Development and Validation of a Customer Satisfaction Measuring Instrument with Laboratory Services at the University Hospital of Kinshasa, Democratic Republic of the Congo (DRC)

Abstract

In clinical laboratory, monitoring customers’ satisfaction is required by laboratory quality standards, such as ISO15189:2012 and ISO17025:2017. However, there is no reliable and valid scale to measure clinical laboratory customers’ satisfaction in DRC. This study aims to develop a theoretical and operational instrument for measuring customer satisfaction with clinical laboratory services. In order to develop a reliable and valid measurement tool, the general methodological approach recommended by Churchill was followed. The developed questionnaire was checked for reliability and validity using exploratory and confirmatory factor analyses. Structural Equation Modeling (SEM) was used to determine whether the hypothesis of the conceptual framework is acceptable in measuring customer satisfaction. The Principal Component Analysis (PCA) results showed a three-dimensionality of Customer Satisfaction. Cronbachs alpha coefficients (0.983, 0.981 and 0.981), Joreskog Rho (0.973, 0.970 and 0.967) and composite reliability values (0.95, 0.92 and 0.93) of the latent variables were greater than 0.9, which confirms the very high reliability of the model. Indicator loadings were all greater than the threshold of 0.7 or higher. Also, all the latent variables have average variance extracted (AVE) greater than 0.5, therefore, convergent validity has been achieved. Both the Maximum Shared Variance (0.195, 0.297 and 0.234) and the Average Shared Variance (0.805, 0.703 and 0.766) were lower than the AVE (0.897, 0.839 and 0.875) for all the constructs in the scale. Therefore, Discriminant validity has been achieved. Fit indices used to assess CFA and structural equation model were found to be at an acceptable level for the two- factor model where chi-square/df was 1.6, p = 0.476, GFI = 0.99, AGFI = 0.99, SRMR = 0.069, RMSEA < 0.001, CFI = 1, NFI = 0.98, RFI = 0.98, IFI = 0.98, TLI = 0.98. The instrument demonstrated acceptable psychometric properties and thus it fits for measuring customers’ satisfaction with laboratory services.