This study applied a choice-based conjoint (CBC) analysis to examine
the criteria by which fitness center users compare and evaluate
alternatives in real selection situations. Unlike prior research that has
mainly focused on perception- or satisfactio...
This study applied a choice-based conjoint (CBC) analysis to examine
the criteria by which fitness center users compare and evaluate
alternatives in real selection situations. Unlike prior research that has
mainly focused on perception- or satisfaction-based outcomes, this
study quantitatively identified how six core attributes—internal
environment, equipment, price, accessibility, operating hours, and human
services—shape users’ choice structures by utilizing data that mimic
actual consumer decision-making.
Data were collected via an online survey conducted in September
2025, yielding responses from 400 fitness center users. An L27
orthogonal design was generated using R to construct nine choice sets,
each presenting three alternatives from which respondents selected their
preferred option. After data cleaning and quality checks, the multinomial
logit model (MNL) was first estimated to derive market-level utility
structures, followed by the mixed logit model (MIXL) to capture
individual-level preference heterogeneity. Market segmentation analyses
were conducted across gender, age, income, exercise frequency, and
exercise purpose, and market share simulations and optimal-profile
predictions were performed based on estimated utility coefficients.
The results indicated that price and accessibility exerted the greatest
influence on choice behavior, whereas operating hours showed a
moderate effect. Internal environment, equipment, and human services
served as essential quality attributes, though their importance increased
within certain consumer groups. MIXL analysis revealed significant
variance in individual preferences, demonstrating that fitness center
choices cannot be explained by a single uniform pattern. Segmentation
analyses further showed that attribute priorities varied across
demographic and behavioral characteristics, confirming substantial
heterogeneity among user groups. Market-share simulations suggested
that economically efficient configurations yielded the highest selection
probabilities; however, quality-enhanced profiles showed greater utility
for specific segments, implying a diversified and stratified demand
structure.
By applying CBC to the fitness context, this study moves beyond
perception-based evaluations and provides an empirically grounded
utility structure derived from actual choice behavior. The findings offer
meaningful academic and practical insights for product development,
segmented market strategies, operational planning, and location decisions
within the fitness industry.