Aim: To examine the construct validity of the Trait Emotional Intelligence Questionnaire Short form.
Background: Emotional intelligence involves the identification and regulation of our own emotions and the emotions of others. It is therefore a potentially useful construct in the investigation of recruitment and retention in nursing and many questionnaires have been constructed to measure it.
Design: Secondary analysis of existing dataset of responses to Trait Emotional Intelligence Questionnaire Short form using concurrent application of Rasch analysis and confirmatory factor analysis.
Method: First year undergraduate nursing and computing students completed Trait Emotional Intelligence Questionnaire-Short Form in September 2013. Responses were analysed by synthesising results of Rasch analysis and confirmatory factor analysis.
Results: Participants (N = 938) completed Trait Emotional Intelligence Questionnaire Short form. Rasch analysis showed the majority of the Trait Emotional Intelligence Questionnaire-Short Form items made a unique contribution to the latent trait of emotional intelligence. Five items did not fit the model and differential item functioning (gender) accounted for this misfit. Confirmatory factor analysis revealed a four-factor structure consisting of: self-confidence, empathy, uncertainty and social connection. All five misfitting items from the Rasch analysis belonged to the 'social connection' factor.
Conclusions: The concurrent use of Rasch and factor analysis allowed for novel interpretation of Trait Emotional Intelligence Questionnaire Short form. Much of the response variation in Trait Emotional Intelligence Questionnaire Short form can be accounted for by the social connection factor. Implications for practice are discussed.
Keywords: TEIQue-SF; emotional intelligence; factor analysis; gender; nursing; rasch analysis; recruitment; retention; trait emotional intelligence questionnaire short form.
© 2015 John Wiley & Sons Ltd.