One purpose of the present study was to develop and test the factor structure of a multidimensional and hierarchical Norwegian Principal Self-Efficacy Scale (NPSES). Another purpose of the study was to investigate the relationship between principal self-efficacy and work engagement. Principal self-efficacy was measured by the 22-item NPSES. Work engagement was measured by a modified version of the Utrecht Work Engagement Scale (UWES). The participants in the study were 300 principals randomly drawn from the population of Norwegian principals. Data were collected by means of an electronic questionnaire. Both the NPSES and the UWES were investigated through confirmatory factor analyses (CFA) before two structural equation models were tested. Both models specified principal self-efficacy as an exogenous variable and work engagement as an endogenous variable. The data were analyzed by means of a SEM analysis for latent variables using the AMOS 18 program. Three different models of NPSES were tested. Both a first- and second-order CFA confirmed that principal self-efficacy constitutes eight dimensions. Furthermore, both structural models had an acceptable fit to data and revealed that principal self-efficacy was positively related to work engagement. The results of the study are discussed together with limitations and suggestions for further research. (Contains 3 figures and 4 tables.)