The role of anxiety factors in predicting smartphone addiction mediated by metacognition among university students: a two-stage SEM-ANN approach.

Abstract

This study aims to investigate the associative factors related to smartphone addiction among university students and to further examine the mediating roles of positive and negative metacognition in the relationship between anxiety factors and smartphone addiction. The researchers randomly surveyed 760 university students from three universities in Henan province of China using a structured questionnaire to measure their self-reported responses on six constructs: academic anxiety, social anxiety, future anxiety, positive metacognition, negative metacognition, and smartphone addiction. By applying the Structural Equation Modeling-Artificial Neural Network (SEM-ANN) approach, the study interprets the non-compensatory and nonlinear relationships between predictor factors and smartphone addiction. The findings underscore that negative metacognition plays a significant mediating role in the relationship between social anxiety, future anxiety, and smartphone addiction, highlighting its critical influence in this association. The study found no significant association between academic anxiety and smartphone addiction, nor did positive metacognition mediate the associations between anxiety and smartphone addiction. Furthermore, based on the normalized importance derived from the Multi-layer Perceptron, the study identified the most significant predictive factor to be negative metacognition (100%), followed by future anxiety (46.7%), social anxiety (28.0%), positive metacognition (15.7%), and academic anxiety (10.2%). Finally, the study proposes theoretical and practical implications regarding associations with smartphone addiction among university students.