An assessment of the emotional intelligence of health workers

Abstract

Purpose This study aims to assess health workers’ level of emotional intelligence (EI) in Accra North and recommend a simple but robust statistical technique for compulsorily validating EI measurement scales. Design/methodology/approach The researchers used a self-reported questionnaire to collect data from 1,049 randomly selected health workers. Two non-nested models, BNK MODEL and CMODEL, were compared to see which of them better fits the study population and yields a better level of EI. The one-sample and independent-samples t-tests, exploratory factor analysis and confirmatory factor analysis were used to present results. Findings The study found that health workers were appreciably emotionally intelligent for both models at the 5 per cent significance level. However, EI was higher for the CMODEL. The CMODEL also better fits the study population (χ2 = 132.2, p = 0.487, Akaike information criterion = 124.932) and thus better underlies EI in it. This study recommends proper validation of the two EI scales evaluated in this study, and possibly other scales, before the use of their data in research, as failure to do so could lead to unrealistic results. Originality/value Apart from its contribution to the literature, this study provides a robust statistical approach for assessing health workers’ EI and validating EI scales. By comparing two models of EI in the validation process, this paper suggests that the researcher’s choice of a measurement scale can influence his/her results.

Construction of a Scale Assessing Patients’ Perceptions Regarding Sanitation and Hygiene in a Clinical Setting

“…We used confirmatory factor analysis to further confirm our 2-factor solution realized in principal component analysis. Interestingly, the results of the confirmatory analysis reflect ample discriminant and convergent validity of the questionnaire based on recommended baselines 8,14,16,20 . Some previous studies have produced reliable scales based on less robust statistical procedures, 9,16 but not limited to the use of principal component analysis.…”

“…Internal consistency of the scale was assessed with Cronbach α coefficient values and factor loadings estimated in confirmatory factor analysis. The rule of thumb is α ≥ 0.7 for each dimension or factor and the overall scale 8,20 . Each item should also produce a factor loading of 0.5 or greater 8,16 .…”

“…Two measurement models were then fitted to estimate the psychometric properties of the scales or confirm that their factor solutions reached in principal component analysis. Multivariate normality of the data, which is a requirement for confirmatory factor analysis, was met in the 2 measurement models based on criteria applied elsewhere 8,19–22 . In agreement with some studies, 15,16,23–25 Pearson correlation test and multiple regression analysis were used to assess concurrent validity.…”

“…Multivariate normality of the data, which is a requirement for confirmatory factor analysis, was met in the 2 measurement models based on criteria applied elsewhere. 8,[19][20][21][22] In agreement with some studies, 15,16,[23][24][25] Pearson correlation test and multiple regression analysis were used to assess concurrent validity. Cohen criterion (r < 0.3 = small; 0.3 ≤ r < 0.5 = medium; and r ≥ 0.5 = large) 23 was applied in the correlation analysis.…”

“…The rule of thumb is α ≥ 0.7 for each dimension or factor and the overall scale. 8,20 Each item should also produce a factor loading of 0.5 or greater. 8,16 Both criteria were met (Tables 1, 2).…”