Refinement of the technology readiness index scale A replication and cross-validation in the self-service technology context

Purpose - The purpose of this paper is to replicate and refine Parasuraman's 36-item technology readiness index (TRI) across contexts and cultures to enhance its applicability and generalizability for both researchers and practitioners. Design/methodology/approach - Based on psychometric procedures of scale development, four separate research phases, each one building on the previous, are performed using several samples. Measurement invariance analyses are performed across demographics, industries, and cultures to ascertain the stability of the refined versus the original scale. Findings - A refined 16-item TRI scale demonstrates sound psychometric properties based on findings from various reliability and validity tests, as well as scale replications employing several samples. The four dimensions remain stable across techniques and samples, while the utility of the refined scale increases due to ease of application. Measurement invariance analyses across demographic groups, industries, and cultures provide further support for the superior stability of the refined TRI. Research limitations/implications - Assessment of TRI across different contexts and cultures enhances validity, utility, and. generalizability by reducing the number of items, building a nomological network, and verifying stability. Practical implications - Service firms should pay more attention to measurement of customers' technology readiness. For both researchers and practitioners, the refined 16-item scale benefits from reduced complexity and enhanced utility of TRI across contexts and cultures. Service managers will find the refined TRI less complicated and easier to apply in customer surveys, which greatly benefits service firms attempting to better understand customers' TR when implementing self-service technologies. Originality/value - Replication and cross-validation of new concepts play a valuable role in determining the scope and limit of empirical research findings; they allow researchers to demonstrate how broadly and in what circumstances such concepts can be used. While Parasuraman calls for studies to assess the generalizability of the TRI scale, the current lack of support for TRI's generalizability is an important gap that needs to be addressed. The current study fills that gap, increasing the applicability and generalizability of the TRI scale through refinement, replication and validation across several samples, contexts, and cultures.