Psychometric properties of the Multifactor Leadership Questionnaire among nurses

Title. Psychometric properties of the Multifactor Leadership Questionnaire among nurses Aim. This paper reports a study investigating the psychometric properties of the Multifactor Leadership Questionnaire among nurses. Background. Although the Multifactor Leadership Questionnaire is one of the most widely used instruments to measure the multifactor leadership theory in organizational sciences, inconsistent research findings have been reported for its psychometric properties. Little is known about these properties. Method. Data were gathered by postal questionnaires in 2001–2002 with nurses working in different healthcare organizations in Finland. A follow-up study was performed 1-year later. The sample consisted of 601 nurses and nurse leaders, and the follow-up study had 78 respondents. The internal consistency of the Multifactor Leadership Questionnaire was explored using Cronbach's α coefficient and item-analysis, and the test–retest reliability using Pearson product moment correlation coefficient and intra-class correlation coefficient (single measure intra-class correlation coefficient). The factor structure was studied using exploratory and confirmatory factor analyses. Results. The internal consistencies of the leadership subscales were satisfactory. The Multifactor Leadership Questionnaire was fairly stable measured at the 1-year interval. Although the data fail to support the full nine-factor model, a reduced set of items from the Multifactor Leadership Questionnaire appears to show evidence of the three- and six-factor structures. Exploratory factor analysis showed evidence for the three-factor structure consisting of three distinctive, yet partly related leadership subscales. Confirmatory factor analysis showed evidence for the six-factor structure. Conclusion. The psychometric data suggest that a modified version of the Multifactor Leadership Questionnaire is a highly suitable instrument to measure multidimensional nursing leadership. Validity and reliability were supported. The results can be used in studies applying the Multifactor Leadership Questionnaire.