Reliability and validity testing of a Shortened Athlete Leadership Questionnaire

Mason B. Sheppard, Todd M. Loughead, Matthieu Boisvert 

Athlete leadership has predominately been measured using two inventories; the Leadership Scale for Sports (LSS; Chelladurai & Saleh, 1980) and Differentiated Transformational Leadership Inventory (DTLI; Callow et al., 2009). Although both inventories have been used to assess athlete leadership behaviours, they were originally conceptualized to measure leadership behaviours in other contexts. As such, researchers have started testing the psychometric properties of these inventories in relation to athlete leadership. To date, researchers have shown that both the LSS and DTLI have demonstrated adequate factorial validity (Callow et al., 2009; Vincer & Loughead, 2010). As noted by Vaughan and Laborde (2018), psychometric testing is an ongoing process to assess validity and reliably and the re-testing of prevalent scales should be a common practice in questionnaire development. While Boisvert et al. (2024) found adequate factorial validity when testing the LSS and DTLI independently, high intercorrelations were found between the dimensions of the LSS and DTLI. Consequently, Boisvert et al. combined both the LSS and DTLI into a shorter version, labelled the Athlete Leadership Behaviour Questionnaire (ALBQ), and found excellent model fit. The purpose of the current study was to further test the reliability and validity of the ALBQ. The convergent and discriminant validity was tested by calculating the Average Variance Extracted (AVE) of each construct. Further, in order to test reliability, McDonald’s Omega was analyzed. The results confirm convergent and discriminant validity, along with acceptable reliability coefficients of the ALBQ.