IJHM_ServiceRituals2026

Description

Data for IJHM

Steps to reproduce

Design and ethics. The data come from a cross-sectional online survey of customers of a ritual-intensive hotpot restaurant (Haidilao). The study received institutional research-ethics approval, and all respondents gave informed consent. Instrument. The questionnaire adapted established, previously validated scales for each construct. It opened with a short definition and three photographs of the focal service rituals (noodle-making performance, face-changing show, birthday celebration) to anchor recall. Items were translated into Chinese using Brislin's (1970) back-translation. Response formats varied deliberately as a procedural common-method remedy: novelty, personalisation, entertainment and experience co-creation used five-point scales; functional, social and emotional value and memorable experience used six-point scales; subjective well-being used seven-point scales. A marker variable (attitude toward the colour blue) and a single global perceived-value item were included for common-method-variance and redundancy checks. The instrument was pretested through cognitive interviews. Sampling and administration. Adult diners who had eaten at Haidilao and experienced the rituals were recruited through Chinese social-media platforms and completed the questionnaire online. After screening and data cleaning (removing inattentive or straight-line responses, duplicate IP/device entries, and outliers), 499 valid cases remained. Dataset. The shared Excel has a Data sheet (499 cases × 51 model variables) and a Codebook sheet documenting each variable, its construct, role and scale. Analysis workflow. Descriptive statistics and Mardia's multivariate normality were computed in IBM SPSS 29. The model was estimated in SmartPLS 4 (Ringle, Wende & Becker, 2024). Perceived value was specified as a higher-order formative construct via the disjoint two-stage approach. We ran the PLS-SEM algorithm, then bootstrapping (10,000 subsamples, one-tailed, percentile confidence intervals), modelled experience co-creation as a second-stage moderator with the index of moderated mediation, included the five controls and the marker, and assessed predictive validity with PLSpredict and CVPAT plus a Gaussian-copula endogeneity check.

Institutions

• University of Malaya

  Kuala Lumpur, Kuala Lumpur

Categories

Licence