reliability
(Cronbach alpha) of every measured construct
in the questionnaire exceeded 0.7, which we considered
the minimum level required for reliable analyses.
The response rate was a favorable 51.3%, with
112 useful questionnaires received out of 218 CEOs.
We performed a test for non-response bias to minimize
the chance that the results we report are affected
by some unknown factor that systematically differs
between respondents and non-respondents [43]. We
compared average scores on two variables, hospital
size and region, for early respondents, late respondents
and non-respondents using both Chi-square tests and
independent-samples t-tests. Hospital size was measured
by the number of beds [21,41]. Hospital region
was a dummy variable that distinguished hospitals in
regions with a relatively long history of health care
management autonomy from hospitals in regions with
a more recent autonomy in health care management.1
The results inTablesA1andA2in AppendixAshowno
sign of non-response bias.Table 1 shows the descriptive
statistics and Table 2 shows the demographics characteristics
of our sample of CEOs. These data show that
the majority of CEOs is male (91.1%). CEOs’ ages
lie between 30 and 59 years, with 77.7% between 40
and 50 years and an average age of 44.2 years. Most
CEOs spent between 3 and 10 years in managerial
positions. Regarding CEO clinical and administrative
backgrounds, allCEOshave a university master degree.
The majority of CEOs had a dominant administrative
background (56.2%), rather than a clinical background
(43.8%).2