Predicting collaborative consumption behaviour: a meta-analytic path analysis on the theory of planned behaviour

Abstract

Purpose Collaborative consumption (CC), a unique business model, provides several monetary and non-monetary benefits to customers. Several adapted theory of planned behaviour (TPB)-based models were developed and tested to understand this consumption behaviour with the findings inconsistent and fragmented. Thus, this study aims to develop a general and consistent TPB model using a meta-analytic path analysis to better understand customers’ CC adoption behaviour. Design/methodology/approach Using 37 studies, a meta-analysis was performed adopting several analytical methods; bivariate analysis, moderation analysis and path analysis. Findings The universal TPB model shows that factors, that is, trust, attitude, perceived environmental responsibility and communication facilities, drive both perceived usefulness and CC. However, subjective norms, such as perceived behavioural control and emotional value, drive only perceived usefulness. Moderation analysis shows that the relationships between variables used in the proposed TPB model tends to vary depending on five moderators, that is, countries’ economic development level, type of CC, sample size, sample type and survey administration method. Research limitations/implications The consideration of only quantitative papers and papers written in English language in this meta-analysis may bias the study’s findings. Practical implications Based on the findings regarding important factors that consumers consider when adopting CC, this study provides insightful recommendations to companies facilitating CC. Originality/value By developing the universal TPB model, this study theoretically contributes to the TPB model, and by conducting the moderation test, the study contextually contributes to the TPB literature in the CC context.

Online service failure: antecedents, moderators and consequences

“…(2014) adopted a meta-analysis approach to quantitatively synthesize the literature on “service failure experience”. It is important to note that a structured SLR is a different method from a meta-analysis (Ashaduzzaman et al. , 2022; Maseeh et al.…”

“…Van Vaerenbergh et al (2014) adopted a meta-analysis approach to quantitatively synthesize the literature on "service failure experience". It is important to note that a structured SLR is a different method from a metaanalysis (Ashaduzzaman et al, 2022;Maseeh et al, 2021a). That is, although both methods are used to synthesize the existing research in a domain, the former qualitatively synthesizes the entire volume of research in terms of theories, methods and constructs studied, while the latter makes a statistical assessment of extant quantitative studies in a domain (Maseeh et al, 2021b;Piper, 2013;Pati and Lorusso, 2018;Paul and Criado, 2020).…”

Anti‐consumption behavior: A meta‐analytic integration of attitude behavior context theory and well‐being theory

“…Various studies have been conducted on anticonsumption during this period; however, wide inconsistencies have been observed across these studies in terms of effect size of variables affecting anti-consumption behavior. Meta-analysts suggest that the variation in research design artifacts can be a possible reason for these inconsistencies (Ashaduzzaman et al, 2022;Lewin & Donthu, 2005;Maseeh et al, 2021b). A moderation analysis enables researchers to identify the possible reasons for inconsistencies in the effect sizes across the studies included in a meta-analysis (Van Vaerenbergh et al, 2014).…”

Artificial intelligence in marketing: A meta‐analytic review

“…Therefore, using the underpinnings of the theory of reasoned action and the unified theory of acceptance and use of technology 2, this study proposes an integrated Metaanalytic framework to synthesize the extant literature on artificial intelligence and to reveal concrete relationships between the drivers and customer behavioral responses to artificial intelligence. Furthermore, the literature suggests that several contextual and methodological factors are likely to account for inconsistencies in the results produced in empirical research (Ashaduzzaman et al, 2022;. For instance, in the artificial intelligence context, the results of empirical studies may vary depending on the type of artificial intelligence tools, such as virtualization, self-service, sensory enabled, and image interactive technology.…”

“…Furthermore, the literature suggests that several contextual and methodological factors are likely to account for inconsistencies in the results produced in empirical research (Ashaduzzaman et al, 2022; Maseeh, Jebarajakirthy, Pentecost, Arli, et al, 2021). For instance, in the artificial intelligence context, the results of empirical studies may vary depending on the type of artificial intelligence tools, such as virtualization, self‐service, sensory enabled, and image interactive technology.…”