Restaurant Physical Environment and Diners’ Loyalty: A Multigroup Structural Equation Model of Occasional and Frequent Customers in Northern Peru

Read full text
Authors: Marco Agustín Arbulú Ballesteros, Christian David Corrales Otazú, Sarita Jessica Apaza Miranda, Sofía Irene Delgado Wong, Irma Rumela Aguirre Zaquinaula, Ericka Julissa Suysuy Chambergo, Luisa Angelica Orejuela Guerrero, Cecilia Paula Luisa Gómez Zúñiga, Victor Ángel Ancajima Miñán, Mirian Azucena Vilela Ubillús
Cited by: 1

Abstract

Diner loyalty is a critical asset for the restaurant industry, and servicescape models typically assume that all customers weigh the physical environment, perceived price, and satisfaction equally. This study submits that assumption to a formal test in full-service restaurants of northern Peru, across five cities: Chiclayo, Trujillo, Piura, Cajamarca, and Tumbes. A cross-sectional survey of 705 diners was analyzed with covariance-based structural equation modeling in Mplus 8.3 (robust maximum likelihood; 10,000 bootstrap draws with bias-corrected confidence intervals), complemented by measurement invariance testing, a multigroup comparison of occasional (n = 304) and frequent (n = 401) customers, and an importance-performance analysis computed in RStudio. Thirteen of the sixteen direct and mediation hypotheses were supported: decor and ambient conditions shaped price perception and satisfaction, satisfaction was the near-exclusive antecedent of loyalty (β = .940; R² = .883), and the serial chain servicescape → price → satisfaction → loyalty held for decor and ambient conditions but not for spatial layout. None of the four between-segment differences reached significance (Wald χ²(2) = 3.56, p = .169), although the decor–satisfaction path was significant only among frequent customers. The evaluation chain proves structurally robust across segments: managers can anchor strategy on satisfaction while directing decor investment toward the repeat clientele.

Publication details

Journal / source
Figshare
Publication date
4 July 2026
Document type
Dataset