Date of Award


Degree Name

Doctor of Philosophy


Educational Studies


Structural equation modeling (SEM) was used in this dissertation to investigate the factorial invariance (i.e., equivalence) of the leadership construct as perceived by different groups of people in education. Specifically, the author focuses on three groups (364 teachers, 419 principals, and 369 superintendents) from a statewide survey, where the central concern is whether components of the measurement model of leadership are invariant across those particular groups. In seeking evidence of multigroup invariance, the author was interested in finding the answer to the following research questions: (1) to what extent do some leadership factorial structures fit a group better than others? and (2) to what extent is the factorial structure of leadership invariant across the three groups of teachers, principals, and superintendents? Respondents’ attitudes toward leadership are related to how they perceive and interpret mental models for this construct. Hence, evidence to support the invariance of factorial structure across groups serves to strengthen the research on educational leadership. The author drew on leadership theory to develop the basic structure of two measurement models, then LISREL 8 was used to estimate the parameters describing the relationships between the observed variables and the factors proposed in these models. As to the best fitting model, teachers’ conception on leadership is more consistent with the four-factor model, while principals’ and superintendents’ conceptions are more consistent with the two-factor model. Furthermore, configural and weak invariance across groups for the two measurement models are supported by the data. Therefore, limited inferences were made (from a more robust measurement position) about the relationship between respondents’ administrative level and the emergence of differences in the mental models held by them for the leadership construct. Implications of the findings are discussed in the dissertation.

Access Setting

Dissertation-Open Access