ter controlling for all main effects,
would be evidence of significant interactions and indicative of moderator effects. As
shown in Table 29, only main effects were found to be significantly correlated with the
criteria. Main effects o f attitudes, beliefs/perception, and intentions were previously
discussed. Main effects of status are not interprelable. No significant interactions were
found. Thus, the fifth hypothesis, which predicted a m oderator effect of status with
attitudes, beliefs/perceptions and intentions, was not upheld.

77
Table 29: Status as moderator: results of regression analyses for hypothesis 5.

Main Effects
P red ictors o f N u m ber o f
F am ily S u p p o rt S ervices

Scale

Interaction

Status

____ ______H

Effects
- U _______ R

Attitude scales
Negative attitudes to FSS

-.22**

-.13*

-.02

.30

Parent involvement in planning

.13*

-. 15**

.04

.24

Positive altitudes to FSS

.26**

-.17

.05

.32

17**

-.09

.04

.26

.28**

-.24**

.01

.32

Administrative support

.20**

_

2 ^**

-.02

.26

Staff support

. 14**

-.15**

-.11

.26

-.10
^ 1**

-.13*
_ 19**

.02

.22

.07

.37

-.03

-.06

.20

Behavioral intention scales
Expansion/im provem ent o f FSS
Agency commitment
Belief/perception scales

Agency capacity
Family and com m unity support

P red ictors o f A d m in istra tiv e F o r m a liz a tio n
Attitude scales
Negative attitudes to FSS

-.18**

Parent involvement in planning

.10

-.05

.04

.14

Positive attitudes to FSS

.20**

-.06

.02

.22

-.05

-.02

.10

Behavioral intention scales
Expansion/im provem ent o f FSS
Agency commitment

-.00
.34**

-.16**

.10

.32

Administrative support

.21**

_ 14 **

.00

.22

Staff support

.19**

-.06

.00

.20

Agency capacity

.03

-.03

.09

.14

Family and comm unity support

.30**

-.09

-.02

.32

Belief/perception scales

* p < .05, * p < .01.

78
O verall P red ictors o f F am ily S u p p ort C on figu ration s
H ypothesis 6: The sixth hypothesis stated that beliefs/perceptions and attitudes
would be more predictive o f family support configurations than agency and context
characteristics. A series of simultaneous entry multiple regression analyses, using a
correlation matrix which had been corrected for attenuation, were used to test the
hypothesis. First, in order to reduce the num ber of variables with high inter-correlations in
the regression analyses, the demand variable was chosen to reflect the overall population
size, density, and demand for services. As was shown in Table 26, dem and had the
strongest correlations with the criteria and was equally highly correlated with population
size (corrected r = .98) and density (corrected r = .96). Dem and was used along with
agency budget and service area wealth indicators as agency and context characteristic
predictors of the criteria.
Second, separate multiple regression analyses (at the agency level of analysis) were
used to determine the most significant predictors from the sets of attitude and
belief/perception variables and from the set of agency and context predictors. In
conducting the analyses, it became evident that one o f the attitude scales, positive attitudes
toward family support, although the most highly correlated of the attitude scales with the
criteria, was too highly intercorrelated with the other attitude and belief/perception
variables. This multi-collinearity resulted in unacceptably small (less than .01) tolerance
estimates for this variable. Thus, the item was deleted from the m ultiple regression
analyses.
Regarding the number of different services, Table 30 indicates that beliefs/perceptions
regarding family advocacy and com m unity support for services (/i = .78, p < .01) and
attitudes toward parent involvement in services planning (/j = .53, p < .01) were the only
two attitude and belief/perception variables with significant Beta coefficients. In regard to
agency and context predictors of num ber of services, demand for services (/i = .75, p <
.05) was the only significant agency/context characteristic predictor.

Table 30: Predictors o f num ber of family support services. Results o f multiple regression
analyses.
Separate Analyses
P

R

Com bined Analysis
P

R2 Change

Attitude scales:
Negative attitudes to FSS

.44

Parent involvem ent in planning

.53**

.23*

Positive attitudes to FSS (tolerance
too small for procedure)
Belief/perception scalcs:
Administrative support

.27**
.03

Staff support

-.06

Agency capacity

-.04

Fam ily/com munity support

.78**

.69

.46**

.48

.30**

Agencv/context characteristics:
Agency budget

-.28

Dem and (subsidy enrollment)

.75*

Avg. personal income

.06

Governm ent spending per capita

.08**

-.06

* p < .05, * * p < .01

In order to determine the relative predictive value between attitude/belief and
agency/context variables to the number of services criterion, significant predictor variables
were submitted to a second multiple regression analysis. The results o f this analysis are
shown in the right hand colum ns of Table 30. This second analysis revealed that both of
the attitude and belief/perception variables as well as demand remained as a significant
predictors of the num ber o f family support services. The extent to which each construct
(i.e., individual versus agency/context variables) uniquely contributed to the prediction was

80
calculated by determining the increase in R2 as each set of variables (i.e., the attitude and
belief/perception variables combined compared with the demand variable) entered last into
the equation. A significant change in R2 indicates that a variable provides unique
information about the criterion that is not available from other independent variables in the
equation (Norusis, 1988). Results, as shown in Table 30, indicated that both variables
provided significant unique contributions to the regression model. The sixth hypothesis
predicted that attitudes/beliefs would be more predictive of the number of different services
provided. Because the change in 7?2 indicated that attitudes/beliefs independently
contributed more to the prediction of the num ber of services, the sixth hypothesis was
supported.
Sim ilar analyses were used to determine predictors of the degree of formalization. As
shown in Table 31, perceptions of family and com m unity support for services {(3 = .75, p
< .01) and beliefs regarding agency capacity (/3 = .33, p < .01) were significant predictors
of adm inistrative formalization. Demand was also a significant predictor (/j = .85, p <
.05). No other agency or service area characteristics were significant predictors. Once
again, in order to determine the relative predictive value o f attitude/belief and
agency/context variables to administrative formalization, significant predictor variables
were subm itted to a second multiple regression analysis. The results of this analysis are
shown in the right hand colum ns of Table 31. This second analysis revealed that only the
attitude and belief/perception variables remained as significant predictors of formalization.
Thus, the sixth hypothesis, which predicted that altitudes/beliefs would be more predictive
o f form alization, was supported.
S u m m a ry o f h y p o th ese s te s ts . Table 32 presents a sum m ary o f the results of the
hypotheses tests. In general, results indicated that variability in the two criteria was
explained by different models of prediction. Utilizing first-order correlations, organization

81
Table 31: Predictors of administrative formalization. Results of stepwise multiple
regression analyses.
First Analyses
P

R

Second Analysis
p

R2 Change

Attitude scales:
Negative attitudes to FSS

.46

Parent involvement in planning

.30

Positive attitudes to FSS (tolerance
too small for procedure)
Belief/perception seales:
Administrative support
Staff support

. 11
-.01

Agency capacity

.33**

Family/community support

.75**

.30*
.62

.52**

.29**

.30

.09

.01

Agencv/context characteristics:
Agency budget
Dem and (subsidy enrollment)
Avg. personal income
Governm ent spending per capi ta

* p < .05, ** p < .01

-.62
.85*
-.23
.04

82
Table 32: Review of hypotheses results.
Configuration Criteria
Hypothesis:

Services

Formalization

1: Organization size will be positively correlated with

Supported

Not supported

family support services (FSS) configurations.

2: Service area population size, density, wealth and
demand will be positively correlated with

Table 26

Supported

Not supported

Table 26

FSS configurations.

3: Attitudes o f stakeholders regarding FSS will be
positively correlated with FSS configurations.

4: Beliefs of stakeholders regarding factors which
encourage or discourage the implementation of FSS

Supported

Not supported

Table 27

Supported

Supported

Table 27

Table 27

will be positively correlated with family support
configurations.
5: Organization status and level of decision-making
participation of stakeholders will moderate the

Not supported Not supported
Table 29

Table 29

Supported

Supported

Table 30

Table 31

predictive value o f stakeholder altitudes and beliefs
to FSS configurations.

6: Beliefs and attitudes o f stakeholders will be more
predictive o f FSS configurations than agency and
service area characteristics.

83
and service area characteristics, and attitudes o f stakeholders toward family support were
predictive of variability in the number o f services but not the degree o f formalization. Only
beliefs/perceptions regarding factors which inhibit or encourage family support, and
behavioral intentions were predictive of both services and formalization criteria.
Although organizational and decision-making status o f the stakeholders correlated with
their attitudes, beliefs/perceptions, and behavioral intentions, the interaction between status
and altitudes/beliefs/intentions did not moderate the predictive value of these variables to the
criterion. A series of simultaneous entry regression analyses revealed that attitudes/beliefs
of stakeholders were more predictive of both the services criterion and the formalization
criterion.
B eyond the H ypotheses: E xp loratory Path M odel o f P red ictors
Research and theory regarding the adoption and implementation o f innovative services
have suggested that organization and context characteristics, as well as individual
characteristics, are predictive of implementation. The hypotheses tested in this study so far
support this general theory. As a further exploration of this theory, the path model which
was described in chapter one was examined for each criteria using ordinary least squares
path analysis (Hunter & Ham ilton, 1992). This analysis was conducted as an exploration
of the relevance of Lewin's (1951) field theory and Ajzen and Fishbein's (1980) theory of
reasoned action to the implementation of family support services, thereby testing the
mediating effect of behavioral intentions to the relationship between attitude/belief
predictors and the criteria. In path analysis, when a mediating variable is entered as an
explanatory variable, a non-significant path coefficient for the direct effect indicates that the
variance is explained as hypothesized. However, a significant path coefficient after
accounting for effects of mediating variables would indicate a direct effect that was not
accounted for by the mediating variables.
In testing the path model, only those variables with significant correlations with the
respective criterion variables (see Tables 26 and 27) were entered. Variables used in the

84

Fam ily/
Com m unity
y S u p p o rt J

FSS
Attitudes

Agency
C om m itm ent

Expansion
Preferences

N um ber of
Services

Dem and/
Resources

Figure 2
Hypothesized path model of predictors of num ber of different services

evaluation of the model also needed to be significantly correlated with antecedent and
consequent path variables. Further, item and scale measures which were highly inlercorrelaled and similarly correlated with the criterion variable were combined into more
general constructs, the variable with the strongest correlation with the criterion was used in
the path analyses.
P a th m odel o f p r e d ic to r s o f n u m b e r o f se rv ic e s. As show n in Figure 2, the
num ber of services implem ented was hypothesized to be a function of demand/resources,
perceptions o f family and com m unity support for services, and attitudes toward family
support. Previous correlational analyses indicated that both positive altitudes and negative
attitudes toward family support (FSS) were significantly correlated with the criterion and
were highly intercorrelated. In order to eliminate problems o f multi-collinearity, positive
attitudes toward family support, which had the strongest correlation with the criterion, was

85
entered in the path analysis. The relation of the criteria to family and comm unity support
was hypothesized to be mediated by the behavioral intention, agency com m itm ent to hiring
and training staff for family support. The relation of the criteria to family support attitudes
was hypothesized to be mediated by preference for expansion and improvement of
services. Further, as hypothesized/modeled, beliefs and altitudes were correlated.
In regard to demand for services and agency resources, as was shown in Table 26,
demand, population, population density, and agency budget were highly inter-correlated.
Thus, the four variables were m odeled as a four part demand/resources construct. The
assum ption was that areas which have a larger population base create a higher demand for
services which agencies then use to justify and argue for larger budgets. As demand for
services was the strongest predictor o f the number of services agencies implemented, it was
used in the path analysis. The model also depicts the hypothesized correlation between
dem and and stakeholders' perceptions of family and com m unity support for services.
Path coefficients are presented in Figure 3, along with respective 95% confidence
interval estimates. As hypothesized, attitude toward family support was significantly
correlated with perception of family and com m unity support (corrected r = .71, +.30).
Demand was also significantly correlated with perception of family and comm unity support
for services (corrected r = .35, ±.32).
As hypothesized, the num ber of different services provided by agencies was explained
by a pattern of relationships which included demand/resource characteristics, and attitudes
and beliefs/perceptions of stakeholders mediated by behavioral intentions. W hen number
o f services was regressed onto agency com m itm ent, expansion preferences, and
dem and/resources (R = .67; ± .20), demand (/) = .42; ± .24) and agency com m itm ent (/? =
.43; ± .28) remained as significant predictors (i.e., the confidence interval did not include
0) of the num ber o f services implemented, while expansion preferences approached
significance (/? = .21; ± .30). Analysis of the model as a whole (%2 (df = 7) = 4.34; n.s.)

86

Fam ily/
Com m unity
Support
.71 (+ 3 0 )
FSS
Attitudes

R= 7 3 (+201
Agency
Com m itm ent
.76 (±.18)
Expansion
Preferences

.43
(+.28)
.21 (+.30)

N um ber of
Services

.35 (+.32)
Dem and/
Resources

Figure 3
Examined path model of predictors of family support services

indicated that the hypothesized path adequately explained all of the direct effects of the
exogenous variables (i.e., all departures could be explained by sampling error). Therefore,
the hypothesis that behavioral intentions would mediate the effect of attitudes and
beliefs/perceptions on the criteria was supported. The paths as modeled accounted for 42%
o f the variance (R = .65; + .20).
P a th m odel o f p re d ic to rs o f fo rm a liz a tio n . Research regarding innovation
implem entation suggests that different models of prediction will be found for different
types of configurations, such as an innovation in practice/technology or administration.
For exam ple, research by Kimberly and Evanisko (1981) found predictors of technological
innovations to be different from predictors of administrative innovations. Also, hypotheses
tested for this research suggest differing models of prediction for services implementation

87

Fam ily/
Com m unity
V Support

J

A dm inistrative
Support

Agency
Com m itm ent

Adm inistrative
Formalization

D em and/
Resources

Figure 4
Hypothesized path model of predictors of administrative formalization

and administrative formalization. W hile variation in services was a function of both
attitudes and beliefs/perceptions, variation in administrative formalization was primarily a
function of beliefs/perceptions, since no attitude variables were significantly correlated with
this criterion (see Table 27). Thus, the hypothesized model which was examined included
only two primary paths: one associated with belief/perception variables, the other
associated with agency/context characteristics.
As shown in the hypothesized path model in Figure 4, the predictive value of
perceptions of family and com m unity support and administrative support to the criterion
was mediated by agency com m itm ent (behavioral intention). Further, demand was
modeled as a direct predictor of the criterion, and also as a correlate of perceptions of
family and com m unity support. However, given the results of zero-order correlational

88

Fam ily/
Com m unity
V uppon

.62(±.34)

f

R =.8lf+ .18)

*
~
\
Agency
A dm inistrative
istrative J------► ^C o m m itm en t
35^ S u p p o r

1_____ Jl6(±3A)

.50

■■K=.53.(±22.)__
A dm inistrative
Form alization

J
^

(+ 3 2 )

s Dem and/
Resources
V

\

l9(+.28)

)
Figure 5

Examined path model of predictors of administrative formalization

analyses and the multiple regression tests of hypothesis 6 for this criterion (see Table 31),
demand will most likely not remain as a significant predictor of administrative
formalization.
As shown in Figure 5, the direct link of formalization on agency com m itm ent (i.e., the
behavioral intention) was significant (/3 = .50; ± .26). The links o f agency com m itm ent on
perceptions o f family and community support (/? = .62; ± .34) and adm inistrative support
(/3 = .36; ± .34) were also significant and accounted for 66% o f the variance (R = .8 1; ±
.18). Demand/resources, as predicted, was significantly correlated with family and
com m unity support (corrected r = .35; ± .32), however, dem and/resources was not
significantly associated with formalization (/J = . 19, ± .28). Analysis of the model as a
whole (%2 (df = 3) = 1.28; n.s.) indicated that the hypothesized path adequately explained
all of the direct effects of the exogenous variables (i.e., all departures could be explained by

89
sampling error). Therefore, the hypothesis that behavioral intentions would mediate the
effect of beliefs on the criteria was supported. The paths as m odeled accounted for 28% of
the variance (R = .53; ±.22).

CHAPTER 4
D is c u s s io n

The purpose of this research was to improve upon three limitations in previous work on
innovation implementation: the scarcity of studies examining the ecology o f implementing
innovative services, including the combined effects of individual, organizational, and
contextual factors; the lack of research focusing on altitudes and beliefs/perceptions as
individual predictors of implementation; and the limited statistical sophistication. This
research was focused on family support services in order to better understand the factors
which account for the wide diversity in implementation of this innovative practice initiative
so that potential avenues for enabling further implementation may be determined. These
issues were addressed in the context of testing the general hypothesis that implementation is
determined by characteristics of the agency and its service area together with characteristics
of individuals in the organization.
In general, the findings support previous research (Baldridge & Burnham, 1975;
Castellani et al., 1986; Ireys et ah, 1985; Kimberly & Evanisko, 1981; Rogers, 1983) and
theory (Havelock, 1971; Rogers, 1983) which suggested that characteristics of the
organization and its context are predictive of innovation implementation. For the num ber of
different services implemented, relatively moderate correlations were found with agency
and service area characteristics, specifically budget, demand, service population and
density. These findings are sim ilar to previous findings (Castellani et ah, 1986; Ireys el
ah, 1985) which suggested that agencies with larger budgets and higher service area

90

91
population densities provided m ore services. Apparently, areas which have a larger
population base create a higher demand for services. Agencies most likely utilize this
perceived demand to justify increased budget requests, and then use the expanded
resources to provide more diverse services. It is also possible, as discussed in the literature
review, that larger agencies potentially have more capacity to shift resources to support the
im plem entation of new services. Therefore, agencies with larger population bases, and
larger budgets may be able to provide a greater diversity in services from existing resources
rather than merely obtaining new resources to fund the services. Both of these
explanations are possible within any given system. Yet, in recognition that the relationship
between demand/resources and num ber of different services implemented was only
m oderate, in some cases perceived demand and funding may be used to expand a particular
service instead of expanding the diversity of services offered. This explanation may help to
understand why longitudinal research of Herman and Hazel (1991) failed to find a
statistically significant relationship between increases in funding over time and the resultant
diversity in services implem entation. However, other explanations are possible.
As has been suggested by advocates of family support (Agosta, Bradley, Rugg et al.,
1985; Knoll et ah, 1990; Taylor, 1989), attitudes which are in harm ony with the innovative
practice, such as those favoring the innovative philosophy and encouraging of expansion,
are im portant to the adoption and implementation of more diverse aspects of a 'loosely
bundled' innovation such as fam ily support. This research supports previous research
(B acker et ah, 1986; Fairw ealher et ah, 1974; Hage & Dewar, 1973; Rogers, 1983) and
theory (Agosta, Bradley, Rugg et ah, 1985; Rogers, 1983; Sabatier & M azmanian, 1981;
Scheirer, 1981; Taylor, et ah, 1989) which suggested that characteristics of individuals,
their altitudes and beliefs/perceptions, would be predictive of innovation implementation.
Relationships between attitude and belief/perception scales and the num ber of services
implem ented were moderate and sometimes stronger in comparison with relationships with
dem and/resources. In regard to attitudes, reasoned action theory (Ajzen & Fishbein, 1980)

92
suggested that attitudes toward family support philosophy and implementation would be
predictive o f behavior, i.e., implem entation. Results supported this theory.
Reasoned action theory also suggested that perceptions o f support from peers,
administration, and the wider community, i.e., the subjective norm, would be predictive of
behavior. This research also supported this aspect o f the theory. Perceptions of greater
family advocacy and com m unity support for services were significantly related to more
services being implemented. Further, beliefs regarding the subjective norm, such as
perceptions of family, community, and administrative support were related to the
implementation o f more formal administrative structures.
In regard to the relative importance of agency, context and individual predictors,
individual characteristics were more important than agency and service area characteristics
for the prediction o f services implementation. This finding is in contrast to previous
research (Baldridge & Burnham, 1975; Kimberly & Evanisko, 1981) which found agency
and context characteristics to be more important. The discrepancy can he accounted for by
the fact that previous research focused on individual characteristics such as age, education,
gender, etc., and failed to measure attitudes and beliefs/perceptions of organization
personnel. Thus, individual characteristics were not found to be relevant. In general,
results pertaining to both criteria support theoretical argum ents (Agosta, Bradley, Rugg et
al., 1985; Berman & M cLaughlin, 1978; Browning et ah, 1981; G laser et ah, 1983; Knoll
et ah, 1990; M cLaughlin, 1987; Zaltman et ah, 1973) and Hage & Dewar's (1973)
findings that individual values/beliefs are more predictive of diversity in implementation
than agency characteristics. Also contrary to previous research (Kimberly & Evanisko,
1981), factors related to organizational structure and context were not significant predictors
of administrative changes. Instead, what seemed to be important in understanding
implementation of administrative practices for family support services were factors
associated with agency personnel, their perceptions of the need and com m unity support for
services, and perceptions of agency capacity for providing services.

93
As was indicated in the background literature review, previous research focused
primarily on univariate analyses. Multivariate analyses, when utilized, were geared at
defining the relative importance of agency and context characteristics. This approach, as
this research has shown, limits our understanding by neglecting to account for the various
multiple correlates of the criterion. In order to extend our understanding of innovation
implementation beyond a discussion of relative importance of agency/context versus
individual characteristics, path models were hypothesized and tested based on theoretical
considerations. Theories which emphasize the role of people's perceptions of their
environm ent and their attitudes toward the innovation (i.e., field theory and reasoned action
theory) were used to formulate causal models for multiple predictors of the criteria.
Implementation was modeled as a function of agency/context forces, and individual
altitudes and perceptions of the social norms for implementation which were mediated by
their behavioral intentions to implement the innovation. As proposed, the number of
different services implemented was predicted by attitudes toward family support, mediated
by intentions to expand services. Diversity in services implementation was also predicted
by perceptions of family and community support for services mediated by the behavioral
intention to hire and train staff for family support. Further, structural characteristics of the
agency and service area were also predictive of the number o f services. Agencies with
larger budgets, more populated service areas, and higher demand for services implemented
more services. Since demand for services was correlated with perceptions of support from
families and the comm unity for services, and these perceptions were highly correlated with
attitudes, it is difficult to separate out the relative importance of attitudes/beliefs and
structural characteristics.
Consistent with field theory and reasoned action theory, individuals evaluate the
pressures to implement or not implement family support services in forming their attitudes
toward the innovation and behavioral intentions. These pressures include the actual
demand for services and the agency's capacity to implement services, along with perceived

94
support from others lor implementation. The important point is that beliefs/perceptions and
attitudes are just as important as fiscal resources and demand to the prediction of services
implementation. Demand and resources may be necessary, but are not sufficient to
encourage diversity in family support services implementation.
The predictive value of individual characteristics, i.e., beliefs/perceptions, was even
stronger in regard to administrative formalization of services. Formalization was best
understood as primarily a function of beliefs/perceptions. In the path model, formalization
was a function of stakeholders' perceptions of family and com m unity support along with
perceptions of administrative support mediated by agency comm itm ent to hire and train
staff for family support (i.e., the behavioral intention). No attitude or agency/context
characteristics were found to be important to the prediction of formalization.
Thus, the findings support previous research (Kimberly & Evanisko, 1981) which
suggested that the variability in the two criteria, number of services and administrative
formalization, would be explained by different models of prediction. Results suggest that
antecedent conditions for the implementation of formal administrative structures, and thus
the routinization o flh e program, differ from conditions to implement more diverse
services, in that formalization was not predicted by demand/resource explanations or
attitudes toward family support philosophy and implementation. Perhaps,
formalization/routinization is based on altiludinal resistance theories which suggest that
people resist innovations that appear to require too much change or are perceived to devalue
their know ledge and skills (Glaser et ah, 1983; Legge, 1984). The formalization measure
included practices that require new views of the role of the professional-parent relationship
(parent advisory committees) and a restructuring of standard operating procedures
(program structure for family support). Those whose support is required for such changes
in adm inistrative structures, i.e., the administration, expressed less favorable altitudes
toward parent involvement in services planning. Stakeholders' perceptions of
administrative support was an important predictor of the formalization criterion.

95
Yet, correlations between services and formalization suggest that as more family
support services are implemented, more formal administrative procedures are also
implemented. Perhaps, as more services are implemented, people in the organization may
become familiar with family support and thereby more supportive o f the innovation. As the
subjective norm for the behavior becomes supportive, people may then be more inclined to
explore possibilities of routinizing their efforts. Conversely, routinizalion m ay also
facilitate implementation of more diverse services.
These considerations suggest that a better conceptualization of the role of attitudes,
beliefs/perceptions and structural characteristics may be found in system dynamic models
and the concept o f feedback loops (cf. Levine, Van Sell & Rubin, 1992). In this case,
attitudes and beliefs/perceptions not only affect services implementation, but services
implementation may also change attitudes and beliefs/perceptions which then further impact
services implementation and routinizalion. Also, as demand increases, perceptions of
demand may increase, encouraging the implementation of services, which then fosters
recruitment of clients, which then can increase demand for services as more families
become aware of the services that are being provided. Longitudinal research which focuses
on change over time may help in the determination of whether a system dynamic model, as
com pared to linear regression models, presents a better understanding of the process of
innovation implementation and routinizalion.
Despite predictions suggested by conflict of interest theories that organizational status
will moderate the predictive value of attitudes to implementation (Hasenfeld & Brock,
1991; Sabatier, 1987b; Scheirer, 1981), no significant attitude/belief by status interaction
effects were found for either criteria. This finding suggests that for family support
services, contrary to the 'top down' approach to implementation, altitudes and
beliefs/perceptions of all those in decision making roles are important in determining
performance. However, attitudes and beliels/perceptions were correlated with role and
decision making status. Staff expressed more positive attitudes toward family support

96
philosophy and implementation, and perceptions of lesser agency comm itm ent, lesser
family and comm unity support for services, and lesser adm inistrative support for services.
Those more involved in decision making and of higher organizational status expressed less
favorable attitudes toward expansion and im provem ent of services. This, in com bination
with staffs more positive attitudes toward family support philosophy, may help to explain
why staff saw administration as less supportive of family support and their agency as less
committed to implementing support services than administrators did. Staff may perceive a
conflict between what adm inistrators say they support, and administrators' behavioral
intentions as identified by their greater hesitancy to expand and improve services.
Certain constraints on the generalizabilily of the findings should be noted. First,
because family support services are widely accepted and implem ented across M ichigan's
mental health services system, the findings may only be generali/.able to other similar
service systems and to points in time beyond the initial stages of policy and services
implementation. Initial implementation may be more or less dependent on individual
characteristics. For example, the conflict of interests model, which suggests that
organizational status will moderate the predictive value of attitudes to implementation, may
be more important to understanding initial stages of implementation. Once an innovative
service or policy has been widely accepted and routinized, there may be more sim ilarity in
attitudes of staff and administration, thus the conflicts of interests model may no longer be
operative. Once again, longitudinal assessments may be necessary to address this issue.
Specifically, do attitudes and beliefs change over time, and if so, are these changes
reflected in changes in implementation behavior? Also, does the relative importance of
individual, organizational and context characteristics change over time, and if so, how? At
what point in time, if any, is the conflict of interests model im portant to understanding
implementation behavior?
The widespread implementation of family support services in M ichigan may also have
produced a situation where agency personnel attitudes and beliefs/perceptions, in general,

97
were positive towards the implementation of family support services. Results of the scale
developm ent analyses suggested that altitudes were skewed toward the supportive end of
the scale. M any recognized the need for family support and would like to see services
expanded or improved. Participants in this study also perceived a general sense o f support
from administration, sta ll and the community for the implementation of family support
services. Although they identified several harriers to implementing services, including lack
o f resources within the agency and the surrounding community, and capacity to provide
services, they also identified several factors which have enhanced their agency's ability to
provide services. These factors included support from agency personnel, state level
funding allocations, training opportunities and hiring of staff with relevant expertise.
The skewed results o f the attitude measures suggested people may have responded to
item wording and content in a 'politically correct' manner. In the m onths ju st prior to and
during the time this research was being conducted, advocacy efforts both by parent groups
and state level administration to improve the availability and accessibility of family support
services may have generated a sense that attitudes disfavoring family support were not
acceptable. The skewed results of the altitude measures may have limited this research's
ability to adequately determine the relative predictive value of altitudes on the criteria.
A further difficulty related to the scale development analyses was the common content
o f the items which were deleted. First, an often discussed theme in the family support
literature is the importance of recognizing the value of family oriented services over
individual oriented services. Family support advocates argue that deinstilulionalizalion and
the resultant focus on providing individuals with support services has created a barrier to
the im plem entation o f family support services (Agosta, Bradley, Rugg et al., 1985;
Herm an et ah, 1992; M oroney, 1979). They argue that although a focus on supporting
natural networks, such as the family, will ultimately better serve the individual, many
service agencies continue to focus on the individual as the service recipient rather than the
family. Several attitude and belief items were written to tap into this philosophy. M ost of

98
these items were deleted during the analyses. Com m ents from participants suggest that the
argum ent o f individual versus family based services may be loo simplistic, that both are
necessaiy and, therefore, important. For example, individualized services may be best for
adults with developmental disabilities, while family oriented services may be best for
children. Participants who were drawn from either child or adult oriented services may
hold differing viewpoints. Since age based distinctions were not made in the items
pertaining to this debate, answers may have been unstable, resulting in low internal
reliability and inter-item commonality.
A second theme among the items deleted from the scales was related to change.
Literature on innovation implementation suggested that attitudes toward change would be
related to behavior (Legge, 1984). Items which attempted to measure this construct were
also deleted during the scale developm ent analyses. Com m ents from participants suggest
that the items may have been confusing or unclear, resulting in unreliable responses.
Another explanation may be that, once again, given the widespread implementation of
family support services, attitudes reflecting a resistance to change may no longer be
consistently evident, thereby resulting in low inter-item commonality. Other than these two
exceptions, results of the scale development analyses were consistent with expected
dom ains of measurement. Resultant scales measured attitudes toward family support
philosophy and implementation, perceptions of the external and internal subjective norm for
implem entation, and behavioral intentions to expand and improve services.
In regard to the criteria, a wide diversity of implementation was found within the
configurations studied. Apparently the "loose bundling" of family support has allowed for
this diversity in implementation, with some agencies providing several different services
and roulinizing their efforts with the adoption of several administrative components, while
others have implemented very few services and no administrative components. This
observed variety in implementation and factors which potentially explain the variations
were the focus o f this research. However, a limitation on the generali/ability of the

99
findings related to the services criterion measure should be noted. Respondents'
definitions of what constitutes family support varied, with several services identified as
family support that had not been formally recognized by state policy or in the literature. It
appears that although family support is gaining acceptance, people may still be unclear as to
w hat "family support" means. Perceptions of respondents support this finding; 41%
indicated that people at their agency lacked a clear understanding of what family support
services are. Clearly, advocacy and policy efforts to clarify what is m eant by family
support services are still needed and should be encouraged.
Overall, the results of this research suggest that Lewin's field theory and Ajzen and
Fishbein's theory of reasoned action can guide our understanding of innovation
implementation in public mental health settings, as well as the implementation of family
support services in other states. In regard to 'loosely bundled' mental health services
innovations, such as family support, context demand forces and availability of sufficient
resources may be necessary for implementation, but not sufficient. Organizations may
have the resources, the demand may be evident in the service area, but without the support
from agency personnel, as evident by their attitudes and beliefs/perceptions, few if any
components of the innovation will be implemented. Conversely, altitudes and
belief/perceptions may support services implementation, but if the resources are not
available, nor the demand evident to justify implementation, agencies are not likely to
im plem ent more diverse services.
Further, it is important to recognize that antecedent conditions for routinizalion (Yin,
1981) will differ from conditions for initial implementation. As suggested by this research,
formahzation/routinization, which appears to be correlated with an increase in the num ber
of innovation components implemented, is more a function of perceptions of agency
personnel of the need and comm unity support for the innovation along with administrative
comm itm ent, than of actual demand characteristics of the service area. Therefore, in public
mental health service areas, the stability and continued implementation of innovative

100
policies and procedures will be dependent upon support from agency personnel, especially
adm inistrative support, and their perceptions of comm unity support. In other words,
widespread community recognition and advocacy for the implementation of an innovative
menial health service philosophy, such as family support, will spur mental health agencies
to move towards full implementation and routinizalion of the innovative practice.
Both of the path models predicting number of services and formalization point to the
importance o f family advocacy and community support for services, mediated by agency
comm itm ent to implementing family support. Several policy and advocacy implications can
be suggested based on these findings. First, as more families inquire about and enroll in
the family support subsidy, perceptions regarding the demand for family support will
change, potentially leading to further implementation and routinization of family support
services to match the perceived demand. Also, efforts to directly affect these perceptions,
such as advocacy campaigns directed at agency personnel to heighten their awareness of the
need for services and families' desire to receive services, can also help to spur the
implem entation and routinization of family support services. Second, perceptions of
comm unity agency referrals for services and the availability of sufficient community
resources were included in the measurement of family and community support for services.
Therefore, policies and practices which encourage collaboration and coordination of
services among community agencies that serve people with developmental disabilities and
their families may help to spur implementation and routinization by impacting the agency
personnel's perception of comm unity support for services. Services coordination efforts
may also increase the visibility of services provided by mental health agencies which then
could increase referrals, thus potentially affecting perceptions of demand for services.
Finally, it is clear from the literature and this research that attitudes toward family
support and attitudes regarding parental involvement in services planning (which were
highly correlated, r = .93, with attitudes toward family support) are important to the
implem entation of family support services. Policy and advocacy efforts which are geared

101
toward encouraging parents to get involved, and also encouraging service providers to
work with parents as co-planners should be promoted. Program s should encourage
families' active participation in planning the service system , empowering them to design
and select the services they receive. In M ichigan, family support services policy requires
that agencies utilize parent advisory committees in services planning. However, this policy
has not been widely implem ented. Efforts by the state to enforce this policy should be
directed at administration, since their attitudes were found to he less favorable toward
parent involvement. These efforts would not only benefit parents by encouraging their
involvement, but may also help to increase the num ber of family support services
com ponents being im plem ented across the state. M ichigan's family support services'
evolution began with parent advocacy, the continued success can also rely on parental
advocacy and involvement. Efforts to encourage parent groups to advocate for their rights
and involvement in services planning locally, can have a ripple effect, increasing positive
attitudes toward family support and parental involvem ent and thereby, increasing efforts to
m eet perceived demands by further developing and enriching current efforts to implement
family support services.

A P P E N D IX A
Stakeholder Survey and Accompanying Letters

A P P E N D IX A
Stakeholder Survey and Accompanying Letters

Community Mental Health Professionals'
Opinion Survey
Regarding Family Support Services

Kelly L. Hazel, Principal Investigator
Department of Psychology
Psychology Research Building
Michigan State University
East Lansing, Ml 48824 -1 1 1 7

This research is supported in part by the Michigan Depanment of Mental Health and the
Michigan Developmental Disabilities Council

102

103

Instructions and Clarifications
*

T h is s u rv e y c o n c e rn s is s u e s s u rro u n d in g s u p p o rt s e rv ic e s p ro v id e d to fa m ilie s w h o a re c a rin g a t h o m e fo r a
m e m b e r w h o h a s a d e v e lo p m e n ta l d is a b ility . T h e s e s e r v ic e s a re o fte n r e f e r r e d to a s " f a m ily s u p p o r t
s e r v i c e s " , a n d m a y in c lu d e :

re s p ite c a r e (in- o r o u t-o f h o m e c a r e , a n d th e la m ily frie n d m o d e i)
fam ily s e r v ic e s c o o rd in a tio n (fam ily c e n t e r e d c a s e m a n a g e m e n t s e r v ic e s )
p a r e n t, fam ily, o r sib lin g s u p p o rt g r o u p s
fam ily c o u n s e lin g
b e h a v io ra l m a n a g e m e n t tra in in g
p a re n tin g skills tra in in g a n d o th e r p a r e n t e d u c a tio n a ctiv itie s
fam ily s u p p o rt o rie n te d n e w s le tte r s
p a r e n t- to - p a r e n t n e tw o rk in g
s p e c ia l in -h o m e s e r v ic e s (n u rsin g , s p e c ia l th e r a p ie s , b e h a v io ra l tra in e r, e tc .)
h o m e r e n o v a tio n s a n d a d a p t iv e e q u ip m e n t
s u m m e r o r a fte r s c h o o l re c re a tio n p ro g r a m s
fam ily s u p p o rt s u b s id y a n d o th e r fo rm s o t fin a n c ia l a s s is ta n c e
o th e r s e n /i c e s r r a c tiv itie s p ro v id e d to fa m ilie s u n iq u e to y o u r b o a r d /a g e n c y
Y o u r b o a rd /a g e n c y m ay p ro v id e a fu ll ra n g e o f " fa m ily s u p p o r t s e rv ic e s" , a few s e le c t s e rv ic e s (s u c h a s
re s p ite c a re a n d fam ily s e rv ic e s c o o rd in a tio n ), o r n o n e a t a ll. W h a te v e r th e c a s e m a y b e fo r y o u r
b o a rd /a g e n c y , w e arc in te re s te d in w lia t y o u h av e to say .
*

A ll q u e s tio n s s h o u ld b e a n s w e r e d b a s e d o n y o u r o w n o p in io n s a n d k n o w le d g e . P le a s e d o n o t g iv e th is
su rv e y to s o m e o n e e lse to a n s w e r . Y o u w e re s e le c te d to a n s w e r th is s u r v e y b a s e d o n y o u r p o s itio n w ith
y o u r lo c a l c o m m u n ity m e n ta l h e a lth c e n te r o r c o n tra c te d a g e n c y . B e c a u s e w e w a n t o p in io n s fro m a v a rie ty
o f c o m m u n ity m e n ta l h ea lth p ro fe s s io n a ls , in c lu d in g a d m in is tra to rs , b o a rd c o m m itte e m e m b e rs , a n d d ire c t
se rv ic e s sta ff, y o u r a n s w e rs a r c im p o rta n t. Y o u r p e rs o n a l r e s p o n s e w ill e n s u r e r e p re s e n ta tiv e n e s s .

*

W e u s e th e te rm b o a r d / a g e n c y to in d ic a te a c tiv itie s o f th e lo c a l c o m m u n ity m e n ta l h e a lth c e n te r a n d it’s
c o n tra c te d a g e n c ie s . Y ou m a y b e an e m p lo y e e o r v o lu n te e r o f th e c e n te r o r a c o n tra c te d a g e n c y . W h ic h e v e r
th e c a s e , re fe re n c e s to a c tiv itie s p ro v id e d b y y o u r b o a rd /a g e n c y in c lu d e b o th d ire c tly o p e ra te d p ro g ra m s o f
th e c e n te r a n d p ro g ra m s p ro v id e d th ro u g h c o n tr a c t a g e n c ie s . I f y o u a r e a n e m p lo y e e o f a c o n tra c te d ag e n c y
o f a c o m m u n ity m e n ta l h e a lth c e n te r, p le a s e g iv e p n g a n s w e r to e a c h q u e s tio n w h ic h b e s t r e f le c ts y o u r
o p in io n s r e g a rd in g b o th y o u r a g e n c y a n d y o u r c o m m u n ity m e n ta l h e a lth c e n te r .

*

P le a s e a n s w e r a il q u e s tio n s as; b e s t a s y o u c a n . D o n o t s k ip q u e s tio n s u n le s s d ir e c te d to d o so . P o r
q u e s tio n s in w h ic h a ra n g e o f a n s w e rs is p ro v id e d , u n le ss o th e rw is e in d ic a te d , p le a s e c irc le o n ly o n e a n s w e r
to e a c h q u e stio n .

*
*

D o n o t w rite y o u r n a m e o n d ie s u rv e y .
A fte r y o u h av e c o m p le te d th e su rv e y , p le a s e re tu rn it u s in g th e a d d re s s e d r e tu r n e n v e lo p e p ro v id e d to:
K elly L. H a z e l, M .A .
D e p a rtm e n t o f P sy c h o lo g y
M ic h ig a n S ta te U n iv e rs ity
E a st L a n sin g , M I 4 8 8 2 4 -1 1 1 7

If y o u h a v e a n y q u e s tio n s o r n e e d clarification, p le a s e c all Kelly L. H a z e l a t (5 1 7 )3 9 4 -6 4 2 8 , o r le a v e a
m e s s a g e a t th e P sy c h o lo g y D e p a r tm e n t a t M ich ig an S ta te U niv ersity , (5 1 7 ) 3 5 5 -9 5 6 1 , a n d s h e will
re tu rn y o u r call a s s o o n a s p o s s ib le .

‘Thank-you fo r your cooperation

10 4

3

1 . Before we get to your opinions, we would like you to think about the services your
board/agency (i.e., the community mental health center and its contracted agencies')
provides. In your view, what services and activities does your board/agency currently
provide to support families who are caring at home for their member(s) who have a
developmental disability?

2 . The following items are designed to determine what community mental health service
professionals, such as yourself, feel about support services for families caring at home for
their membeiis) with a developmental disability, commonly referred to as "family support
services". Please respond to the items in terms of your own perceptions, or how you feel
about your board/agency's implementation of support services for families. Circle the
response that most closely indicates how you feel about each statement.
Strongly
Strongly I
Disagree Disagree Neutral Agree Agree |

I
I feel that programs and services for people
with disabilities should take priority over
support services for families.

SD

D

N

/V

SA

I would like to see my board/agency expand
or improve its family support services.

SD

D

N

A

SA

Since most families with children with
SD
disabilities are supported by the school system,
it is not necessary for mental health to also
provide services for them.

D

N

A

SA

I am personally committed to the development
and/or expansion of family support services at
my board/agency.

SD

D

N

A

SA

Family support services require too much
change in the way my board/agency provides
services to our clients.

SD

D

N

A

SA

105

4

Strongly
Strongly I
Disagree Disagree Neutral Agree Agree J

What do you think?

I

Parents are very knowledgeable and competent S D
resources and should be involved in all services
planning for their family member who has a
developmental disability.

D

N

A

SA

There are needs in our community that are
much more pressing which should be
addressed first, before we develop family
support services any further.

SD

D

N

A

SA

Family support services take up too much
staff time that would be better devoted to
other, more needed services.

SD

D

N

A

SA

I think that family support should be a greater
percentage of my board's overall budget.

SD

D

N

A

SA

I feel that the focus of services should be on
the person with disabilities, not the family.

SD

D

N

A

SA

1 would like to encourage and help other service S D
providers implement or improve family support
services.

D

N

A

SA

I would like to see my board/agency
SD
discontinue providing family support services.

D

N

A

SA

I personally do not see a strong enough need in
our community to warrant any further
development of family support services at our
board/agency.

D

N

A

SA

I feel that all support services to people with a S D
developmental disability who are living at home
are more effective when they are family
centered and family driven.

D

N

A

SA

I think that my board/agency should redirect
current resources to provide more support
services for families.

SD

D

N

A

SA

Too many families are not capable of
functioning as care givers for their family
member who has a developmental disability.

SD

D

N

A

SA

D

N

A

SA

I feel that supporting families should be a top
priority at my board/agency.

SD

SD

106

5

Strongly
Strongly
Disagree Disagree Neutral Agree Agree

What do you think?

I

Many activities and services related to family
support are not within the domain of mental
health responsibility, other service providers
should be providing these supports.

SD

D

I feel that parents' views and opinions
regarding needed services and service delivery
practices are very important to overall services
planning.

SD

D

I am not at all convinced that family support
services are an effective or appropriate way of
helping people with disabilities.

SD

I personally do not think there is a need for
family support services.

A

SA

N

A

SA

D

N

A

SA

SD

D

N

A

SA

I think that my board/agency should find out
more about family support services before we
go any further.

SD

D

N

A

SA

Families have a lot of resources for help and
support from other service providers, mental
health should not be spending its scarce
resources too.

SD

D

N

A

SA

I think parents should be actively involved in
planning overall services for people with
developmental disabilities.

SD

D

N

A

SA

I feel that family support concepts and practices S D
should be incorporated into the overall way of
providing services to individuals with
disabilities and their families. Family support
should be my board/agency's way of doing
business.

D

N

A

SA

1 would like to know what other
boards/agencies are doing in regards to family
support services.

SD

D

N

A

SA

SD

D

N

A

SA

I think that staff resources should be redirected
to put more emphasis on supporting families

N

10 7

3 . We are interested in your opinion about how people at your board/agency feel about mental
health service agencies providing support services for families. What do you think people
at the following levels in your board/agency’s administrative structure feel? Based on your
experience, are they neutral or do they highly object to, somewhat object to, somewhat
approve of or highly approve of mental health agencies providing family support services?
Circle your response. If the description of an individual or group is not relevant to you,
please indicate so by putting a NA next to the description.

In regards to family
support services, how do the
following feel:

f

Highly Somewhat
Object
Object

The staff vou supervise?........................H O
Your professional peers?................ ...HO
Your immediate supervisor?.............. ...H O

Neutral

SO
SO

Somewhat Highl)
Approve Approv

N
N

SA

HA

SA

HA

N

HA

N

SA
SA

HA

N

SA

HA

N

SA

HA

The board's executive director?........ ...H O

so
so
so
so

Your board/agency's top
administration?................................ ...H O

so

N

SA

HA

Your board/agency’s developmental
disabilities services administration? ... HO

so

N

SA

IIA

so

N

SA

HA

so
so

N

SA

HA

N

SA

HA

Your board's chairperson?............. ...HO
Other board committee members?.... ...HO

Your board/agency's developmental
disabilities services staff?............... ...H O
Your board/agency's family
support services staff?.................... ...H O
Y ourself?.......................................... ...H O

4 . There is a wide variation across the state in the extent to which boards/agencies are
providing support services for families. We are interested in determining what community
mental health professionals are concerned about in regards to providing family support
services. Please respond to the items in terms of vour own perceptions, or how you feel
about the issues. Circle the response that most closely indicates your opinion to each
statement.
Very Somewhat
Somewhat Very
Untrue
Untrue Neither True
True

I

My board/agency has sufficient capacity to
VU
provide adequate support services to all families
in our community who need/want them.

SU

N

ST

VT

In general, people at my board/agency are not
interested in providing services for families.

SU

N

ST

VT

VU

108

7

Very Somewhat
Somewhat Very
Untrue
Untrue Neither True
True
What do you think?
My board/agency lacks sufficient staff
resources to provide adequate family support
services.

VU

SU

N

ST

VT

Providing support services for families is a top
budget priority at my board/agency.

VU

SU

N

ST

VT

State Department funding allocations have
helped my board/agency's ability to provide
family support services.

VU

SU

N

ST

VT

My board/agency has provided the necessary
opportunities for staff to improve their
knowledge and learn the skills required to
provide support services to families.

VU

SU

N

ST

VT

There is a sufficient number of qualified
VU
service providers (e.g. for respite care, in-home
nursing, in-home behavioral management
training, etc.) in our community to support my
board/agency's efforts to provide family
support services.

SU

N

ST

VT

Families in our community have been very
active in advocating for the development and/or
expansion of family support services at my
board/agency.

VU

SU

N

ST

VT

In general, people at my board/agency lack a
clear understanding of what family support
services are.

VU

SU

N

ST

VT

My board/agency has lacked the funding
capacity to develop family support services.

VU

SU

N

ST

VT

My board has hired people who have the
expertise needed to provide support services to
families.

VU

SU

N

ST

VT

The various agencies and professionals who
work with families in our community readily
refer families to my board/agency for the
services we provide.

VU

SU

N

ST

VT

Our mental health board committee members
are committed to the development and/or
expansion of family support services.

VU

SU

N

ST

VT

100

Very Somewhat
Somewhat Very I
Untrue Untrue Neither True
True I
What do you think?
My board/agency has not been able to keep u p V U
with the demand for family support services,
families have had to wait to get into the
program or to receive services.

SU

N

ST

VT

Families have been reluctant to take advantage
of the services available at my board/agency.

SU

N

ST

VT

VU

My board/agency's administration is committed VU
to the development and/or expansion of family
support services.

SU

N

ST

VT

There is a general lack of community interest
VU
and involvement in supporting individuals with
disabilities and their families.

SU

N

ST

VT

The system does not recognize the family a s t h e V U
client, only the individual with disabilities,
which hinders our efforts to implement support
services for families.

SU

N

ST

VT

VU

SU

N

ST

VT

The attitudes and practices of most service
VU
providers in our community are that support
services should be provided only to the
individual with disabilities, they do not involve
or support the family.

SU

N

ST

VT

Families in our community really don't want
family support servic s.

5 . What plans (if any) does your board/agency have in regards to family support services in
the near future?

110

9

So that we can compare and contrast your answers with other community mental health
professionals, we would like to know some information about yourself and your role in the
decision making processes at your board/agency.
6 . What impact do you, personally, have on decisions regarding the overall planning and
implementation of services at your board/agency? Circle your answer.
Very Little

L ittle

Som e

Much

Very Much

7 . How frequently do you participate in decisions regarding the overall planning and
implementation of services at your aboard/agency? Circle your answer.
N ever

S eld o m

S o m etim es

O ften

A lw a y s

8 . How frequently do you participate in decisions regarding the implementation and provision
of services for individuals with developmental disabilities and/or their families? Circle your
answer.
N ever

S eld o m

S o m etim es

Often

A lw a y s

9 . What impact do you, personally, have on decisions regarding services and supports to
individuals with disabilities and/or their families at your board/agency? Circle your
answer.
Very Little

L ittle

Som e

Much

Very Much

1 0 . How long have you worked or volunteered for this board/agency? Circle one answer.
A less than one year
B 1-2 years
C 3-5 years
D 6-10 years
E more than 10 years
1 1 . What is your current position at this board/agency? Circle the one answer which best
describes your current position.
A family support services staff
B developmental disabilities services staff (not family support)
C family support services/program administration
D developmental disabilities services/program administration
E service program staff for both MI and DD client services
F service program administration for both MI and DD client services
G contract service agency administration
H community mental health agency administration (e.g. director, asst, director)
I

board committee member

J

board committee chairperson

K other (please specify)_________________________________________

I ll

10

12.

How long have you been in your current position at this board/agency? Circle your
answer.

— A less than one year
B 1-2 years
C 3-5 years
D 6-10 years
E more than 10 years
—►12 a.

If less than one year in your current position, what position did you have prior
to your current one?

13 . Do you, personally, have a caseload of clients/families for which you are responsible?
Circle your answer.
NO

p YES
'

1

> 13a.

If yes, what is the percentage of your caseload that has a developmental
disability (as compared to a mental illness diagnosis)?
__________ % has a developmental disability

1 3 b . What are the age ranges of the persons who have a developmental
disability? Please indicate the percentage of your caseload between the
following ages.
__________ % 0 and 18 years of age
__________ % 19 and 26 years of age
__________ % older than 26 years of age
=100 % of caseload with a developmental disability
1 4 . Do any of your immediate family members (i.e., brother, sister, daughter, son, parent,
spouse) have a developmental disability? Circle your answer.
NO
YES
If you have anything else that you would like to share with us, please write your comments on
the back page of this survey.

‘Thank-you fo r your time and thoughts

112

MICHIGAN

STATE UNIVERSITY
EA ST L A N SIN G • M IC H IG A N • 48824.1 117

D E PA R T M E N T O F P SY C H O L O G Y
P SY C H O L O G Y RESEARCH B U ILD IN G

March 13, 1992

Dear Community Mental Health Professional,
The future direction of community mental health services for individuals with developmental
disabilities has been a major focus of debate among policy makers, service professionals,
individuals and their families. Much of this debate has been focused on services provided to
support families caring at home for their member(s) who has a developmental disability, commonly
referred to as "family support". In Michigan, family support services began in the early 1980s as
pilot demonstrations and have subsequently been incorporated into ongoing services delivery at
several boards. As a community mental health services professional, your opinic is about this
trend and about family support services in general are important to future mental health services
planning.
You are one of a select few individuals who arc being asked to give their opinions about
support services for families. You were selected to answer this survey based on your position
with your local community mental health center or its contracted agency. Because we are interested
in the opinions of a variety of community mental heaidi professionals, including administrators,
board committee members, and direct services staff, your answers are important.
Your participation in this survey is voluntary. However, in order to ensure that the results will
truly represent the thinking of mental health professionals at differing administrative levels across
the state, it is important that every survey be completed and returned by the person to whom the
survey was sent. The survey should only take you 15-25 minutes to complete. The time you
spend now will greatly benefit mental health services recipients, service professionals, and the
community in general. The results of this research will be made available to mental health services
policy makers, service professionals, and other interested citizens.
You may be assured of complete confidentiality. The survey has an identification number for
mailing purposes only. This is so we may check your name off of the mailing list when your
survey is returned. Your name will never be placed on the survey. All research results will be
presented in aggregate form, individual answers will not be identified. The return of an answered
survey will indicate your consent to participate in this project.
If you have any questions or need clarification, please contact Kelly Hazel at (517)394-6428,
or leave a message at the Psychology Department at Michigan State University, (517)355-9561,
and she will return your call as soon as possible.
Thank-you for your assistance.

Sincerely,

Kelly L. Hazel, M.A.
Principal Investigator
S 1 S U is a n A ffir m a tiv e A c tio n /E q u a l O p p o r tu n ity In s titu tio n

113

STATE

JOHN

OF

M ICHIGAN

ENGLER.

Governor

DEPARTM ENT OF M ENTAL HEALTH
LEWIS

CASS

LANSING.
JA M ES

K

BUILDING

MICHIGAN
HAVEMAN

48913
JR

D irector

M a r c h 2,

1992

TO:

Community Mental

Health Services Professionals

FROM:

W i l liam L. M c S hane

SUBJECT:

O p i nion Survey R e g a rding Family Support

Services

Mi c h i g a n State University, in c o ope r a t i o n with the S e rvices Research Unit of the
Depa r t m e n t of Mental Health, is cond u c t i n g a study of the opinions of community
mental
health
professionals.
The purpose of
the
study
is to determine
professionals'
views regarding services for families with
members who have
d e v e l o p m e n t a l disabilities, commonly referred to as "familv support services".
Kelly Hazel, who is u n d e r t a k i n g this study as part of the requirements for the
c o m p l e t i o n of her doctoral dissertation, is the principal investigator.
Dr.
Sandra
E.
Herman
of
the
Services
Research
Unit will
be overseeing
the
D e p a r t m e n t ' s role m
the study.
Your p a r t i c i p a t i o n in this project is strictlv voluntary.
However, since your
answers are important to future planning, I encourage you to complete the
encl o s e d survey as soon as possible.
Your answers will help to ensure that the
results
of
this
study
are
representative
of
c ommunity
mental
health
professionals'
opinions.
You can be assured
that
your
answers will be
confidential.
If you have any q u estions about the survey, please contact Ms.
Hazel at M i c h i g a n State U n i v ersity
'517)
355-9561,
or Dr. Herman at the
D e partment of Mental Health (517) 373-3862.
SEH/cmc

114

MICHIGAN

STATE U N IV ERSITY

EAST L A N SIN G • M IC H IG A N • 48824-1117

DEPA RTM EN T O F PSY C H O LO G Y
PSY C H O L O G Y RESEARCH BU ILD IN G

March 20, 1992

Dear Community Mental Health Professional,
About one week ago a survey was sent to you, seeking your opinion about family support
services. If you have already completed and returned the survey to me, please accept my sincere
thanks. If not, please do so today. Because it has been sent to only a select few, but
representative, number of community mental health professionals, it is extremely important that
you also be included in the study if the results are to accurately represent the opinions of
community mental health professionals in Michigan.
If for some reason you did not receive the survey, or it has been misplaced, please call me right
now (517-349-6428 or 517-355-9561) and I will get another one in the mail to you right away.
.S in n n rn lv

Kelly L. Hazel, M.A.
Principal Investigator

M S U is a n A ffir m a tiv e A c tio n /E q u a l O p p o r tu n ity in s titu tio n

115

M IC HIG A N STATE UNIVERSITY

D EPA R T M EN T O F PSY C H O LO G Y

EA ST L A N S IN G • M IC H IG A N • <48824*1117

P S Y C H O L O G Y RESEARCH BU ILD IN G

March 27,1992

Dear Community Mental Health Professional,
About three weeks ago I wrote to you seeking your opinion on family support services. As of
today I have not yet received your completed survey.
I have undertaken this study because of the belief that opinions of community mental health
professionals such as yourself, should be taken imo account in the formation of future policies
regarding services for people with developmental disabilities and their families.
I am writing to you again because of the importance each survey has to the usefulness of this
study. You were selected to participate based on your position with your local community mental
health board or its contracted agency. Only a few professionals at each administrative level at each
board or contract agency have been being asked to give their opinions about family support
services. In order for the results of this study to truly represent the opinions of community mental
health professionals such as yourself, it is essential that each person selected return their survey.
So that the reporting of the results of this survey be timely, we need to receive all surveys by April
10, 1992.
If for some reason you did not receive the survey, or it has been misplaced, please call me now
(517-394-6428 or 517-355-9561) and I will get another one in the mail to you today.
Thank-you for your assistance.

Sincerely,

Kelly L. Hazel, M.A.
Principal Investigator

M S U is a n A ffir m a tiv e A c tio n /E q u a l O p p o r tu n ity I n s titu tio n

116

M IC H IG A N STATE UNIVERSITY

D E P A R T M E N T O F PSY C H O L O G Y

EA ST L A N S IN G • M IC H IG A N • 48824-1117

P S Y C H O L O G Y RESEARCH B U ILD IN G

April 10, 1992

Dear Community Mental Health Professional,
About five weeks ago, I wrote to you asking you to respond to the 'Community Mental Health
Professionals’ Opinion Survey Regarding Family Support Services". As of today 1 have not yet
received your completed survey.
1 am writing to you again because of the importance each survey has to the usefulness of this
study. You were selected to participate based on your position with your local community mental
health center or its contracted agency. Only a few professionals at each administrative level at each
center or contract agency have been asked to give their opinions about family support services.
This is a personal opinion survey. Only those to whom the survey was addressed should answer
it Please, do not pass it on to another person to answer. In order for the results of this study to
truly represent the opinions of community mental health professionals such as yourself, it is
essential that each person selected return their survey. At this time, we would greatly appreciate
receiving your completed survey no later than April 24, 1992.
The survey should only take you 15-25 minutes to complete. The time you spend now will
greatly benefit mental health services recipients, service professionals, and the community in
general. The results of this research will be made available to mental health services policy
makers, service professionals, and other interested citizens.
In the event that your survey has been misplaced, a replacement is enclosed.
Thank-you for your assistance.

Kelly L. Hazel, M.A.

M S U is a n A f f i r m a t t v t A c t t o n / E q u a i O p p o r tu n ity In s titu tio n

A P P E N D IX B
UCRIHS Approval Letter

A P P E N D IX B
UCRIHS Approval Letter
M I C H I G A N STATE U N I V E R S I T Y
EA ST L A N SIN G • M IC H IG A N • 4K824-1046

O FFICE O F VICE PR ESID EN T FOR RESEARCH
A N D D E A N O F T H E G R A D U A T E SC H O O L

F e b r u a r y 14, 1992

K e l l y L. Hazel
P s y c h o l o g y De p a r t m e n t
135 S n y d e r Hall
RE:

C O N F I G U R A T I O N S OF F A M I L Y S U P P O R T SE R V I C E S IN MI C H I G A N :
STRUCTURE, C O N T E X T A N D A T T I T U D I N A L INFLUENCES, IRB # 9 2 - 0 5 2

ORGANIZATION

Dear Ms. H a z e l :
The above pr oj e c t is e x e m p t fr om full UCRIHS review.
T h e p r o p o s e d r e sear ch
protocol has been r e v i e w e d by a n o t h e r co m m i t t e e member. T h e rig h t s and w e l f a r e
of human subjec ts app e a r to be p r o t e c t e d and y o u have approv al to c o n d u c t the
research.
You are re m i n d e d that U C R I H S approval is valid for one c a l e n d a r year.
If you
plan to c o n t i n u e this p r o j e c t beyond one year, p l e a s e m a k e p r o v i s i o n s for
o b t a i n i n g a p p r o p r i a t e U C R I H S approval one m o nth p r i o r to F e b r u a r y 14, 1993.
An y c h a nge s in p r o c e d u r e s in v o l v i n g human subjects mu s t be r e v i e w e d by U C RIHS
p r ior to in itia tio n of the change. U C R I H S must also be n o t i f e d p r o m p t l y of any
p r o b l e m s ( u nexp ect ed side e f fect s, compla int s, etc.) in v o l v i n g h u m a n subjects
d u r i n g the cou r s e of the work.
T h a n k y o u for b r i n g i n g this p r o j e c t to my attention.
help, ple a s e do not h e s i t a t e to let me know.

If I can be of any future

Sincerely,

Da vid E. Wright, Ph.I
U n i v e r s i t y Co m m i t t e e
H u man S u bjec ts (UCRIHS)

Involving

D E W/ d e o
cc:

Dr. Fr ank Floyd

M S U u a n A f fir m a tiv e A c tio n /E q u a l O p p o r tu n ity I n s titu tio n

117

A P P E N D IX C
Preliminary Analyses Results Tables

A P P E N D IX C

Preliminary Analyses Results Tables

Table C -l: Services identified as supportive ol'fam ilies of people with developmental
disabilities.
Participants
Family Support Services

% (/V=339)

Respite Care (in general)

81

In-home Respite Care

17

Out-ol-home Respite Care

20

Family Friend Respite

15

Case M anagement

68

Advocacy

11

Information and Referral

13

Support Groups (in general)

11

Fam ily Support Groups

3

Parent Support G roups

9

Sibling Support G roups

5

Counseling or Therapy (in general)

22

Family Counseling

30

Individual Counseling

8

Behavior M anagement Training or Consultation
In-home Behavioral Training or Program Aid
Parent Education/Skills Training

30
4
20

Parent to Parent Network

9

Family Support Newsletter

8

In-home Services (in general)

19

In-hom e Nursing

7

Special Therapies

4

Adaptive Equipment

16

Home Renovations

9

After School or Latch Key Program

3

Recreation or Family Activity Program

8

Sum mer Recreation Program

6

118

119
Table C -l (cont'd).
Participants
Fam ily Support Services
Cam pership

% (N - 3 3 9 )
7

M odel W aiver 2 Program

25

Subsidy

54

Financial Assistance

4

Em ergency or Crisis Services

3

Parent Advisory Committee

4

Other Services and Supports
Perm anency Planning

4

Guardianship or Adoption Planning

3

Psychological or Psychiatric Consultation

9

A ssessm ents

6

M edical or Health Related Consultation

3

Foster Care

5

Day Activity or Partial Day Program

12

W ork Activity or Supported Em ploym ent Program

8

Residential Services

5

C lient Education Program (e.g., sexuality, life skills, etc.)

2

Outpatient or Life Consultation Program

4

Transportation

3

120
Table C-2: Descriptive statistics of attitude items.

Strongly

Strongly

Disagree Disagree Neutral

Agree

Agree

1

2

3

4

5

W hat do you think?

M cantSD l %

%

%

%

%

I feel that programs and services for

2.54(.91)

8

49

26

16

4.29(.67)

1

1

7

52

1,75(.84)

41

49

4

4

4.11 (85)

2

2

14

47

57

11

4

people with disabilities should take
priority over support services for
families.
I would like to see my board/agency

40

expand or improve its family support
services.
Since m ost families with children with
disabilities are supported by the school
system , it is not necessary for mental
health to also provide services for them.
I am personally committed to the

35

developm ent and/or expansion o f family
support services at my board/agency.
Fam ily support services require too

1.96(.81) 27

m uch change in the way my board/
agency provides services to our clients.
Parents are very knowledgeable and
com petent resources and should be
involved in all services planning for
their family m em ber who has a
developmental disability.

4.27(.86)

2

7

44

45

121
Table C-2 (cont'd).

Strongly

Strongly

Disagree Disagree Neutral
1
W hat do vou think?
There are needs in our comm unity that

Meant SD) %
2.36(.89) 13

2
%
54

Agree

Agree

3

4

5

%

%

%

20

13

1

42

42

43

17

64

17

are much more pressing which should
be addressed first, before we develop
family support services any further.
I feel that all support services to people

4.19(.89)

1

with a developmental disability who are
living at home are more effective when
they are family centered and family driven.
Fam ily support services take up too

1.77(.66) 33

60

much staff lime that would be better
devoted to other, more needed services.
I think that family support should be a

3.67(.88) 0

greater percentage of my board's overall
budget.
I lcel that the focus o f services should

2.21 (.88)

17

57

14

be on the person with disabilities,
not the family.
I would like to encourage and help other 3.94(.70)

1

16

service providers implem ent or improve
family support services.
I would like to sec my board/agency
discontinue providing family support
services.

1.36(.69) 72

25

122
Table C-2 (cont'd).

Strongly

Strongly

Disagree Disagree Neutral
1

2

3

Agree
4

Agree
5

W hat do you think?__________________ Mean (SD) %______ %______ %______ %______ %
I personally do not see a strong enough

1.58(.70) 52

41

5

2

0

3.48(.97) 2

15

32

37

14

2.30(.97) 19

49

19

12

2

21

47

24

need in our comm unity to warrant any
further developm ent o f family support
services at our board/agency.
I think that my board/agency should
redirect current resources to provide
more support services for families.
Too many families are not capable of
functioning as care givers for their family
m em ber who has a developmental
disability.
I leel that supporting families should be

3.87(.89)

1

6

a top priority at my board/agency.
M any activities and services related to

2.72(1.12)

12

38

20

24

5

4.48(.53) 0

0

2

49

49

family support are not within the domain
of mental health responsibility, other
service providers should be providing
these supports.
I feel that parents' views and opinions
regarding needed services and service
delivery practices are very important to
overall services planning.

123
Table C-2 (cont'd).

Strongly

Strongly

Disagree Disagree Neutral
1
W hat do

you

think?

2

3

Agree
4

Agree
5

Mcan(SD) %______%_______%______ %_______%

I am not at all convinced that family

1.60(.61) 46

50

4

1

0

1.37(.56) 66

32

1

1

0

1.77(.68) 35

55

7

2

0

4.38(.67) 0

1

5

47

46

4.04(.80) 0

4

18

49

30

1

11

59

29

support services are an effective or
appropriate way of helping people
with disabilities.
I personally do not think there is a need
for family support services.
Families have a lot of resources for help
and support from other service providers,
mental health should not be spending its
scarce resources too.
I think parents should be actively
involved in planning overall services
for people with developmental
disabilities.
I feel that family support concepts and
practices should be incorporated into
the overall way of providing services
to individuals with disabilities and their
families. Family support should be my
board/agency's way of doing business.
I would like to know what other boards/ 4.15(.66) 0
agencies are doing in regards to family
support services.

124
Table C-2 (cont'd).

Strongly

Strongly

Disagree Disagree Neutral
1

2

3

Agree
4

Agree
5

W hat do vou think?___________________ Mean(SD) %______%______ %______ %_______%

I think that staff resources should be
redirected to put more emphasis on
supporting families

N=339

3.57(.88)

1

9

37

39

14

125
Table C-3: Descriptive statistics of perceptions of support for services from others.

Mean OSD)

Highly

Som ewhat

Som ewhat

Highly

Object

Object

Neutral

NA

1

2

3

4

5

%

%

%

%

%

%

Approve Approve

Staff you supervise

4.43(.76)

38

<1

<1

7

19

35

Professional peers

4.35(.75)

7

<1

2

7

38

45

Immediate supervisor

4.46(.74)

13

1

1

7

29

50

Board's chairperson

3.96(.85)

14

<1

3

22

35

25

3.87(.83)

15

<1

4

23

38

20

4.15(.81)

12

0

3

15

36

34

4.09(.87)

12

<1

4

15

36

32

4.42(.78)

14

<1

2

7

28

49

4.47(.76)

7

<1

2

7

28

55

support services staff

4.67(.63)

11

<1

<1

5

17

66

Self

4.73(.53)

2

<1

0

2

21

74

Board committee
members
Agency's executive
director
Agency's top
administration
Agency's developmental
disabilities services
administration
Agency's developmental
disabilities services staff
Agency's family

N = 339

126

Table C-4: Descriptive statistics o f belief/perception items regarding forces which hinder or
facilitate the implementation o f family support services.

W hat do vou think?
M y board/agency has sufficient

Very

Som ewhat

Som ewhat Very

I Untrue

Untrue

Neither

True

1

2

3

4

5

MeantSD) %

%

%

%

%

2.09(1.16) 39

36

4

20

1

1.63(.91) 59

27

43

30

True

capacity to provide adequate support
services to all families in our
com m unity who need/want them.
In general, people at my board/
agency are not interested in
providing services for families.
My board/agency lacks sufficient

3.77(1.17) 5

15

2.84(1.11)

13

28

28

27

3.24( 1.26)

16

13

13

48

10

16

11

48

20

staff resources to provide adequate
family support services.
Providing support services for
families is a top budget priority
at my board/agency.
Stale Department funding allocations
have helped my board/agency's ability
to provide family support services.
M y board/agency has provided the
necessary opportunities for staff to
improve their knowledge and learn
the skills required to provide support
services to families.

3.62(1.11) 5

127
Tabic C-4 (cont'd).

Very
Untrue
1

Som ewhat

Somewhat Very

Untrue

Neither

True

2

2

4

True
5

W hat do you think?___________________ Meant,SD) %______ %_______%______%_____ %
There is a sufficient number o f

2.40(1.23) 26

40

5

23

5

2.97(1.25) 14

27

16

33

10

2.92(1.24) 15

28

16

33

8

3.68(1.17) 5

15

14

38

18

3.64(1.13) 5

15

12

45

22

3.98(.99) 2

9

10

46

32

qualified service providers (e.g. for
respite care, in-home nursing, in-home
behavioral m anagement training, etc.)
in our com m unity to support my board/
agency's efforts to provide family
support services.
Families in our community have been
very active in advocating for the
developm ent and/or expansion of family
support services at my board/agency.
In general, people at my board/agency
lack a clear understanding of what
family support services are.
My board/agency has lacked the
funding capacity to develop family
support services.
My board has hired people who have
the expertise needed to provide support
services to families.
The various agencies and professionals
who work with families in our community
readily refer families to my board/agency
for the services we provide.

128
Table C-4 (cont'd).

Very
Untrue
1
W hat do vou think?
Our mental health board committee

Som ewhat

Som ewhat Very

Untrue

Neither

True

2

3

4

True
5

Mean(SD) %______ %_______%______%______%
3.44(.92) 2

13

33

42

10

3.59(1.25) 8

15

13

38

27

2.47(1.12) 23

33

19

24

1

3.56(.92) 1

13

26

46

13

2.77(1.08) 10

38

22

26

4

3.26(1.26) 10

22

18

33

18

members are committed to the
development and/or expansion o f family
support services.
My board/agency has not been able to
keep up with the demand lor family
support services, families have had to
wait to get into the program or to receive
services.
Families have been reluctant to take
advantage of the services available at
my board/agency.
M y board/agency's administration is
committed to the development and/or
expansion of family support services.
There is a general lack o f com m unity
interest and involvement in supporting
individuals with disabilities and their
families.
The system does not recognize the
family as the client, only the individual
with disabilities, which hinders our
efforts to implem ent support services
for families.

129
Table C-4 (cont'd).

Very

W hat do vou think?

The attitudes and practices of most

Somewhat

Som ewhat Very

Untrue

Untine

Neither

True

I

2

3

4

5

Mean(SD) %

%

%

%

%

18

41

17

21

4

1.59(.83) 59

27

11

3

1

2.51(1.12)

Tme

service providers in our community
are that support services should be
provided only to the individual with
disabilities, they do not involve or
support the family.
Families in our community really
don't w ant fam ily support services.
(N=339)

130
Table C-5: Principal axis factoring of attitude items after varimax rotation.
Factors
I

II

III

IV

V

VI

8.0

2.0

1.6

1.4

1.2

1.0

27.8

5.5

3.6

3.0

2.2

1.8

D on’t think there is a need for FSS (R)

J_6

.21

.14

.06

.12 .00

Not convinced FSS is effective or appropriate (R)

JS7

.29

.04

.12

.20 .08

Parents views are important to services planning

.55

.39

.11

.13 -.00 .01

Like to see agency discontinue FSS (R)

.53

.16

.09

.06

.13 .11

.52

.24

.39

.12

.13 .10

.49

.01

.23

.26

.34 .28

M

.39

.13

.19 -.04 .03

.36

.19

.28

.23 .31

.35

.06

.15

.26

.25 .09

.18

37

.05 -.01

.04 .18

.12

3 2 -.02

.29

.18 .25

.34

.48

.22

.19

.15 ■
-.01

Services more effective when family centered and driven

.20

.47

.16

.11

-.21
.13 ■

Personally com m itted to FSS

.23

36

.36 -.09

.18 .07

Supporting families should be top priority at agency

.27

36

.29

.29

.12 ■
-.02

Like to help others implement FSS

.15

39

.23

.09

.01

Agency should redirect resources to provide more FSS

.08

.06

34

.19

.03 .09

Staff resources should be redirected, more em phasis FSS .11

.25

M

.19

.02 ■-.11

FSS should be greater percentage of agency budget

.19

.20

.61

.13 -.05 .36

W ould like to see agency expand or improve FSS

.36

.29

.43 -.19 -.14 .27

AW ould like to know what other agencies are doing

.22

.11

.27

Factor statistics
Eigenvalue (before rotation)
Percent of variance (after rotation)
Item s (abbreviated)

Do not see strong enough need to warrant further
developm ent of FSS (R)
Families have help from others, mental health
should not be using resources too (R)
Parents should be actively involved in services planning
Family support takes up too much staff lime
better devoted to other services (R)
AFam ilies supported by school system, no need
for mental health to also do so (R)
Parents are knowledgeable and competent resources
and should be involved
Fam ilies not capable care givers (R)
Fam ily support concepts should be agency's way of
doing business

.02

.06 -.25 .05

131
Table C-5 (cont'd).
Factors
Items (abbreviated)_______________________

I

II ITT

TV

V VI

Focus on person with disability, not family (R)

II

.11 .21 J>8 -.05 .02

11

.11 .06

^48

.12 .07

12

.0 4 .12

.06

.58 -.01

FSS require too much change (R)

16

.19 -.1 0 -.0 0

.5 0 .02

AFSS not in domain of mental health responsibility (R)

09

.14 .18

.19

.33 .19

AOther needs should be addressed fust (R)

26

.10 .33

.22

.09 ,4 4

Services for individuals priority over services
for family (R)
Agency should find out more about FSS before
going further (R)

(R) identifies items which were reverse scored for factor analysis.
A identifies items deleted in final factor analysis.
Note. Six factors accounted for 57% of the variance prior to rotation.

132
Table C-6: Principal axis factoring of belief/perception items after varimax rotation.
Factors
Factor statistics
Eigenvalue (before rotation)
Percent o f variance (after rotation)

I

II

III

IV

8.2

2.6

2.2

1.9

1.5 1.0

25.1

7 .0

5.3

4.4

2.7 1.7

V

VI

Item s (abbreviated')
Board comm ittee members are supportive

33

.21 -.12

.09 .21

Agency's executive director is supportive

3A

.23

.03

.03 .1 1 .32

Agency's top administration is supportive

14

.33

.08

.10 .08 .03

Board chairperson is supportive

31

.38

.01

.08 .06 .08

31

.02

.26

.03 .07 -.11

32

.10

.21 -.02 .1 1 -.12

People at agency are not interested in providing FSS (R)

11

.18

.19 -.01 .16 .00

Providing FSS is a top budget priority

15

-.01

.36

.12 .06 -.05

People at agency lack clear understanding of FSS (R)

M

.10

.23

.01 .28 -.12

Self is supportive of FSS

.12

34

.09

.05 -.11

Agency's developmental disabilities staff is supportive

.23

3 1 -.04

FSS staff is supportive

.13

31

.13 -.02 .04 -.12

Professional peers are supportive

.14

M

.1 1 -.08 .15 .09

Staff supervised are supportive

.18

.64

.15

.07 .18 .22

Developmental disabilities administration is supportive

.48

H

-.11

.19 .08 -.14

Im m ediate supervisor is supportive

.29

.49

.20

.0 6 - .14 .30

-.00

.39

1 1 -.05 .23 .05

Com m unity agencies refer families to agency for services

.15

.10

.53 -.1 2 --.08 .09

Families active in advocating for FSS

.18

.01

.47

.05 .15 -.04

AAgency hired staff with FSS expertise

.32

.14

A5

.1 1 .13 -.17

Families reluctant to take advantage of services (R)

.04

.16

A l -.15 .15 .01

AAgeney provided FSS staff training opportunities

.35

.16

.37

.26 .06 -.05

.09 -.08

.32

.19 .22 .02

.03

.32

.31

.13 .13

.06 -.03 -.02

15

.12 .12

.23

Board committee members committed to
develop/expand FSS
Agency's administration is comm itted to
develop/expand FSS

Fam ilies don't want FSS (R)

.02

.05 .19 -.15

ASufficient resources in comm unity to support agency's
efforts to provide FSS
AState department funding allocations support
agency's FSS
Agency has sufficient capacity

.09

133
Table C-6 (cont'd).
Factors
Items (abbreviated!______________________________

III III

IV

V

VI

Agency lacks funding capacity (R)

.04

.11

.1 0

JLL.0 0 - .0 3

Agency lacks sufficient staff (R)

.12

.01

.05

J>9.0 7 - .1 4

Agency not able to keep up with demand, families wait (R )-.03 -.03 -.3 0

*49 .02 .04

Attitudes and practices in comm unity support individual
services, not family services (R)

.18

.15

.13

.02 J58 .19

.12

.13

.15

.03 *43 -.09

-.00 -.0 0

.15 .40 -.07

Lack of com m unity interest in supporting individuals and
families (R)
System does not recognize family as client (R)

.28

(R) identifies items which were reverse scored for factor analysis.
A identifies items not included in final factor analysis.
Note. Six factors accounted for 57% of the variance prior to rotation.

134
Table C-7: Attitude factors IV and V and belief/perception factor V internal reliability.

Inter-item r
Items (abbreviated')________________

J_2

Corrected
Item-total r

Alpha

Attitude factor IV :
Individual program priority over FSS
Focus o f service on individual, not family

.38
.38

.38

.55

Attitude factor V (behavioral intentions):
FSS require loo much change
Should find out more about FSS

.37
.37

.37

.54

Belief/perception factor V :
Lack of community interest
System does not recognize family

.24
.21

.28

Attitudes and practices of community
service providers don't support families

.24

.23

.29

.47

135
Table C-8: Correlation between criteria and items and factors not included in final predictor
scales.

Criteria
Num ber of
Factors/item s (abbreviated)__________________

Administrative

Services_F ormalization

Other needs in com m unity are more pressing

-.39**

FSS not within mental health domain

-.24

.00

.12

-.11

-.25

-.21

. 11

.26

-.27*

-.10

Individual program priority over FSS

-.26

-.18

Focus of service on individual, not family

-.18

.05

Like to know what other agencies are doing

-.05

Fam ilies supported by school system, not
necessary for menial health to do also
State funding allocations have helped

Attitude factor IV :

Attitude factor V (behavioral intentions):

-.35**

-.34*

FSS require too much change

-.24

-.27*

Should find out more about FSS

-.39**

-.33*

-.07

-.15

Lack of comm unity interest

-.09

-.13

System does not recognize family

.09

-.08

-.16

-.1 4

Belief/perception factor V :

Attitudes and practices of community
service providers don't support families

* p < .0 5 , ** p < .01, two-tailed.

BIBLIOGRAPHY

Alexander, E. (1985). From idea to action: Notes for a contingency theory o f policy
implementation process. Adm inistration and Society. JL6(4), 403-426.
Agosta, John. (1989). Using cash assistance to support fam ily efforts. In G. H. S. Singer
& L. K. Irvin (Eds.), Support for care giving families: Enabling positive adaptation to
disability (pp. 189-204). Baltimore: Brookes.
Agosta, J. M. & Bradley, V. J. (Eds.) (1985). Fam ily care for persons with developm ental
disabilities: A growing com m itm ent. Boston, MA: Human Services Research Institute.
Agosta, J., Bradley, V., Rugg, A., Spence, R. & Covert, S. (1985). D esigning program s
to support family care for persons with developmental disability: Concepts to practice.
Boston, MA: Human Services Research Institute.
Agosta, J. M., Jennings, D. & Bradley, V. J. (1985). Statewide fam ily support program s:
National survey results. In J. M. Agosta & V. J. Bradley (Eds.), Fam ily care for
persons with developmental disabilities: A growing com m itm ent (pp. 94-112). Boston,
MA: Human Services Research Institute.
Ajzen, I. & Fishbein, M. (1980). Understanding attitudes and predicting social behavior.
Englewood Cliffs, NJ: Prentice Hall.
Arneaud, S. & Herman, S. E. (1989). Fam ily support subsidy program: Annual report to
the governor and the legislature. Lansing: M ichigan Departm ent of Mental Health.
Attewell, P. & Gerslein, D. R. (1979). Governm ent policy and local practice. Am erican
Sociological Review . 44. 311 -327.
Ayer, Sam. (1984). Com m unity care: Failure of professionals to m eet family needs. Child:
Care. Health and D evelopm ent. 10. 127-140.

136

137
Backer, T. E., Liberm an, R. P., Kuehnel, T. G. (1986). D issem ination and adoption of
innovative psychosocial interventions. Journal of Consulting and Clinical Psychology.
54(1), 111-118.
Baldridge, J. V. & Burnham, R. A. (1975). Organizational innovation: Individual,
organizational and environm ental impacts. Administrative Science Quarterly. 20. 165176.
Bardach, Eugene. (1977). The implementation game: W hat happens after a bill becomes a
law . Cam bridge, MA: M IT Press.
Beckm an, Paula J. (1983). Influence of selected child characteristics on stress in families
o f handicapped infants. Am erican Journal of Mental Deficiency. 88(2). 150-156.
Berman, P. & M cLaughlin, M. W. (1977). Federal program s supporting educational
change. Vol. VII: Factors affecting implementation and continuation. Santa M onica,
CA: Rand Corporation.
Berman, P. & M cLaughlin, M. W. (1978). Federal program s supporting educational
change. Vol. VIII: Implementing and sustaining innovations. Santa M onica, CA: Rand
Corporation.
Blacher, Jan. (Ed.) (1984). Severely handicapped young children and their families:
Research in review . New York: Academic Press.
B lakely, C. H., M ayer, J. P., G otlschalk, R. G., Schm itt, N„ D avidson, W . S., et al.
(1987). The fidelity-adaptation debate: Implications for the implementation of public
sector social programs. A m erican Journal of Com m unity Psychology. 15. 253-268.
Boluck, S. & W insberg, B. G. (1991). Effects of respite on mothers o f school-age and
adult children with severe disabilities. Mental Retardation. 29(1), 43-47.
Breslau, N. & Prabucki, K. (1987). Siblings of disabled children: Effects o f chronic stress
in the family. Archives o f General Psychiatry. 44. 1040-1048.
Brom ley, B. & Blacher, J. (1989). Factors delaying out-of-hom e placem ent o f children
with severe handicaps. American Journal o f Mental Retardation. 94(3), 284-291.

138
Bronfenbrenner, U. (1986). T he ecology of the family as a context for human
developm ent: Research Perspectives. Developmental Psychology. 22(6). 723-742.
B row ning, R. P., M arshall, D. R. & Tahh, D. H. (1981). Im plem entation and political
change: Sources of local variations in federal social programs. In D. A. M azm anian and
P. A. Sabatier (Eds.), Effective policy implem entation (pp. 127-146). Lexington, MA:
Lexington Books.
Bruininks, R. H. & Krantz, G. C. (Eds.) (1979). Fam ily care of developm entallv disabled
members: Conference proceedings. Minneapolis: University of Minnesota.
Byrne, E. A. & Cunningham , C. C. (1985). The effects of mentally handicapped children
on families: A conceptual review. Journal of Child Psychology & Psychiatry & Allied
D isciplines. 26(6). 847-864.
C astellani, P. J., Dow ney, N. A. Tausig, M. B. & Bird, W. A. (1986). A vailability and
accessibility of family support services. Mental Retardation. 24(2), 71 -79.
Cavanagh, J. & Ashm an, A. F. (1985). Stress in fam ilies with handicapped children.
Australian & New Zealand Journal o f Developmental Disahilities. 11 (3). 151-156.
C enter on Hum an Policy (1987, Sept.). Fam ilies for all children. Syracuse, NY: author.
Cernoch, Jennifer. (1989). Respite care: A gift of time. NICHCY News D igest. 12. 1-9.
C hetw ynd, Jane. (1985). Factors contributing to stress on m others caring for an
intellectually handicapped child. British Journal of Social W ork. 15(3), 295-304.
Cohen, Shirley. (1982). Supporting families through respite care. Rehabilitation Literature.
43(1-2), 7-11.
C ohen, S. Agosta, J., Cohen, J. & W arren R. (1989). Supporting fam ilies o f children
with severe disabilities. Journal of the Association for Persons with Severe H andicaps.
14(2), 155-162.
Cole, D. A. & M eyer, L. H. (1989). Impact of needs and resources on fam ily plans to seek
oul-of-hom e placement. American Journal of Mental Retardation. 93(4), 380-387.
C ronbach, Lee J. (1951). Coefficient alpha and the internal structure of tests.
Psychom etrika. 16(3). 297-334.

139
Darling, Rosalyn B. (1983). Parent-professional interaction: The roots of
m isunderstanding. In M. Seligman (Ed.), The family with a handicapped child:
U nderstanding and treatm ent (pp. 95-121). New York: Grune & Stratton.
Darling, Rosalyn B. (1988). Parental entrepreneurship: A consum erist response to
professional dom inance. Journal of Social Issues. 44( 1), 141-158.
Davis, Donald D. (1982). Innovation adoption and organization change: Program
evaluation in gerontology. Unpublished doctoral dissertation, Michigan State
U niversity, East Lansing.
DeVellis, Robert F. (1991). Seale development: Theory and applications. Newbury Park,
CA: Sage.
Dow ns, G eorge W ., Jr. (1976). Bureaucracy, innovation, and public policy. Lexington,
MA: Lexington Books.
Dunst, C. J. & Trivette, C. M. (1981). An enablem ent and em pow erm ent perspective of
case m anagement. Topics in Early Childhood Education. 8(4), 87-102.
Dunst, C., Trivette, C. & Deal, A. (1988). Enabling and empowering families: Principles
and guidelines for practice. Cambridge, MA: Brookline Books.
Dunst, C. J., Trivette, C. M ., Gordon, N. J. & Pletcher, L. L. (1989). Building and
m obilizing informal fam ily support networks. In G. H. S. Singer & L. K. Irvin (Eds.),
Support for care giving families: Enabling positive adaptation to disability (pp. 121141). Baltimore: Brookes.
Edwards, G eorge C. (1980). Implementing public policy. W ashington, DC: Congressional
Quarterly Press.
Elmore, Richard F. (1979). Backward mapping: Implementation research and policy
decisions. Political Science Quarterly. 94(4), 601-616.
Elmore, Richard F. (1985). Forward and backward mapping: Reversible logic in the
analysis of public policy. In K. H anf and T. Tooneu (Eds.), Policy implem entation in
federal and unitary systems (pp. 33-70). Dordrecht, Netherlands: M arlinus Nijhoft.

140
Factor, D. C., Perry, A. & Freem an, N. (1990). Stress, social support, and respite care
use in families with autistic children. Journal o f Autism and Developmental Disorders.
20(1), 139-146.
Fairw eather, G. W. & Tornatzky, L. G. (1977). Experimental methods for social poliev
research. New York: Pergamon.
Fairw eather, G. W „ Sanders, D. H. & Tornatzky, L. G. (1974). Creating change in
mental health organizations. New York: Pergamon.
Fox, John. (1991). Regression diagnostics. Newbury Park, CA: Sage.
Frank, S. J. & Davidson, D. S. (1983). Ideologies and intervention strategies in an urban
sample of drug-abuse agencies. American Journal of Com m unity Psychology. 11(3).
241-259.
G allagher, J. J., Beckm an, P. & Cross, A. H. (1983). Fam ilies of handicapped children:
Sources of stress and its amelioration. Exceptional C hildren. 50, 10-19.
Gallagher, J. J. & Gallagher, G. G. (1985). Fam ily adaptation to a handicapped child and
assorted professionals. In H. R. Turnbull & A. P. Turnbull (Eds.), Parents speak out:
Then and now (pp. 233-242). Colum bus, OH: Charles E. Merrill.
Gallagher, J. J. & Vietze, P. M. (Eds.). (1986). Fam ilies o f handicapped persons:
Research, program s, and policy issues. Baltimore: Brookes.
G hiselli, E. E., Cam pbell, J. P. & Zedeck, S. (1981). M easurem ent theory for the
behavioral sciences. San Francisco: W. H. Freeman.
G laser, E. M., Abelson, H. H. & Garrison, K. N. (1983). Putting know ledge to use:
Facilitating the diffusion o f knowledge and the implementation of planned change. San
Francisco: Jossey-Bass.
Grant, G. & M cGrath, M. (1990). Need for respite care services for care givers of persons
with menial retardation. American Journal of M ental Retardation. 94(6), 638-648.
Hage, J. & Dewar, R. (1973). Elite values versus organizational structure in predicting
innovation. Adm inistrative Science Quarterly. 18. 279-290.

141
Hall, G. E. & Loucks, S. F. (1978). Innovation configurations: Analyzing the adaptations
o f innovations. Austin: University of Texas at Austin, Research and Development
Center for Teacher Education.
Halpern, Peggy L. (1985). Respite care and family functioning in fam ilies with retarded
children. Health & Social W ork. 10(21. 138-150.
Halpern, Robert. (1986). Home-based early intervention: Dim ensions o f current practice.
Child W elfare. 65(4), 387-398.
Havelock, R. C. (1971). Planning for innovation through dissem ination and utilization.
Ann Arbor: University of Michigan.
Hascnfcld, Y. & Brock, T. (1991). Im plem entation of social policy revisited.
Adm inistration and Society. 22(4). 451 -479.
H azel, K. L., Herman, S. E., Thom pson, L., Linklater, A. (1990). Survey of C om m unity
Mental Health Family Support Services. Lansing: M ichigan Department o f Mental
Health.
Herman, S. E. (1983). Fam ily support services: Report on m eta-evaluation study. Lansing:
Michigan Department of Mental Health.
Herman, S. E. (1984). Baseline study of family support services in the State o f M ichigan.
Lansing: Michigan Department of Mental Health.
Herm an, S. E. (1986). Report on family support subsidy program FY 84-85. Lansing:
Michigan Department o f Mental Health.
Herman, S. E. (1991). Use and impact o f a cash subsidy program . Mental R etardation.
29(5), 253-258.
Herman, S. E. & Hazel, K. L. (1987). Fam ily support services in M ichigan: An evolving
system . Lansing: Michigan Department of Mental Health.
Herman, S. E. & Hazel, K. L. (1991, in press). Evaluation of family support services:
Changes in availability and accessibility. Mental Retardation. 29(6).

142
H erm an, S. E., Hazel, K. L. & M arcenko, M. O. (1989). Quality family support
evaluation: Interim report to the Michigan Developmental Disabilities Council. Lansing:
Michigan Department of Mental Health.
H erm an, S. E„ M arcenko, M. O. & Hazel, K. L. (1991). Fam ily support services:
Evaluation of program quality. Lansing: Michigan Department of Mental Health.

H erm an, S. E., Hazel, K. L., Thom pson, T. L. & Linklater, A. (1992a). Fam ily Support
Services in Michigan: Respite Care. Lansing, Ml: Michigan Developmental Disabilities
Council.
H erm an, S. E., Hazel, K. L., Thom pson, T. L. & Linklater, A. (1992b). Fam ily Support
Services in Michigan: Parent Advisory Committees at Community Mental Health.
Lansing, MI: Michigan Developmental Disabilities Council.
H erm an, S. E., Thom pson, T. L., Linklater, A. & Hazel, K. L. (1992). Fam ily Support
Services in Michigan: Community Mental Health Family Support Services. Lansing,
MI: Michigan Developmental Disabilities Council.
Hinckley, E. C. & Ellis, W. F. (1985). An effective alternative to residential placement:
H om e-based services. Journal of Clinical Child Psychology. 14(3), 209-213.
H unter, John E. (1980). Factor analysis. In P. M ongc and J. Capella (eds.), M ultivariate
techniques in human comm unication research (p. 229-257). New York: Academic
Press.
Hunter, J. E. & Hamilton, M ark A. (1992). Path: A program in basica. Statistical software
available from first author, Psychology Department, Michigan State University, East
Lansing, MI 48824-1117.
Ineichen, Bernard. (1986). A job for life? The service needs of mentally handicapped
people living in the community, and their families. British Journal of Social W ork.
16(3), 311-323.
Ingram , H. M. & M ann, D. E. (Eds.). (1980). W hy policies succeed or fail? Beverly
Hills, Sage.

143
Ireys, H. T. & Eichler, R. J. (1988). Program priorities o f crippled children’s agencies: A
survey. Public Health R eports. 103(1). 77-83.
Ireys, H. T., Hauck, R. & Perrin, J. (1985). Variability am ong state crippled children's
service programs: Pluralism thrives. Am erican Journal of Public H ealth. 75(4), 375381.
Johnson, R. W . & O 'Connor, R. E. (1979). Intraagency lim itations on policy
implementation: You can't always get what you want, but sometim es you get what you
need. Adm inistration & Society. 11(2). 193-215.
Joyce, K., Singer, M. & Isralow itz, R. (1983). Im pact of respite care on parents'
perceptions o f quality of life. Mental Retardation. 21 (4). 153-156.
Justice, R. S., O'Connor, C. & W arren, N. (1971). Problem s reported by parents of
mentally retarded children—W ho helps? American Journal of Mental Deficiency. 75(6).
685-691.
Kagan, S. L., Powell, D. R., W eissbourd, B. & Zigler, E. F. (1987). Past
accom plishm ents: Future challenges. In S. Kagan, D. Powell, B. W eissbourd and E.
Zigler (Eds.), America's fam ily support programs: Perspectives and prospects (pp.
365-380) New Haven: Yale University Press.
Katz, R. F. (1984). Em pow erm ent and synergy: Expanding the com m unity's healing
resources. In J. Rappaport, C. Sw ift and R. Hess (Eds.), Studies in em pow erm ent:
Steps toward understanding and action (pp. 201-226). New York: Haworth Press.
Kazak, Anne E. (1987). Fam ilies with disabled children: Stress and social networks in
three sam ples. Journal o f Abnorm al Child Psychology. 15( 1), 137-146.
Kim berly, J. R. & Evanisko, M. J. (1981). Organizational innovation: The influence of
individual, organizational, and contextual factors on hospital adoption of technological
and adm inistrative innovations. Academy of M anagem ent Journal. 24. 689-713.
Kirst, M. & Jung, R. (1980). The utility of a longitudinal approach in assessing
implementation: A thirteen-year view of Title 1, ESEA. Education Evaluation and
Policy A nalysis. 2(5). 17-34.

144
Knoll, J. A., C overt, S., O such, R., O 'Connor, S., Agosta, J., Blaney, B. & Bradley, V.
J. (1990). Family support services in the United States: An end of decade status report.
Cam bridge, MA: Human Services Research Institute.
Krauss, M. W. (1986). Patterns and trends in public services to families with a mentally
retarded m ember. In J. J. G allagher & P. M. Vietze (Eds.), Fam ilies o f handicapped
persons: Research, program s, and policy issues (pp. 237-248). Baltimore: Brooks.
Landesman, S. & Butterfield, E. C. (1987). Normalization and deinstitutionalization of
m entally retarded individuals. American Psychologist. 42(8), 809-816.
Legge, K. (1984). Evaluating planned organizational change. New York: Academic Press.
Levine, R. L., Van Sell, M. & Rubin, B. (1992). System dynam ics and the analysis of
feedback processes in social and behavioral systems. In R. L. Levine and H. E.
Fitzgerald (eds.L Analysis o f dynamic psychological systems Ivol. 1] (pp. 145-266).
New York: Plenum.
Lewin, K. (1951). Psychological ecology. In D. Cartw right (Ed.), Field theory in social
science (pp. 170-187). New York: Harper.
Lewis-Beek, M ichael, S. (1980). Applied regression: An introduction. Beverly Hills:
Sage.
Likert, R. (1932). A technique for the m easurement of attitudes. Archives of Psychology.
No. 140.
Lutzer, V. D. & Brubaker, T. H. (1988). Differential respite needs of aging parents of
individuals with mental retardation. Mental Retardation. 26(1). 13-15.
M ajone, G. & W ildavsky, A. (1978). Implementation as evolution. Policy Studies Review
A nnual. 2. 103-117.
M arc, D. & M acDonald, L. (1988). Respite care: W ho uses it? Mental Retardation. 26(2).
93-96.
M azm anian, D. A. & Sabatier, P. A. (Eds.). (1981). Effective policy im plem entation.
Lexington, MA: Lexington Books.

145
M azm anian, D. A. & Sabatier, P. A. (1983). Implementation and public policy. Chicago:
Scott Foresm an and Co.
M cKinney, B. & Peterson, R. A. (1987). Predictors o f stress in parents of
developm entally disabled children. Journal o f Pediatric Psychology. 12(1). 133-150.
M cLaughlin, M. (1987). Learning from experience: Lessons from policy implementation.
Educational Evaluation and Policy Analysis. 9(2). 171-178.
Meisels, Samuel J. (1989). M eeting the mandate o f public law 99-457: Early childhood
intervention in the nineties. American Journal of O rthopsychiatry. 59(3), 451-460.
M eyers, J. C. & M arcenko, M. O. (1989). Im pact of a cash subsidy program for families
of children with severe developm ental disabilities. Mental Retardation. 27(6). 383-387.
Michigan Department of Commerce. (1990). Michigan rural development strategy: Data
book. Lansing, MI: Author.
Michigan Department of Mental Health. (1984). Public Mental Health Manual: Community
Mental Health Family Support Services for the Developmentally Disabled (Vol. 4, Sec.
2, Chapter. G, pp. 1-6). Lansing, MI: Author.
Michigan Departm ent of Mental Health. (1986). Michigan mental health code [family
support subsidy pp. 9-13]. Lansing: Author.
Michigan Department of Mental Health. (1988). Public Mental Health Manual: Family
support services for persons with developm ental disabilities (Vol. 4, Sec. 2, Chapter.
G, pp. 1-8). Lansing, MI: Author.
Michigan Developmental Disabilities Council. (1988). Family support action plan.
Lansing, MI: Author.
M och, M. K. & M orse, E. V. (1977). Size, centralization and organization adoption of
innovations. American Sociological Review . 42(5). 716-725.
M oroney, R. M. (1979). Allocation of resources for family care. In R. H. Bruininks & G.
C. Kranlz (Eds.), Family care of developmentally disabled members: Conference
proceedings (pp. 63-76). M inneapolis: University of Minnesota.

146
M oroney, Robert M. (1986). Shared responsibility: Families and social policy. New York:
Aldine de Gruyter.
M owbray, C. T. & Herman, S. E. (1991). Using m ultiple sites in mental health
evaluations: Focus on program theory and implementation issues. New Directions for
Program E valuation. 50, 45-57.
Nord, W. R. & Tucker, S. (1987). Implementing routine and radical innovations.
Lexington, MA: Lexington Books.
N orusis, M ariza J. (1988). SPSS-X Advanced statistics guide [2nd ed.J. Chicago: SPSS
Inc.
Parrot, M. E. & Herman, S. E. (1987). Michigan department of mental health report on
family support subsidy program FY 85-86. Lansing: Michigan Department of Mental
Health.
Parrot, M. E. & Herman, S. E. (1988). Fam ily support subsidy program report FY198687. Lansing: Michigan Department of Mental Health.
Perlm an, R. & G iele, J. Z. (1983). An unstable triad: Dependents' dem ands, family
resources, com m unity supports. In R. Perlman (Ed.), Family home care: Critical issues
for services and policies (pp. 12-44). New York: Haworth Press.
Quine, L. & Pahl, J. (1985). Exam ining the causes o f stress in families with severely
mentally handicapped children. British Journal of Social W ork. 15(5). 501-517.
Rappaport, Julian. (1987). Term s o f empowerment/FXemplars of prevention: Tow ard a
theory of com m unity psychology. American Journal of Com m unity Psychology.
15(2), 121-144.
Rice, R. E. Rogers, E. M. (1980). Reinvention in the innovation process. Knowledge:
Creation. D iffusion. U tilization. 1(4), 499-514.
Rimmerman, A. (1989). Provision of respite care for children with developmental
disabilities: Changes in maternal coping and stress over time. Mental Retardation.
27(2), 99-103.
Roberts, R. N „ W asik, B. FI., Casio, G. & Ram ey, C. T. (1991). Fam ily support in the
home: Program s, policy, and social change. Am erican Psychologist. 46(2), 131-137.

147
Rogers, Everett M. (1983). Diffusion o f innovations [Third edition). New York: Free
Press.
Roitman, David B. (1984). Innovation in public sector organizations: A test of the modified
RD&D approach. Unpublished doctoral dissertation, M ichigan State University, East
Lansing.
Rowitz, Louis. (1985). Social support: The issue for the 1980s. M ental Retardation. 23(4),
165-167.
Rubin, S. & Quinn-Curran, N. (1983). Lost, then found: Parents' journey through the
com m unity service maze. In M. Seligman (Ed.), The family with a handicapped child:
Understanding and treatm ent (pp. 63-94). New York: Grune & Stratton.
Sabatier, P. A. (1987a). Knowledge, policy-oriented learning, and policy change: An
advocacy coalition framework. Knowledge: Creation, diffusion, utilization. 8(4), 649692.
Sabatier, P. A. (1987b). Top-down and bottom-up approaches to implem entation research:
A critical analysis and suggested synthesis. Journal of Public Policy, 6, 21-48.
Sabatier, P. A. & M azmanian, D. A. (1981). The implem entation o f public policy: A
fram ew ork of analysis. In D. A. M azmanian and P. A. Sabatier (Eds.), Effective policy
im plem entation (pp. 3-35). Lexington, MA: Lexington Books.
Salisbury, C. L. & Intagliata, J. (1986). Respite care: Support for persons with
developmental disabilities and their families. Baltimore: Brookes.
Scheirer, Mary. A. (1981). Program implementation: The organizational context. Beverly
Hills, CA: Sage.
Schroeder, Pat. (1989). Toward a national family policy. Am erican Psychologist. 44(11),
1410-1413.
Seligm an, Milton. (Ed.) (1983). The family with a handicapped child: Understanding and
treatm ent. New York: Grune & Stratton.
Seltzer, M. M. & Krauss, M. W. (1984). Placem ent alternatives for m enially retarded
children and their families. In J. Blacher (Ed.), Severely handicapped young children
and their families (pp. 143-175). New York: Academic Press.

148
Sherman, Barry R. (1988). Predictors of the decision to place developm entally disabled
family members in residential care. American Journal of Mental Retardation. 92(41.
344-351.
Sherm an, B. R. & Cocozza, J. J. (1984). Stress in fam ilies o f the developm entally
disabled: A literature review o f factors affecting the decision to seek out-of-home
placements. Fam ily Relations. 33. 95-103.
Sigelm an, L., Roeder, P. W. & Sigelman, C. K. (1981). Social service innovation in the
American slates: Deinstitutionalization of the mentally retarded. Social Sciences
Q uarterly. 62(3), 503-515.
Singer, G. H. S. & Irvin, L. K. (Eds.) (1989). Support for care giving families: Enabling
positive adaptation to disability. Baltimore: Brookes.
Slater, M. A. & W ikler, L. (1986). 'Norm alized' fam ily resources for fam ilies with a
developm entally disabled child. Social W ork. 31(5). 385-390.
Soyka, P. (1976). Hom em aker home health aide service for handicapped children. Child
W elfare. 55, 241-251.
Starkey, J. & Sarli, P. (1989). Respite and family support services: Responding to the
need. Child and Adolescent Social W ork. 6(4), 313-326.
Stewart, David W. (1984). Secondary research: Information sources and m ethods. Beverly
Hills: Sage.
Sum m ers, J. A., Behr, S. K. & Turnbull, A. P. (1989). Positive adaptation and coping
strengths o f fam ilies who have children with disabilities. In G. H. S. Singer and L. K.
Irvin (Eds.), Support for care giving families: Enabling positive adaptation to disability
(pp. 27-40). Baltimore: Brookes.
Tausig, Mark. (1988). Personal support networks: Benefits and liabilities. Mental
R etardation. 26(1), 47-49.
Taylor, S. J., Knoll, J. A., Lehr, S. & W alker, P. M. (1989). Fam ilies for all children:
Value-based services for children with disabilities and their families. In G. H. S.
Singer and L. K. Irvin (Eds.), Support for care giving families: Enabling positive
adaptation to disability (pp. 41-53). Baltimore: Brookes.

149
Turnbull, A. P., Brothcrson, M. J. & Sum mers, J. A. (1985). The im pact of
deinstitutionalization on families: A family system s approach. In R. H. Bruininks & K.
C. Lakin (Eds.), Living and learning in the least restrictive environm ent (pp. 115-152).
Baltimore: Brookes.
Turnbull, A. P., Sum m ers, J. A. & Brothcrson, M. J. (1986). Fam ily life cycle:
Theoretical and empirical implications and future directions for families with mentally
retarded members. In J. J. G allagher & P. M. Viet/.e (Eds.), Fam ilies of handicapped
persons: Research, program s, and policy issues (pp. 45-66). Baltimore: Brookes.
Upshur, Carole C. (1982). Respite care for m entally retarded and other disabled
populations: Program m odels and family needs. Mental Retardation. 20(1). 2-6.
Van de Ven, A. H. & Rogers, E. M. (1988). Innovations and organizations: Critical
perspectives. Com m unication R esearch. 15(5). 632-651.
Van M eter, D. S. & Van Horn, C. E. (1975). The policy implem entation process: A
conceptual framework. Adm inistration and Society. 6(4), 445-488.
W eiss, Heather B. (1988). Family support and education programs: W orking through
ecological theories of hum an developm ent. In H. W eiss and F. Jacobs (Eds.),
Evaluating fam ily program (pp. 3-36). New York: Aldine de Gruyter.
W eiss, Heather B. (1989). State family support and education programs: Lessons from the
pioneers. Am erican Journal of Orthopsychiatry. 59(1), 32-48.
W eissbourd, B. & Kagan, S. L. (1989). Fam ily support programs: Catalysts for change.
American Journal of O rthopsychiatry. 5911), 20-31.
W iegerink, R. & Com fort, M. (1987). Parent involvement: Support for fam ilies of children
with special needs. In S. Kagan, D. Powell, B. W eissbourd and E. Zigler (Eds.),
Am erica's family support programs: Perspectives and prospects (pp. 182-206) New
Haven: Yale University Press.
W ikler, Lynn M. (1986). Fam ily stress theory and research on families of children with
mental retardation. In J. J. G allagher & P. M. Vietze (Eds.), Families of handicapped
persons: Research, program s, and policy issues (pp. 167-195). Baltimore: Brookes.

150
W inton, Pamela. (1986). The developmentally delayed child within the family context.
Advances in Special Education. 5, 219-255.
Yin, Robert K. (1981, Jan.). Life histories of innovations: How new practices become
routinizcd. Public Administration Review. 21-28.
Yin, Robert K. (1982). Studying the implementation of public programs. In W. W illiam s
et al„ Studying implementation: Methodological and administrative issues (pp. 36-72).
Chatham , NJ: Chatham House.
Zaltm an, G., Duncan, R. & Holbek, J. (1973). Innovations and O rganizations. New York:
Wiley.
Zigler, E. & Black, K. B. (1989). Am erica's family support m ovem ent: Strengths and
limitations. American Journal o f Orthopsychiatry. 59(11. 6-19.