Do you see what I see? Investigating the validity of an equine personality questionnaire

Abstract

Subjective equine personality questionnaires have the potential to predict a range of industry-relevant outcomes including fear reactivity, compliance with human cues, pain expression and susceptibility to stereotypies, in a time- and cost-efficient manner. However, to produce meaningful measures of target animals’ behavioural tendencies, subjective personality assessment tools must satisfy four criteria: internal consistency, predictive validity, inter-rater reliability, and test-retest reliability. The Equine Personality Test (EPT) has been developed to assess horses on five personality factors based on trait ratings from a familiar observer. While the EPT has been shown to have predictive validity, it has not been assessed for internal consistency, inter-rater reliability or test-retest reliability. To this end, three experienced primary caregivers and three riding instructors assessed 25 familiar horses using the EPT. The internal consistency, inter-rater reliability and test-retest reliability of the five subscales of the EPT were investigated using Cronbach’s α and intra-class correlation (ICC) analyses. The Agreeableness, Neuroticism, Extroversion and Gregariousness towards People subscales had high Cronbach α and inter-rater and test-retest ICC coefficients (α > 0.7; ICC>0.8). By contrast, the Gregariousness towards Horses subscale had low Cronbach α (α = 0.39) and inter-rater ICC coefficient (ICC=0.498), and an adequate test-retest ICC coefficient (ICC=0.784). Primary caregivers had higher ICC coefficients than instructors for most subscales and questionnaire items. The EPT therefore provides internally consistent and highly reliable measures of Agreeableness, Neuroticism, Extroversion, and Gregariousness towards People in equines, although measures of Gregariousness towards Horses should be interpreted with caution. The reliability of EPT scores can be further improved by targeting primary caregivers as raters. Taken together with previous findings demonstrating predictive validity for the questionnaire, these results contribute to making the EPT the only subjective equine personality questionnaire to have been checked against all four criteria of a valid and reliable personality assessment tool. This positions the EPT as a highly relevant equine personality assessment tool that may be used to predict behavioural tendencies in industry or research settings alike.