Perceptions and Opinions of Canadian Hospital Executives: METHODS
Study Procedures
A national sample of 500 individuals, from the total of 2581 CCHSE members (2003 membership year), were selected to receive the survey. The computerized random selection was performed by CCHSE staff. Mail and e-mail addresses were provided by CCHSE, and the survey was administered in both English and French. The survey sample was stratified to include 250 middle management (supervisors, managers, and directors) and 250 senior management members (executives). Although Quebec represents 23% of the total population of Canada, persons from Quebec make up only about 5% of the CCHSE membership. To ensure a representative sample, all Quebec members of the CCHSE were included. In August 2003, the survey instrument was sent by regular mail and by e-mail to the identified recipients, along with a cover letter explaining the purpose of the study. A follow-up letter was mailed to nonrespondents after 3 weeks. A second follow-up letter and another copy of the survey instrument were mailed to those who had still not responded after an additional 3 weeks. In-kind support, including letterhead and signatures from CCHSE representatives, was used to help increase response rate. Survey design and methodology were derived from commonly used sources.
Data Analysis
The responses from all completed surveys were analyzed using SAS version 8.2 software (SAS Institute, Cary, North Carolina). A one-way analysis of variance (ANOVA) was performed to compare average scores for the questions related to medication management competencies in relation to each demographic variable. For demographic variables with only 2 options, data were compared by t tests. Significance for all tests was set at p = 0.05. Post hoc tests were performed for significant ANOVA results using Tukey’s HSD test; these post hoc comparisons were considered significant at p = 0.05. In addition, the respondents’ responses to 3 open-ended questions were analyzed by generating common themes. Similar responses were grouped together within a question, and these clusters were labelled by a member of the research team (C.J.M.).
female viagra online
