Technological innovation is transforming the sports services sector, with mobile applications becoming crucial for fitness centers to maintain user engagement and loyalty. This study aims to determine how customer evaluations of a fitness app influence their recommendations. The research questions address which factors affect future app recommendations and whether age and gender play a role. This investigation utilizes two methodologies – linear models and fuzzy-set qualitative comparative analysis (fsQCA) – to provide a comprehensive understanding of customer perceptions and inform strategies for improving competitiveness. Linear models assess the individual contribution of each variable, while fsQCA explores the combined effects of variables, offering a more nuanced perspective on the causal relationships. The study hopes to provide insights into optimizing app design and user experience for maximum positive recommendations.

The literature review covers new technologies in the sports sector, focusing on the impact of wearable technology and mobile health (mHealth) apps, particularly fitness apps. Studies highlight the use of wearable technology to monitor performance and training, improving athlete outcomes. The review discusses the importance of app quality, measured by scales such as the Mobile App Rating Scale (MARS), which comprises dimensions of engagement, functionality, aesthetics, and information. Existing research emphasizes the role of app quality in user satisfaction and intentions to use fitness apps. However, the review also points out the need for standardized approaches to evaluate app quality and the growing need for research on the impact of mobile devices on health behaviors. The review underscores the importance of app design, functionality, and the accuracy of information provided to enhance user experience and recommendations.

The study employed a sample of 210 participants (54 females, 156 males) using convenience sampling. The Mobile App Rating Scale (MARS), validated in Spanish, measured app quality across four dimensions: engagement, functionality, aesthetics, and information. A 5-point Likert scale assessed user perceptions. Data analysis involved two approaches: hierarchical regression analysis and fuzzy-set qualitative comparative analysis (fsQCA). Hierarchical regression examined the individual influence of MARS dimensions, age, gender, and service fidelity on app recommendations. fsQCA analyzed the combinations of variables influencing high and low app recommendations, revealing causal configurations and equifinality. The fsQCA analysis employed calibration values to transform continuous variables into fuzzy sets, allowing for the identification of necessary and sufficient conditions for high and low levels of app recommendations. Consistency and coverage were used to evaluate model fit. The software used was SPSS and fsQCA 2.0.

Hierarchical regression analysis revealed that aesthetics and information significantly predicted app recommendations (R² adjusted = 0.52). Adding age, gender, and fidelity to the model did not significantly improve predictive power. fsQCA analysis showed no necessary conditions for high or low recommendations. Sufficiency analysis identified several configurations predicting high app recommendations. One key configuration involved a high level of engagement and being female. Other configurations highlighted the importance of information quality. For low app recommendations, low functionality across all paths emerged as an important influence. The fsQCA approach also revealed that older users were more likely to provide positive recommendations, contrasting with younger users who were more likely to offer negative feedback. The analysis demonstrates the complementary nature of the two methodologies; linear models revealed the individual impact of specific variables, while fsQCA highlighted the more complex interactions between them.

The findings demonstrate the crucial role of app quality dimensions (aesthetics and information) in influencing app recommendations. The fsQCA results reveal the complex interplay between app quality and demographics, highlighting the importance of considering multiple interacting factors rather than focusing solely on individual variables. The significant influence of engagement, particularly among women, suggests a need to focus on interactive app features that increase user involvement. The importance of accurate and reliable information reinforces the need for high-quality content in health-related applications. The contrasting influence of age on app recommendations points to the necessity of tailoring app design and features to specific user groups. The study validates the use of complementary methodologies (linear regression and fsQCA) for a thorough understanding of user perceptions and behavior.

This study highlights the importance of app quality, particularly aesthetics and information, in driving positive app recommendations. FsQCA analysis emphasizes the complex interplay of app features and user demographics, revealing that engagement is especially significant for women. Older users showed a greater tendency for positive recommendations than younger users. The complementary use of linear models and fsQCA provides a more comprehensive understanding of the factors influencing app recommendations. Future research could explore the relationship between app recommendations and user loyalty, investigate the impact of app features on different user demographics, and analyze public fitness center apps to explore variations in app design and user perceptions.

The study's limitations include the sample size and the focus on a single private fitness center in Spain, limiting the generalizability of findings. The model only analyzed app quality dimensions and their relationship with recommendations, not considering other variables like user satisfaction or loyalty. Future research should incorporate larger, more representative samples, incorporate subjective factors such as user pathologies, and explore the impact of app features on different user groups beyond gender and age.