Loyalty and public transit: a quantitative systematic review of the literature

Cities are becoming more and more car-dependent. In tandem, public transit systems are facing an increasing loss of ridership. In this scenario, researchers have started investigating loyalty behaviour to understand what encourages user retention. This paper reviews the academic literature on loyalty formation in public transit, including (i) the research context; (ii) the methods applied and the assessment of heterogeneity; (iii) the conceptualisation and operationalisation of loyalty; and (iv) the influence of researched constructs on loyalty. Finally, potential avenues for future research are highlighted. The literature is largely concentrated in Asia, North America, and Europe. Most studies use a combination of quantitative data, cross-sectional design combined with survey data collection, modelling, and, to a lesser extent, clustering. In this sense, structural equation modelling (SEM) is broadly applied. The loyalty construct is usually operationalised according to attitudinal variables, while behavioural measures are less assessed. The research on this field is strongly influenced by marketing theory. Consequently, service quality, satisfaction, and perceived value are often modelled. Both satisfaction and service quality are repeatedly shown to have a strong positive influence on the development of loyalty. Nonetheless, the operationalisation of these constructs can differ greatly across studies, which makes it harder to compare the findings. Additionally, most SEM studies only report direct effects, which inhibit assessing the full influence of the variables on loyalty. There are also small clusters of research on other variables, such as image, involvement, problem experiences, and social psychology variables. To different degrees, these constructs are found to influence loyalty. Sometimes, even to a greater extent than service quality and satisfaction. In this sense, future research could benefit from further exploring these less research variables as they can bring new theoretical and operational perspectives to the current understanding of loyalty behaviour.