Assessing the discriminant validity of transformational leader behaviour as measured by the MLQ

Bass's (1985) proposition that transformational leadership can be defined by distinct constructs (charisma, intellectual stimulation and individualized consideration) was tested. Using LISREL 8 confirmatory factor analysis, a three-factor first-order model (Model 1) was compared with a single-factor model (Model 2) and a hierarchical model (Model 3). The sample consisted of 1440 subordinates who worked in Australia for an international bank. The Multifactor Leadership Questionnaire (MLQ-5X; Avolio, Bass & Jung, 1995) was used to assess transformational leadership. The findings showed that a multidimensional first-order model and a hierarchical model both fined the data. Based on the evidence presented, it was argued that the MLQ assesses a single higher order construct of transformational leadership and that there is little evidence to support the contention that the MLQ measures distinct transformational leader behaviours.