Validating the remote first aid self-efficacy scale for use in training and personal development of remote first responders

Abstract

This study was designed to investigate the psychometric properties of the 30-Item Remote First Aid Self-Efficacy Scale (RFA SES), develop a shorter 15-Item RFA-SES, and gain a preliminary understanding of the psychometric properties of the shorter 15-Item version. The original 30- Item RFA SES was developed for two main purposes: (1) evaluation of wilderness first aid and other types of emergency care training designed for remote geographies, communities, and worksites; and (2) reflection by training participants so they can self-assess their beliefs, confidence, and capacity to respond. Students from Laurentian University (LU) and graduates from Wilderness Medical Associates (WMA) training courses were recruited to respond to an online questionnaire at two different time periods (T1 and T2). A total of 1106 students and 448 graduates from WMA responded at T1. These results demonstrated that the RFA SES was a unidimensional scale with an eigenvalue of 18.1 at T1. The mean inter-item correlation was 0.75 at T1. Test-retest reliability (T1 to T2) was high for both the LU group (r = .91, p < .01) and the WMA group (r = .92, p < .01). Moderate correlations were found between the RFA SES and two other similar scales (included in the questionnaire) measuring different constructs. WMA participants showed higher mean scores than LU students at T1 (t (569) = 16.2, p < .01, twotailed). The 30-Item RFA SES is a unidimensional, reliable, and valid scale for assessing remote first aid self-efficacy. To develop the 15-Item RFA SES, three members of the research team completed an iterative process to reduce items based on expert opinion and statistical performance from the inter-item correlations. Preliminary analysis of the 15-item version of the RFA SES, using data from the WMA group (n=448), also indicated that the scale was reliable and valid. However, a more rigorous validation study with original data is required.