Ahmad Juma’h

This study examines how perceived security and privacy influence user ratings of cryptocurrency applications, which are critical for adoption and satisfaction amid the growing popularity of blockchain technologies and rising concerns over information security in online platforms and mobile apps. The study focuses on mobile applications from the Android app market. It used text mining methods to investigate over 64 thousand text-based user reviews and star ratings of over 140 cryptocurrency-related mobile applications available in the Google Play store. Using a partially supervised machine learning approach, this study first identified reviewer sentiment related to privacy and security, then employed ordinal regression analysis to examine the data to reveal the relationship between perceived security threats, privacy concerns, and app ratings. This study found that crypto apps average 3.84 out of 55 stars, which is higher than Productivity apps (3.46) but lower than FinTech (4.29) and Banking (4.25) apps. Ordinal regression analysis revealed security and privacy threats negatively impact ratings, while robust security measures improve them, with a model Pseudo R² of 0.25. These results have implications for both cryptocurrency app developers and platform managers, offering insights for enhancing user experiences and informing future research endeavors in this domain. It contributes to the literature by integrating the Protection Motivation Theory with sentiment analysis and provides a structured framework for developing an understanding of user behavior in the context of cryptocurrency apps.