4.4 Data Collection ProcedureOnce the questionnaire was finalized and its validity tested, data collection was initiated.The questionnaires were distributed in the selected hospitals. A cover letter was enclosedwith the questionnaire informing the respondents of the purpose and the authenticity ofthe research. The approval of Medical Research Ethical Committee, MREC from theMOH was ensured with the detailed confidentiality clause before conducting the study. Asample of the questionnaire was given to the Ministry of Health for their review so thatthey understand exactly what the research was about in order to secure their approval andget access and cooperation from the director of each hospital to help distribute thequestionnaires.4.5 Operationalization of Measurements /InstrumentsThe measurement of each study variable is discussed in this section. A total of five mainvariables were involved including the demographic variables.The responses were made on a five-point Likert scale as it is the most widely used scalein recent researches (Gwinner, 2006). Moreover, it is also able to measure accurately(Hair, Black, Babin, & Anderson, 2010) and to test the proposed hypotheses (DeVellis,1991). The respondents were able to choose a neutral rating in case some of them feltneutrally about some topics. Neutrality, according to Gwinner (2006), can be described asa legitimate opinion existing among respondents.
4.5.1 Patient SafetyIn the present study, patient safety was measured on two dimensions: frequency ofoccurrence of adverse events and overall of perception of patient safety.184.108.40.206 Frequency of Occurrence of Adverse EventsSix items adapted from Teng et al. (2009) were used to measure the frequency ofoccurrence of medication errors, patient falls, infections, injuries because of care, delayedpatient care, and incomplete or incorrect documentation. The items were measured on afive-point Likert scale, ranging from '1' "not at all" to '5' "frequently, if not always".Participants were asked to indicate how frequently these events occurred. The internalconsistency of the scale was reported to exceed 0.76 (Teng et al., 2009). The items areshown in Table 4.4.Table 4.4Measurement Items of Frequency of Occurrence of Adverse EventsNo. Items 1.Injuries because of care happen in the past year?2.Patient falls occur in the past year?3.Nosocomial infections occur in the past year?4.Medication errors occur in the past year?5.Delayed patient care happen in the past year?6.Incomplete or incorrect documentation occur in the past year?Source: Teng et al. (2009)
220.127.116.11 Overall Perception of Patient SafetyEight items adapted from Ramanujam, Abrahamson, and Anderson (2007) dealing withnurse managers’ perceptions of patient safety designed on a five-point Likert scale,ranging from '1' "strongly disagree" to '5' "strongly agree" was used. Reliability analysiswas reported to yield a Cronbach’s alpha of .874. Participants were asked to indicate theirlevel of agreement or disagreement to the items. Two items were negatively worded.