Can risky behaviors, gaming addiction, and family sense of coherence accurately classify gender among university students?

Abstract

The present study examined whether risky behaviors, gaming addiction, and family sense of coherence classify female and male university students correctly. For this purpose, Logistic Regression Analysis (LRA) was used. The main purpose of LRA is to estimate which group the individual is a member of. LRA is widely used to predict categorical data (i.e., male and female, addicted and non-addicted, single and married). The present study comprised 281 university students (148 females and 133 males) who had been playing digital videogames for at least six months. The measures used included the Risk Behaviors Scale, Digital Game Addiction Scale, and Family Sense of Coherence Scale-Short Form. Analysis demonstrated that risky behavior, gaming addiction, and family sense of coherence predicted gender with 76.5% accuracy. In other words, risky behaviors, gaming addiction, and family sense of coherence scores can predict a high probability that an individual is male or female. More specifically, risky behaviors (school dropout, antisocial behaviors, alcohol use, smoking, and eating habits) and gaming addiction contributed significantly to the classification of an individual being male or female. The biggest contribution in the classification of gender was digital gaming addiction. Family sense of coherence and risky behaviors such as substance use and suicide tendency did not contribute significantly to gender classification. The present study provided important resulted in terms of demonstrating risk factors related to gender.