Predicting the Acceptance of Informal Learning Technologies: A Case of the TikTok Application

Earlier research focused on investigating the acceptance of educational technologies applied in formal learning settings. Understanding the factors that can lead to the adoption of other dominant technologies in social communication for informal learning is an area that remains under-studied. Moreover, previous literature focused on the use of structural equation modeling (SEM) to predict technology acceptance, whereas the application of data mining algorithms is rare in this direction of research. This study, therefore, aims to (1) propose an integrated framework based on the DeLone and McLean information system model, the diffusion theory, the interactivity theory, the intrinsic motivation theory, and the security perceptions, (2) predict the adoption of TikTok as a learning means in an informal educational space, and (3) compare the performance of data mining techniques and SEM in predicting users’ behavioral intention towards TikTok acceptance. A cross-sectional survey research design is adopted to achieve the research goals. Data from 143 participants are collected and analyzed based on the convenience sampling technique. The partial least square, Support Vector Classifier, and Random Forest techniques are used to identify the predictability of the proposed framework. The findings suggest that the most influential constructs on TikTok adoption are perceived enjoyment, interactivity, security perceptions, and perceived satisfaction. Such factors explain about 83.2% of the variance of behavioral intention towards the adoption of TikTok for informal learning. The study also shows a clear similarity between the findings of SEM and data mining techniques in their overall prediction rate. The key implications of this research are twofold. First, it proposes a modified framework that explains a high variance of TikTok acceptance. Second, in informal learning contexts, particular constructs such as enjoyment, security, interactivity, and satisfaction can affect technology adoption more than information or system quality.