College student Fear of Missing Out (FoMO) and maladaptive behavior: Traditional statistical modeling and predictive...

Area of Science:

• Psychology
• Social Sciences
• Data Science

Background:

• Fear of Missing Out (FoMO) is increasingly recognized as a significant factor influencing behavior.
• Understanding the link between FoMO and maladaptive behaviors in college students is crucial for intervention and prevention strategies.

Purpose of the Study:

• To investigate the relationship between college student FoMO and a spectrum of maladaptive behaviors.
• To evaluate the predictive power of FoMO and demographic variables using both traditional statistical methods and supervised machine learning.

Main Methods:

• A cross-sectional study involving 472 college students who completed questionnaires on FoMO and maladaptive behaviors.
• Part 1: Hierarchical regression modeling to assess relationships between FoMO, demographics, and behaviors.
• Part 2: Supervised machine learning techniques (RFE, PCA, logistic regression, random forest, SVM) to quantify predictive accuracy.

Main Results:

• College student FoMO significantly predicts a range of maladaptive behaviors, from substance use to academic misconduct and illegal activities.
• Machine learning models demonstrated high predictive accuracy for classifying offenders versus non-offenders, significantly exceeding baseline rates (e.g., 87% for academic misconduct).
• FoMO and demographic variables were identified as key predictors by machine learning models.

Conclusions:

• FoMO is a robust predictor of diverse maladaptive behaviors among college students.
• Machine learning offers powerful tools for uncovering predictive insights into complex behavioral patterns, complementing traditional statistical approaches.
• Integrating machine learning with established statistical methods can enhance research in psychology and the social sciences.