|| Checking for direct PDF access through Ovid
The purpose of this study was two-fold. Firstly, to examine the measurement/factorial validity and invariance of the Referee Self-Efficacy Scale (REFS) among Norwegian soccer referees. Secondly, extending scale validation, we also tested a structural model in which a second-order version of the REFS was modelled to mediate a set of theoretically informed antecedents and outcomes.Cross-sectional.One hundred and eleven Norwegian elite referees and 81 non-elite referees completed an electronic questionnaire measuring expected antecedents and outcomes of perceived self-efficacy in the role as referees.Analyses provided support for the first – and second order versions of the REFS scale as well as for scale invariance. Further, a task goal and experience as referee related positively to referee positive affect, mediated by referee efficacy. Unexpectedly, physical preparation related directly to positive affect, whereas no mediation effects or direct effects for mental preparation and ego goal were observed.Measurement/factorial validity and invariance of the REFS were fully supported. Validity of the scale in mediation model by Guillén and Feltz (2011) was only partly supported.Testing the factorial validity and invariance of the Norwegian REFS scale.Cross-sectional study using latent variable modeling.Support for validity and invariance of the referee efficacy scale.REFS mediated years of experience and task orientation on positive affect.