Online shopping attributes and its influence on customers satisfaction, trust and behavioural intention : an empirical study

Abstract

Shopping is an integral part of one’s daily life routine. In online shopping, buying and selling goods and services is done on a virtual platform (here, e-stores of retailers) with the help of the Internet. Online shopping helps retailers to get connected with their customers 24x7. Similarly, it allows buyers to place desired orders on the retailers’ e-stores (or web stores) anywhere and anytime. However, customers cannot interact face-to-face with retailers in online shopping. Moreover, they can only physically evaluate goods and services (such as touch, smell or test) once they receive and use them. Therefore, it is challenging to satisfy and build trust among online shoppers towards online shopping compared to traditional shopping. Consequently, it is necessary to evaluate (or a research gap) how e-retailers are satisfying customers and building trust among customers to adopt online shopping platforms. Moreover, it also needs to be checked that if the customers are satisfied with the online shopping platform and the online sellers successfully build trust, do they adopt online shopping into their behavioural intention? This study conducts empirical research to answer the above questions and bridge this gap in the body of knowledge. Based on previous research and theories in the areas of online shopping, the study identified the relevant online shopping dimensions (e.g., product reviews, perceived risk, website interface quality, perceived security, customer trust, customer satisfaction, and customer purchase behavioural intention) that influence customers’ buying behavioural intention on online shopping platforms. The study developed a structured questionnaire to measure the identified online shopping dimensions and analyze the online buyers’ demographic profiles.

The study uses a five-point Likert scale to measure the items/questions of various online shopping parameters. In this five-point Likert scale, five represents “strongly agree”, four represents “agree”, three represents “undecided/neutral”, two represents “disagree”, and one represents “strongly disagree”. The target population of this study is college and university-going students who buy goods and services online. The study applied convenience sampling {as suggested by Gopinath (2021), Sunitha & Gnanadhas (2014), and Dani (2017)} to select institutions, departments, and centres. After that, the study used systematic random sampling techniques {as suggested by Alwan & Alshurideh (2022), Farzin et al. (2022), and Ariansyah et al. (2021)} to collect the responses from online shoppers. The study takes the help of Cochran’s formula (1977) to determine the sample size for an infinite population. The study took a sample of 576 online shoppers to explain the purchase behaviour intention of customers in online shopping platforms. The study considers parameters such as gender (Slyke et al., 2002), age (Khare et al., 2012), educational qualification (Susskind, 2004) and income (Mahmood et al., 2004) to analyse online customer demographic profiles. In addition, the study takes two new variables, called “payment method” and “time spent on the Internet” (suggested by Brown et al., 2003), to gauge consumers’ payment method preferences and Internet experience. The study used Exploratory Factor Analysis (EFA) with Principal Component Analysis (PCA) method to extract the underlying dimensions of online shopping that influence customer buying behavioural intention on online shopping platforms. Furthermore, the study used varimax rotation with Kaiser normalisation to obtain the Rotated Component Matrix (RCM).

The univariate normality of the data is checked with descriptive statistics, such as mean, standard deviation, skewness, and kurtosis. The study takes the help of Mardia’s coefficient test (1970) to test the multivariate normality of the data of online shopping parameters. The study examined the internal consistency in scale items or reliability of the online shopping construct with the help of “Cronbach alpha (α)” and “Composite reliability (CR)”. The validity of an online shopping construct is examined with the help of “discriminant validity” and “convergent validity”. The study established discriminant validity by average variance extracted (AVE) and convergent validity by the Fornell-Larcker test. The study develops an online shopping behaviour intention measurement model, structural model and respecified structural model of customers with the help of a rotated component matrix using statistical software (AMOS). The study established the fit indices of these models with the help of various specified model fit indices, such as the overall fit index (i.e., CMIN), absolute fit index (i.e., GFI, RMSEA, RMR, SRMR, and Normed chi-square), incremental fit index (i.e., NFI, CFI, and RFI) and parsimony fit index (i.e., AGFI and PNFI). The study found that women (57.5 per cent) are more inclined towards online shopping than men (42.5 per cent). Compared to shoppers in other age groups (such as up to 20, 26 to 30 and over 30), shoppers aged 20 to 25 are more interested in online shopping (40.8 per cent). Of the 576 online shoppers, 53.5 per cent are pursuing graduate programs. The study reveals that 33.9 per cent of online shoppers have a household income between Rs 2.5 lakh to Rs 5 lakh. Furthermore, the study shows that 46.5 per cent of online shoppers prefer the cash-on-delivery (COD) option payment method, and 50.2 per cent surf the Internet for 2 to 4 hours per day.

The statistical results of this study show that perceived security (PSEC), product review (PRV), and perceived risk (PRK) affect both customer satisfaction (CSAT) and trust (CTRT). Web interface quality (WIQ) affects customer trust (CTRT) but does not affect customer satisfaction (SAT). Customer satisfaction (CSAT) is influenced by customer trust (CTRT) in the online shopping platform. Furthermore, the study shows that the behavioural intention (BI) of customers in online shopping platforms is directly influenced by customer satisfaction (CSAT) and trust (CTRT) and, indirectly, by perceived security (PSEC), product reviews (PRV), perceived risk. (PRK), and Web Interface Quality (WIQ). The square multiple correlations (R2) of customer satisfaction (CSAT) and customer trust (CTRT) in the online shopping platform are 0.43 and 0.38, respectively. This means that the proposed model (i.e., a model for estimating customer’s purchase behaviour intention in online shopping platforms) explains 49 and 38 per cent variation in customer satisfaction (CSAT) and customer trust (CTRT), respectively, with the help of taking online shopping factors in this study. The square multiple correlations (R2) of customer behavioural intention (BI) in the online shopping platform is 0.32. This means that the proposed model explains a 32 per cent variation in customer purchase behaviour intention (BI) with the help of taking online shopping factors in this study. There are some research limitations of this study. This study does not consider the responses of other online shoppers (such as housewives, senior citizens and professional online shoppers). This study proposed a customer purchase behaviour model on online shopping platforms considering relevant dimensions (such as product reviews, perceived risk, perceived security, website interface quality, customer trust, customer satisfaction and customer behavioural intentions). However, these online shopping dimensions are only indicative lists and not exhaustive lists of online

shopping dimensions. Since online shopping uses technology and the Internet, it can be a new dimension if any technological innovation is adopted to make online shopping convenient. Hence, online shopping dimensions are dynamic as technological innovations are dynamic. Thus, assessing customers’ buying behaviour on online shopping platforms is dynamic and continuous, and the research on online shopping is considered a never-ending process.