Objective: As social networking services (SNS) have become embedded in daily life, their excessive or inappropriate use has been associated with sleep disturbances, depression, anxiety, and related health problems, and SNS addiction has increasingly been recognized as a social concern in Japan. In the health sciences, determining how to weigh and pragmatically combine a population strategy (population-wide primary prevention) with a high-risk strategy (targeted support for individuals at elevated risk) requires empirical evidence regarding the latent structure of the target construct—specifically, whether the construct is best represented as a dimensional continuum or as a qualitatively distinct categorical subgroup. However, the latent structure of SNS addiction tendencies has not been tested using taxometric methods. This study examined the continuity/discontinuity of SNS addiction tendencies among Japanese adults and provided foundational evidence for the design of health support strategies.
Methods: A total of 1,788 Japanese adults, including university students, completed the Japanese version of the Social Networking Addiction Scale (SNAS-J). Six subscale scores—salience, mood modification, tolerance, withdrawal symptoms, conflict, and relapse—served as indicators. Taxometric analyses used three complementary procedures: Mean Above Minus Below A Cut (MAMBAC), Maximum Eigenvalue (MAXEIG), and Latent Mode (L-Mode). For each procedure, empirical curves were compared with simulated data generated under dimensional and categorical models, and model proximity was quantified using the Comparison Curve Fit Index (CCFI; CCFI < .45 = dimensional, CCFI > .55 = categorical).
Results: Across all procedures, empirical curves closely matched the simulated dimensional model and no peaks or multimodal patterns characteristic of a categorical structure were observed. The CCFI values were .313 (MAMBAC), .350 (MAXEIG), and .283 (L-Mode). All values fell below the .45 criterion, and the mean CCFI was .315.
Conclusion: SNS addiction tendencies are best conceptualized as a continuously distributed health risk among adults in contemporary Japanese society, where SNS use has become embedded in social life, rather than as a qualitatively distinct pathological group. These findings support prevention frameworks that prioritize population-based approaches while supplementing them with targeted support for vulnerable groups rather than relying on rigid dichotomous classifications.
View full abstract
