Psychometric adaptation and performance of the Scale of Physical and Emotional Torment towards Animals

Abstract

Introduction: Physical and emotional torment towards animals during childhood and emerging adulthood has been associated with the diagnosis of various behavioral disorders. The aim of this study was to adapt and validate the Animal Physical and Emotional Torment Scale (PET) in a sample of university students from the Colombian Caribbean.

Materials and methods: This was a quantitative instrumental study with a cross-sectional design. Three models were exploratively evaluated in 100 university students, and two solutions of the instrument were confirmed in 140 university students.

Results: In the exploratory factor analysis (EFA), two models met the factor allocation criteria. The first model yielded a two-dimensional factorial structure that accounted for 68.39% of the variance in the data, with a significant Bartlett's sphericity (331,099, df=36, p<.002), an acceptable Kaiser-Meyer-Olkin measure of sampling adequacy (KMO=.73), and adequate total and dimensional internal consistency (ordinal total α=.825; ordinal subscale 1α=.93, and ordinal subscale 2 α=.75). The second model resulted in a unidimensional factorial structure that explained 73.16% of the variance, with appropriate indicators of sphericity, sample size, and consistency (216.369, df=6, p<.000; KMO=.73, and ordinal total α=.877). The explored solutions were confirmed, as both models showed satisfactory goodness-of-fit indices (two-dimensional model: X2/df=2.33, GFI=.991, AGFI=.969, CFI=.944, NFI=.970, RMSEA=.070, and SRMR=.070; unidimensional model: X2/df=1.98, GFI=.998, AGFI=.987, CFI=.968, NFI=.994, RMSEA=.068, and SRMR=.040).

Discussion and conclusions: The PET scale in its original and reduced versions is a valid and reliable instrument for assessing this construct in the Colombian sociocultural context.