Full text of DOI:10.1375/bech.18.2.114

Organisational Correlates to Staff Attitudes
about Behavioural Programs
Patrick W. Corrigan, Sarah Lickey Diwan, Stanley G. McCracken, Kelvin Oliver,
Annette Backs, Fadwa Rashid, John Campion, and Ruth Davies Farmer
University of Chicago Center for Psychiatric Rehabilitation
In an attempt to facilitate the dissemination and regular implementation of behaviour treat-
ments in real-world settings, research has examined staff attitudes that may pose barriers to
these treatments. The purpose of the study reported in this paper is to examine the construct
validity of perceived organisational barriers to behavioural programs. One hundred and eigh-
teen staff working in community-based treatment programs for adults and children with
severe mental illness completed the Barriers to the Implementation of Behavior Therapy Test.
An analysis from a previous study (Corrigan, Kwartarini, & Pramana, 1992) identified two
reliable and valid factors that suggested perceived organisational barriers: institutional con-
straints and insufficient collegial support. Results of a confirmatory factor analysis on data
from this study supported the earlier factor structure. Research participants also completed
team functioning measures of individual staff burnout, collegial network satisfaction, organi-
sational culture, and team leadership. Attitudes about institutional constraints and insufficient
collegial support were found to be associated with burnout, collegial network satisfaction, and
organisational culture. Strategies for disseminating behavioural programs must target organi-
sational barriers too.
s behaviour therapies have been shown to of teachers and mentors expert in these areas,
ameliorate a variety of problems that arise and improved staff attitudes about behaviour
A
from psychiatric disorders, researchers have dis- therapies (Axelrod, 1992; Backer, Liberman, &
cussed strategies for increasing the use of these Kuehnel, 1986; Persons, 1997; Wilson, 1997).
therapies in real-world settings. This discussion Because they may undermine use of behavioural
has led to the identification of validated treat- therapies in real-world settings, staff attitudes
ments and corresponding treatment manuals for particularly targeted for change include low self-
specific disorders (Kazdin, 1996, 1998; Task expectations about the potency of interventions
Force on the Promotion and Dissemination of and low sense of self-efficacy in carrying out the
Psychological Procedures, 1995). Development program (Bandura, 1989; Kirsch, 1985).
of these manuals has, in turn, led to dissemina-
Most treatment dissemination strategies
tion strategies that facilitate the broadest use of focus on improving the individual practitioner;
these treatments in a manner faithful to the hence, they are largely educational in nature,
empirically tested protocol. These strategies seeking to improve information and attitudes
include more information for professionals and about behavioural interventions for specific
consumers about treatments, greater availability populations. Many treatments, however, are
This study was made possible, in part, by grants from the Illinois Office of Mental Health and the Rehabilitation
Services Administration of the US Department of Education.
Address for correspondence: Patrick W. Corrigan, University of Chicago Center for Psychiatric Rehabilitation, 7230
Arbor Drive, Tinley Park, IL, 60477, USA. Email: p-corrigan@uchicago.edu
BEHAVIOUR CHANGE
114
Vol. 18, No. 2 2001 pp. 114–123

ATTITUDES ABOUT BEHAVIOURAL PROGRAMS
provided by teams and programs rather than validity of the factors representing organisational
individual practitioners. This is especially true barriers to implementing behaviour therapies.
for seriously disordered groups such as
Our research attempts to replicate earlier
persons with severe mental illness, develop- findings by using staff burnout and collegial
mental disability, and/or substance abuse support as indicators of team (organisational)
1
(
Corrigan & Giffort, 1998). Clearly, group- functioning. Two additional markers are also
level or organisational factors affect the team included in this study: team leadership and
member’s perceptions of specific interventions organisational culture. Teams with effective
and the team’s ability to implement programs leaders are more cohesive (Sluyter, 1995) and
based on these interventions (Sedikides, provide better clinical services (Garman &
Schopler, & Insko, 1998). Hence, organisa- Corrigan, 1998). Hence, we examined the rela-
tional variables must be added to any paradigm tionship between perceptions of team leaders
targeting treatment dissemination.
and barriers to behavioural interventions.
Organisational psychologists have researched
several models for understanding leadership. A
particularly useful theory to arise out of this
research is the multifactor model of leadership
The need for combined organisational and
educational factors in disseminating
behavioural treatments has been supported in
studies of staff perceptions about the implemen-
tation of behaviour therapy. For example, find-
ings from one study identified barriers related to
educational and organisational factors that
impede the use of behavioural interventions
(
Bass, 1990; Bass & Yammarino, 1991). The
model has been validated on teams working in
mental health (Corrigan, Garman, Lam, &
Leary, 1998) and rehabilitation settings
(Corrigan, Garman, Canar, & Lam, 1999). The
(Emerson & Emerson, 1987). Factor analysis of
multifactor model distinguished two factors rel-
evant to the effective leader:
a survey of 332 staff working in a California
psychiatric hospital and developmental centre
further specified those organisational constructs 1. Transformational leadership. Effective lead-
(
Corrigan et al., 1992). The two organisational-
ers help team members transform programs
to meet the ever-evolving needs of their
clientele.
level constructs that emerged were institutional
constraints (administrative practices and bud-
getary limitations result in inadequate resources 2. Transactional leadership. Effective leaders
to implement behavioural treatments) and insuf-
ficient collegial support (staff perceive that
peers are not interested in or will not support
behavioural programs). In an effort to test the
validity of these constructs across mental health
service settings, one goal of the investigation
reported in this paper is to confirm the factor
attend to the day-to-day tasks that need to
be completed to operate the program
smoothly.
Transformational goals are achieved through
charisma, inspiration, intellectual stimulation,
and individual consideration of staff interests.
Transactional goals are pursued using contin-
structure from the earlier study on staff working gent reward and management-by-exception
in community settings.
(responding to organisational issues only when
Subsequent research examined correlates to they arise as a problem or an “exception” to
these barriers. Results of a path analysis showed standard practices). These effective styles are
that manifest variables representing organisa- contrasted with what has been shown to be a
tional and educational factors accounted for fundamentally ineffective approach: laissez-
independent variance in overall barriers to faire leadership. The laissez-faire leader
behavioural programs (Corrigan et al., 1998). remains aloof, uninvolved, and uninterested in
Other research showed that such organisational the day-to-day activities of the treatment team.
variables as staff burnout and collegial support
Organisational culture represents a set of
are associated with these barriers (Corrigan et mutually acceptable and important beliefs held
al., 1996). A second goal of the study reported in by a cohesive program staff. Like the multifac-
this paper is to further examine the construct tor model of leadership, these beliefs may be
115

CORRIGAN ET AL.
transformational (reflecting settings where there al., 1992). These factors, and items that com-
is a sense of purpose and high work goals) and prise them, are summarised in Table 1. In subse-
transactional (organisations that focus on rela- quent research, we distinguished the first two
tionships in terms of explicit and implied con- factors (labelled as organisational or team fac-
tractual relationships) (Bass & Avolio, 1993a). tors) from the latter three factors (called individ-
We expect to find that team members with ual or education-based factors) (Corrigan et al.,
transformational team leaders and cultures will 1996). Other studies supported the reliability
endorse fewer organisational barriers to and validity of the first two, organisational fac-
behaviour therapy.
tors, but not the latter three, which examined
education-based constructs (Corrigan et al.,
1
992; Corrigan, McCracken, Kommana,
Method
Edwards, & Simpatico, 1996). Hence, the focus
of the confirmatory factor analysis and subse-
quent construct validity reported in this paper is
on the first two factors of the BIBT.
One hundred and eighteen staff members were
recruited and informed for this study. All con-
sented and completed the battery of measures.
Research participants were recruited from 18
treatment programs provided by four mental
health agencies in Illinois. Programs included
residential, vocational, and skill-oriented outpa-
tient services for children and adults with severe
mental illness. No differences in data were
found across programs, so staff responses were
collapsed onto one group for the remaining anal-
yses. All current staff, regardless of tenure, were
included in the study; previous research has
failed to show significant correlations between
length of time on the job and barriers to
behaviour therapy (Corrigan et al., 1992).
The factor structure obtained from our
sample was compared to the one obtained from
the previous study (Corrigan et al., 1992) using
SAS confirmatory factor analysis (CFA) gener-
ated by structural equation modelling.
Specifically, the factor loadings in the linear
equations were determined from the earlier
factor structure (Corrigan et al., 1992); the eight
items of the BIBT were loaded into equations
that corresponded with the two organisational
factors that were found in the first study. We
examined the fit between our previous model
and the data from this study using several
indices of goodness of fit as well as significance
tests for individual CFA factor loadings. The
model for these data was subsequently modified
to improve goodness of fit based on examina-
tion of the LaGrange multipliers.
Measurement of Perceived Barriers
to Behaviour Therapy
Staff were administered the Barriers to
Implementation of Behavior Therapy (BIBT)
test. The BIBT comprises 18 statements about
barriers to introducing behavioural innovations
into the subject’s program (Corrigan et al.,
Organisational Correlates to Perceived Barriers
Research participants completed four other
measures representing team functioning.
1
992). BIBT statements (e.g., “There are not
enough staff to adequately carry out behavioural
interventions”) are rated on a 6-point Likert Maslach Burnout Inventory. Participants
scale (l = strongly agree). Results of an reported their level of burnout using the Maslach
exploratory factor analysis yielded five factors: Burnout Inventory (MBI), a 22-item self-report
institutional constraints, insufficient collegial measure that assesses burnout in helping profes-
support, philosophical opposition (staff do not sions (Maslach & Jackson, 1986). Research par-
believe that behaviour therapy is effective or ticipants rated the frequency with which they
appropriate for their clients), client dissatisfac- experienced job-related stressors on a 7-point
tion (clients and their families do not approve of Likert scale (7 = highly agree). Prior analysis of
behavioural programs), and collateral interfer- the MBI has uncovered three reliable and valid
ence (clients, other patients, and family mem- factors that are included in this study: emotional
bers will not comply with, or otherwise exhaustion (e.g., “I feel emotionally drained
undermine, behavioural strategies) (Corrigan et from work”), depersonalisation (e.g., “I feel I
116

CORRIGAN ET AL.
ATTITUDES ABOUT BEHAVIOURAL PROGRAMS
TABLE 1
faction with his or her support network. When
completing this pencil-and-paper measure,
research participants provided the initials of
individuals who have supported them in generic
life problems that are described in seven test
questions (e.g., “Who can you really count on to
be dependable when you need help?”). Subjects
then reported satisfaction with their network vis-
a-vis each life problem on a 6-point Likert scale
Items and Factors from the Barriers to Implementation
of Behavior Therapy Test
Factor 1: Institutional Constraints
Item
9
4
1
Residence overcrowded
Insufficient staff to carry out programs
2
1
Insufficient tangible rewards available to rein-
force clients
(
6 = very satisfied). The seven stimulus ques-
tions were rewritten slightly to reflect staff
members’ perceptions of collegial support at the
work setting (e.g., “Who can you really count on
at work to be dependable when you need
help?”). The SSQ yields a measure of total satis-
faction with one’s support network equal to the
ratings across the seven stimulus questions. The
revised measure of collegial satisfaction has
been shown to be internally consistent (Corrigan
et al., 1998; Corrigan, Holmes, & Luchins,
1
Restrictive administrative practices
Factor 2: Insufficient Collegial Support
Item
5
1
6
Fellow staff uninterested
5
Lack of support from fellow staff
Lack of effective communication between
shifts
7
Excessive staff turnover
Factor 3: Philosophical Opposition
Item
1995) and to significantly predict perceived bar-
riers (Corrigan et al., 1996).
2
3
Believes behaviour therapy does not work
The Multifactor Leadership Questionnaire,
Version Y. Staff were instructed to rate 44 items
representing leadership skills according to “how
you view your team leader” (Bass & Avolio,
Believes behaviour modification inappropriate
for population
1
Views behaviour modification as unethical
Factor 4: Client Dissatisfaction
Item
1993b). Items were rated on a 5-point frequency
scale from 1 (frequently if not always) to 5 (not
at all); lower scores represent greater endorse-
ment of the scale’s construct. The MLQ has
been widely investigated and shown to have
excellent internal consistency, test-retest relia-
bility, and construct validity. Results of factor
analyses have shown that the test yields eight
reliable factors. Four factor scales represent
transformational leadership: charisma, inspira-
tion, intellectual stimulation, and individualised
consideration. Three scales comprise transac-
tional leadership: contingent reward, active man-
1
1
7
6
Clients do not favour behavioural programs
Clients do not understand behavioural pro-
grams
1
8
Families of clients do not favour behavioural
programs
Factor 5: Collateral Interference
Item
1
1
3
4
Relatives interfere with behavioural programs
Other residents interfere with behavioural
programs
agement-by-exception,
and
passive
management-by-exception. The nonleadership
treat some of the consumers as if they were factor is called laissez-faire leadership.
impersonal objects”), and personal accomplish-
Organisational Development Questionnaire.
ment (e.g., “I feel exhilarated after working
Research participants completed the
closely with my consumers”).
Organisational Development Questionnaire
Modified Social Support Questionnaire (M- (ODQ) to describe their program’s culture. The
SSQ). I. Sarason, Levine, Basham, & B. Sarason ODQ comprised 28 items (e.g., “People are hesi-
tant to say what they really think”) that were to
be answered either true, false, or can’t say (Bass
(
1983) developed the SSQ to measure an indi-
vidual’s perceptions about the size of and satis-
117

CORRIGAN ET AL.
&
Avolio. l993b). Staff members were instructed BIBT (Corrigan et al., 1982). Two models were
to answer these items about “the treatment team tested before goodness-of-fit indices met
on which you work”. These items are summed to acceptable levels. In Model 1, the equation for
generate two reliable and valid scores: the trans- Factor 1 comprised BIBT Items 9, 4, 12, and
formational culture and transactional culture 11. The equation for Factor 2 included Items 5,
score. Higher scores represent greater endorse- 15, 6, and 7. The convergence criterion was met
ment of a cohesive and positive culture.
after 38 iterations, suggesting that the CFA was
successful for the 118 participants in this study.
All the CFA loadings were satisfactory: the
size of the standardised coefficients was small
to large, the standard errors did not approach
Results
Confirmatory Factor Analysis
The indices representing the goodness of fit and zero, and the t values were significant (Hatcher,
CFA factor loadings are summarised in Table 2. 1994). However, review of the goodness-of-fit
The confirmatory factor analysis only consid- indices showed that none met criterion levels.
ered the first two factors originally found on the The chi-squared statistic was highly significant,
TABLE 2
CFA Indices Representing Goodness of Fit Generated for Two Models for the Two Organisational Factors
Model 1
Model 2
Model 1
Model 2
Fac 1: Items 9, 4, 12, 11
Fac 1: Items 9, 4,
CFA factor loadings continuded
Item/Factor = 7/2
SC = .61
1
2, 11
Fac 2: Items 5, 15, 6, 7
Fac 2: Items
5
, 15, 6
SE = .15
Goodness of fit
Chi-square
t = 6.32***
Item/Factor = 9/1
SC = .32
= 9/1
=
.001
48.55 df = 19 p < .001
= 22.79 df = 13 p
<
= .33
Chi-square/df = 2.56
Bentler’s CFI = .87
= 1.75
= .94
= .91
SE = .14
= .14
t = 2.98*
= 3.08**
= 11/1
= .68
Bentler & Bonett’s NNFI = .87
CFA factor loadings
Item/Factor = 4/1
SC = .56
Item/Factor = 11/1
SC = .70
= 4/1
SE = .14
= .15
= .56
t = 7.14***
= 6.83***
= 12/1
= .66
SE = .16
= .17
Item/Factor = 12/1
SC = .64
=
5.61***
= 5.48***
= 5/2
Item/Factor = 5/2
SC = .54
SE = .16
= .16
= .57
t = 6.51***
= 6.53***
= 15/2
= .86
SE = .16
= .15
Item/Factor = 15/2
SC = .81
t = 5.54***
Item/Factor = 6/2
SC = .68
= 5.85***
= 6/2
SE = .13
= .14
= .66
t = 9,18***
= 9.46***
SE = .15
= .16
t = 7.31***
= 6.99***
Note. CFI = comparative fit index; NNFI = non-normed fit index; SC = standardised coefficient; SE = standard error.
p < .05. **p < .01. ***p < .001.
*
118

ATTITUDES ABOUT BEHAVIOURAL PROGRAMS
the chi-square-to-df ratio was greater than 2, straints met the Bonferroni criterion for signifi-
and the CFI and NNFI were less than .90. cance. Staff who reported greater satisfaction
Examination of the LaGrange multipliers (a with peers reported fewer institutional con-
conservative strategy for modifying CFA straints. Staff who rated their organisation’s cul-
models) suggested dropping BIBT Item 7 to ture as high on transformational qualities were
improve the goodness of fit. Hence, Model 2 less likely to endorse institutional constraints.
included all items from Model 1 except Item 7.
No leadership variables were significantly asso-
Once again, CFA factor loadings were ade- ciated with institutional constraints.
quate, with standardised coefficients, standard
Table 4 summarises the results of a multiple
errors, and t values meeting criterion levels. regression analysis to determine how institu-
Review of goodness-of-fit indicators mostly tional constraints were independently associated
supported the results. Once again, the chi- with these measures of team functioning.
Independent variables in this analysis included
the four team functioning variables that were
associated with institutional constraints in Table
square statistic was highly significant, which
does not support goodness of fit. However, this
statistic may lead to Type II error, especially in
small data sets (Hatcher, 1994). Significance of
the chi-square statistic has been alternatively
interpreted in terms of the chi-square-to-df
statistic. Goodness of fit is supported when this
ratio is less than 2; in this study, the chi-square-
to-df was 1.75. Moreover, the CFI and NNFI
were above .90. Hence, these results support a
factor structure for the organisational factors
representing institutional constraints and insuf-
ficient collegial support.
3
. We also included the BIBT Factor 2 in this
analysis, given its high association (r = .55)
with Factor 1. Team functioning variables asso-
ciated with Factor 1, after Factor 2 is partialled
out, will suggest variance associated uniquely
with institutional constraints. Results show that
both Factor 2 and depersonalisation on the MBI
account for unique variance in institutional con-
straints. Hence, burnout seems to account for
independent variance in BIBT Factor 1. The
two variables together account for 33.6% of the
variance in institutional constraints.
Correlates of Perceived Barriers
Table 3 also lists the correlations between
insufficient collegial support (BIBT Factor 2)
and the team functioning measures. Like insti-
tutional constraints, insufficient collegial sup-
port was significantly associated with emotional
exhaustion and depersonalisation on the MBI.
Insufficient collegial support was also associ-
ated with network satisfaction; staff who were
more satisfied with their network of colleagues
reported more collegial support on the BIBT.
Insufficient collegial support was signifi-
cantly associated with both ODQ measures,
albeit in an inverse manner. Staff who described
their organisational culture as more transforma-
tional and less transactional were likely to
Pearson product–moment correlations were
determined among the BIBT factors and the
measures of team functioning. Results are sum-
marised in Table 3. Note that the two subscales
that represent organisational or team factors
(
institutional constraints and insufficient colle-
gial support) were highly correlated (r = 0.55, p
.001). Although not reported here, it is also
<
interesting to note that none of the three sub-
scales representing individual or education-
based factors (found from Corrigan et al., 1992)
were significantly associated (p < .10) with the
team functioning scales.
According to Table 3, four team functioning
measures were significantly associated with report greater collegial support for behavioural
BIBT Factor 1, institutional constraints. Staff interventions. Insufficient collegial support was
who reported higher emotional exhaustion and associated with five variables from the MLQ.
depersonalisation on the MBI were more likely Research participants who characterised their
to endorse institutional constraints as a barrier leaders with greater charisma, inspiration, and
to behavioural programs. The correlation coeffi- individualised consideration were likely to
cient representing the relationship between report more collegial support for behavioural
emotional exhaustion and institutional con- interventions. Conversely, staff who described
119

CORRIGAN ET AL.
TABLE 3
Pearson Product–Moment Correlations Among Factors from the Barriers to Implementation of Behavior Therapy
Test and Measures of Burnout, Collegial Network, Organisational Culture, and Team Leadership
Factor 1
Institutional constraints
Factor 2
Insufficient collegial support
Burnout (MBI):
Emotional exhaustion
Depersonalisation
–.31***
–.28**
–.03
–.39***
–.25*
.00
Personal accomplishment
Collegial network (SSQ):
Network size
.01
.11
Network satisfaction
Organisational atmosphere (ODQ):
Transformational
.26**
.27**
.20*
–.14
.47***
Transactional
–.32***
Team leadership (MLQ):
Charisma
.01
–.10
–.03
.01
–.18*
–.20*
–.01
–.23*
–.09
.01
Inspiration
Intellectual stimulation
Individualised consideration
Contingent reward
.07
Active management-by-exception
Passive management-by-exception
Laissez-faire
–.06
–.05
.03
.20*
.23*
Note. MBI = Maslach Burnout Inventory; SSQ = Social Support Questionnaire; ODQ = Organisational Development
Questionnaire; MLQ = Multifactor Leadership Questionnaire. Remember that lower scores on the BIBT and MLQ reflect
greater endorsement of corresponding factors.
*
p < .05. ** p < .01. *** p < .001.
TABLE 4
Results of Multiple Regression Examining the Relationship Between Predictors and Two BIBT Factors:
Institutional Constraints and Insufficient Collegial Support
Dependent variable: BIBT Factor 1: Institutional constraints
Independent variables
BIBT Factor 2
Cumulative R
F-to-enter
41.57
.54
.58
MBI: Depersonalisation
4.50
Dependent variable: BIBT Factor 2: Insufficient collegial support
Independent variables
ODQ: Transformation
BIBT Factor 1
Cumulative R
F-to-enter
31.99
17.13
9.11
.51
.62
.66
.68
MBI: Emotional exhaustion
MLQ: Laissez-faire
5.13
120

ATTITUDES ABOUT BEHAVIOURAL PROGRAMS
their leaders as laissez-faire and practicing pas- four domains — organisational culture, staff
sive management-by-exception were likely to burnout, and team leadership — accounted for
report insufficient collegial support for independent variance in this organisational bar-
behavioural programs.
rier. Like institutional constraints, results
The bottom of Table 4 summarises the showed that staff who were emotionally
results of the multiple regression analysis with exhausted, depersonalised, and dissatisfied with
insufficient collegial support as the dependent their support network reported insufficient sup-
variable. After partialling out variance due to port for behavioural programs. Hence, staff
BIBT Factor 1, the analysis showed insufficient burnout and dissatisfaction with one’s collegial
collegial support to be independently associated network seem to be broad indicators of organi-
with one variable from the organisational cul- sational barriers.
ture (ODQ transformation), burnout (MBI emo-
The two factors representing organisational
tional exhaustion), and leadership (MLQ culture had inverse relationships with insuffi-
laissez-faire) domains. The four independent cient collegial support. Staff who viewed their
variables together accounted for 46.2% of the culture in a transformational light (i.e., recog-
nising a sense of purpose and aspiring to a
higher quality of care) were likely to report that
their colleagues were supportive of behavioural
interventions. Conversely, staff who viewed
their interactions with peers as contractual
reported less support for behavioural interven-
tions. The first finding is consistent with Bass’s
variance in insufficient collegial support.
Discussion
The purpose of this study was to examine the
construct validity of perceived organisational
barriers to behavioural programs. Results of a
confirmatory factor analysis supported findings
from an earlier study that identified two organi-
sational barriers: institutional constraints and
insufficient collegial support. As in the earlier
factor analysis, institutional constraints
reflected lack of staff and monetary resources as
well as bureaucratic red tape that impeded the
implementation of behavioural programs.
Insufficient collegial support included lack of
interest and poor communication among peers.
Subsequent analyses sought to further
describe organisational barriers by examining
concurrent correlates to each. Institutional
(1990) view of transformational goals. Namely,
staff who seek to improve the plane of services
are more likely to be more innovative and to
implement effective strategies in their setting.
The second finding about transactional culture
was more of a surprise, namely, that viewing
colleagues vis-à-vis business-like exchanges
actually interferes with innovations. This find-
ing seems to undermine an organisational
behaviour-management approach towards
changing teams that seeks to use applied
behaviour analysis to improve staff functioning.
Note, however, that Bass (1990) has not found
constraints were found to be significantly asso- that transactional cultures have this kind of neg-
ciated with indices representing individual ative influence.
staff burnout, collegial network, and organisa-
Several leadership factors were found to be
tional culture. Staff who were emotionally associated with insufficient collegial support.
exhausted, depersonalised, and dissatisfied Staff who viewed their leader as charismatic,
with their network of colleagues reported inspiring, and individually considerate had
greater institutional constraints. Moreover, more positive views towards colleagues being
staff who viewed their organisational culture in supportive of behavioural innovations.
a transformational light were less likely to Conversely, staff who reported that their leader
endorse institutional constraints to implement- was passive or laissez-faire admitted to little
ing behavioural interventions.
support. This finding may suggest that leaders
Attitudes about insufficient collegial support are essential for cohesive teams that are willing
were associated with all four team functioning to implement innovative and effective programs.
variables. Moreover, results of a multiple However, these results do not suggest the spe-
regression analysis suggested that three of these cific direction of association: whether cohesive
121

CORRIGAN ET AL.
teams yield transformational leaders or whether
transformational leaders produce tight work
groups. This issue needs to be examined in
future research.
Endnote
1
We used the terms team and organisation inter-
changeably in the remainder of the paper. Although
we realise that organisation is a much broader con-
cept, our research thus far has focused on the organi-
sational dynamics with in the team.
Results from this study suggest that atti-
tudes about barriers to implementing
behavioural interventions are influenced by
several organisational factors. This research
does not, however, suggest whether these atti-
tudes directly influence behavioural practices,
that is, whether staff who perceive greater bar-
riers actually implement behavioural interven-
tions less often. Other research has shown a
relationship between perceived attitudes and
actual practices (Corrigan et al., 1995). This
association needs to be replicated in future
research. In particular, investigations need to
examine how specific barriers affect the team’s
ability to implement a behavioural program.
Findings from this study — that burnout,
organisational culture, and leadership account
for unique variance in the organisational bar-
rier of collegial support — suggest that several
factors are needed to change this organisa-
tional barrier. Future research needs to identify
strategies that will address these factors.
Perhaps the greatest message of this study,
however, is to remind the reader that dissemi-
nation of behavioural strategies is not solely an
issue of changing individual practices, espe-
cially in program-based approaches that are
provided by a team of practitioners. A variety
of organisational strategies must be added to
educational attempts to improve the dissemina-
tion and eventual implementation of state-of-
the-art behavioural therapies. Previous
research by our group has examined some of
these strategies (Corrigan & McCracken,
References
Axelrod, S. (1992). Disseminating an effective educa-
tional technology. Journal of Applied Behavior
Analysis, 25, 31–35.
Backer, T., Liberman, R., & Kuehnel, T. (1986).
Dissemination and adoption of innovative psychoso-
cial interventions. Journal of Consulting
and Clinical Psychology, 54, 111–118.
Bandura, A. (1989). Regulation of cognitive processes
through perceived self-efficacy. Developmental
Psychology, 25, 729–735.
Bass, B.M. (1990). From transactional to transforma-
tional leadership: Learning to share the vision.
Organizational Dynamics, 18, 19–31.
Bass, B.M., & Avolio, B.J. (1993a). Transformational
leadership and organizational culture. Public
Administration Quarterly, 17, 112–121.
Bass, B.M., & Avolio, B.J. (1993b). Transformational
leadership development: Manual for the Multifactor
Leadership Questionnaire. Palo Alto, CA:
Consulting Psychologists Press.
Bass, B.M., & Yammarino, F.J. (1991). Congruence
of self and others leadership ratings of naval officers
in understanding successful performance. Applied
Psychology: An International Review, 40, 437–454.
Corrigan, P.W., Garman, A.N., Canar, J., & Lam, C.
(1999). Characteristics of rehabilitation team lead-
ers: A validation study. Rehabilitation Counseling
Bulletin, 42, 186–195.
Corrigan, P.W., Garman, A.N., Lam, C., & Leary, M.
(1998). What mental health teams want in their lead-
ers. Administration and Policy in Mental Health.
Corrigan, P.W., & Giffort, D.F. (Eds.). (1998). New
directions for mental health services: No. 79.
Building teams and programs for effective rehabili-
tation. San Francisco: Jossey–Bass.
Corrigan, P.W., Holmes, E.P., & Luchins, D. (1995).
Burnout and collegial support in state psychiatric
hospital staff. Journal of Clinical Psychology,
5
1, 703–710.
1
997; Garman & Corrigan, 1998); they include
Corrigan, P.W., Kwartarini, W.Y., & Pramana, W. (1992).
Staff perception of barriers to behavior therapy
at a psychiatric hospital. Behavior Modification,
staff needs assessment regarding current pro-
gram shortfalls, participative decision making
to develop program components that meet
these shortfalls, Socratic questioning to exam-
ine specific decisions, and continuous quality
improvement to evaluate the impact of pro-
gram changes over time.
1
6, 132–144.
Corrigan, P.W., & McCracken, S.G. (1997). Interactive
staff training: Rehabilitation teams that work.
New York: Plenum.
Corrigan, P.W., McCracken, S.G., Kommana, S.,
Edwards, M., & Simpatico, T. (1996). Staff
perceptions about barriers to innovative behavioral
122

ATTITUDES ABOUT BEHAVIOURAL PROGRAMS
rehabilitation programs. Cognitive Therapy and
Research, 20, 541–551.
Kirsch, I. (1985). Self-efficacy and expectancy:
Old wine with new labels. Journal of Personality
and Social Psychology, 49, 824–830.
Maslach, C., & Jackson, S.E. (1986). Maslach Burnout
Inventory manual (2nd ed.). Palo Alto, CA:
Consulting Psychologists Press.
Persons, J.B. (1997). Dissemination of effective meth-
ods: Behavior therapy’s next challenge. Behavior
Therapy, 28, 465–471.
Sarason, I.G., Levine, H.M., Basham, R.B., & Sarason,
B.R. (1983). Social support as an individual differ-
ence variable: Its stability, origins, and relational
aspects. Journal of Personality and Social
Psychology, 44, 127–139.
Corrigan, P.W., Williams, O.B., McCracken, S.G.,
Kommana, S., Edwards, M., & Brunner, J. (1998).
Staff attitudes that impede the implementation
of behavioral treatment programs. Behavior
Modification, 22, 548–562.
Emerson, E., & Emerson, C. (1987). Barriers to the
effective implementation of habilitative behavioral
programs in an institutional setting. Mental
Retardation, 25, 101–106.
Garman, A.N., & Corrigan, P.W. (1998). Developing
effective team leaders. In P.W. Corrigan & D.F.
Giffort (Eds.), New directions for mental health ser-
vices: No. 79. Building teams and programs for
effective rehabilitation (pp. 45–54). San Francisco:
Jossey–Bass.
Sedikides, C., Schopler, J., & Insko, C.A. (Eds.). (1998).
Intergroup cognition and intergroup behavior.
Mahwah, NJ: Lawrence Erlbaum.
Hatcher, L. (1994). A step-by-step approach to using
the SAS system for factor analysis and structural
equation modeling. Cary, NC: SAS Institute.
Sluyter, G.V. (1995). Mental health leadership training:
A survey of state directors. Journal of Mental
Health Administration, 22, 201–204.
Kazdin, A.E. (1996). Validated treatments: Multiple per-
spectives and issues — introduction to the series.
Clinical Psychology: Science and Practice, 3,
Task Force on the Promotion and Dissemination
of Psychological Procedures. (1995). Training in and
dissemination of empirically-validated psychologi-
cal treatments: Report and recommendations.
The Clinical Psychologist, 48(1), 3–23.
2
16–217.
Kazdin, A.E. (1998). Treatment manuals in clinical prac-
tice: Introduction to the series. Clinical Psychology:
Science and Practice, 5, 361–362.
Wilson, G.T. (1997). Behavior therapy at century close.
Behavior Therapy, 28, 449–457.
ꢀ
123