This slightly revised model demonstrated excellent
fit: NFI = .98, NNFI = 1.18, CFI = .94. PFI (.55) is somewhat
low, but is deflated by the modest sample size.
The
results with the feasibility measure (Figure 1) support four of
the five predicted paths. Quality of working relationship has
a strong negative path to organizational resistance, and the path
from feasibility to quality of working relationship is positive.
(One of the paths added to improve fit was the one from feasibility
to organizational resistance, which indicates that quality of
working relationships is not a complete mediator of the feasibility/organizational
resistance linkage.) Two of the context variables (information
processing capacity and resource availability) show direct paths
to feasibility. Paralleling the results for control, three other
paths were added: from experience to working relationship and
to organizational resistance, and from information processing
capacity to organizational resistance.
DISCUSSION
A dramatic
interorganizational response (Community Mental Health Boards suing
the Ohio Department of Mental Health) was conceptualized within
an organizational resistance framework, and predicted using a
model drawing on the strategic issue diagnosis and interorganizational
relationship literatures. The results were consistent with the
key linkages in our model: contextual factors affect the way Board
decision makers frame the issue, which affects the quality of
working relationship with ODMH, which, in turn, affects the extent
of Board resistance to institutional pressure.
Our
findings support the argument that the quality of working relationship
plays a key role in interorganizational responses to strategic
issues. Quality of relationship completely mediated the effects
of control frame on resistance and partially mediated this effect
for feasibility frame. The direct effect of feasibility on resistance
suggests that if Boards felt they had feasible options for implementing
the Act, they were somewhat able to resist pressure from ODMH,
independent of effects due to the state of their working relationship.
We
found support for the usefulness of contextual factors at the
individual, group and organizational level for understanding resistance,
which supports Thomas et al.'s (1994) model and extends it to
interorganizational responses. Four of six predictions linking
context to frames were supported, indicating that contextual factors
work through interpretations. Work experience affected control,
but not feasibility, and information processing capacity showed
the opposite pattern. Resource adequacy was linked to both control
and feasibility, perhaps because of the centrality of resource
adequacy in the lawsuit. Modifications to improve the fit of the
models also support the importance of context. In both models
experience had direct effects on working relationship and resistance,
and information processing capacity had a direct effect on resistance.
This suggests that tests of the sensemaking model need to examine
both the indirect (through interpretations) and direct effects
of contextual factors on organizational response and outcomes.
This
study has several strengths. Very few studies on strategic issue
framing have examined a consequential, explicit organizational-level
response, and even fewer have dealt with interorganizational responses.
This study was strengthened by the inclusion of both subjective
and objective measures. Two of the six measures were objective
(i.e., work experience, organizational resistance), reducing the
plausibility of shared method variance as an alternative explanation.
Through happenstance, the data were gathered prospectively. Even
though we did not anticipate the lawsuit, the timing of our survey
was ideal: in advance of the litigation but not too far in advance.
We were fortunate to be in a research setting where 53 organizations
were considering litigation against a single institutional target.
Finally, the results supported key portions of our a priori model,
and the final model fit well despite a useable sample of 48 organizations.
The
modest sample size was one of the limitations of the study, in
that it reduced statistical power and forced us to include a limited
number of variables in our model. Because the lawsuit was not
anticipated, we were limited to using data gathered in a survey
that was not designed to predict this specific response. Had we
anticipated the litigation, we would have measured other variables
relevant to organizational resistance and to the quality of working
relationships.
Research
Home - Page 1 | 2
| 3 | 4 | 5
| < PREV NEXT
>
