Machine learning-structural equation modeling algorithm on online shopping customer satisfaction measurement

Abstract

Purpose-The purpose/aim of this empirical research was to know the impact of ease of use, service, convenience, reliability and security on online customer satisfaction. Further, to investigate the role of reliability and security on online customer satisfaction. There are three categories of variables mainly influencing on online customer satisfaction. They are 1.Control variables: gender, age, education, service and income. 2. Main variables: ease of use, service and convenience. 3. Interaction variables: reliability and security. Design/Methodology/Approach: Using a structured closed ended questionnaire to collect the data from 400 respondents. The questionnaire validated with pilot study followed by standardized loaded factors validated through confirmatory factor analysis and model fit was validated through GFI, AGFI, CFI and TLI >0.90. Researcher applied hierarchical regression analysis to see the impact of moderator on customer satisfaction. Findings- The outcome of the research proved that, ease of use, service, convenience, reliability and security have shown significant positive impact on online customer satisfaction followed by the moderating variables: security and reliability strengthened in such manner which leads to high customer satisfaction. The factor analysis also proved through standardized loadings with good model fit. Research implications- The outcome of the research can be generalized under any circumstances, situations and conditions. The proposed model was useful to understand online customer satisfaction level. In fact, the research was strengthened through literature review. Social implications- present study can be useful to judge the pulse of the online customer. Originality/value- This study provides new insight to the readers, which has not been designed earlier to understand the online customer satisfaction level. Paper type- by nature it is a research paper.