Red Box, Green Box - Psychometric evaluation of a self-report behavioral frequency measurement approach for behavioral addictions research: Data and research objects.

Published: 15 August 2025| Version 1 | DOI: 10.17632/rpbrhknnb3.1
Contributors:
Matthew Stevens,
,
,
,
,
,
,

Description

This dataset contains psychometric validation data for the novel "Red Box, Green Box" method of measuring gaming behavior, specifically designed to improve identification of gaming disorder (GD) risk compared to conventional total gaming time measures. Our research hypothesis was that Red Box hours (gaming instead of fulfilling responsibilities) would demonstrate superior diagnostic accuracy for identifying GD compared to traditional weekly gaming hours and Green Box hours (gaming during free/leisure time). The data shows that gamers with GD reported significantly higher Red Box hours (M=21.1, SD=11.3) than those without GD (M=8.7, SD=8.4; p<.001), with a greater Red Box proportion (41.9% vs. 26.8%). Most notably, Red Box hours demonstrated excellent diagnostic accuracy for ICD-11 GD (AUC=0.86, sensitivity=0.94, specificity=0.63) and good accuracy for DSM-5-TR IGD (AUC=0.76, sensitivity=0.88, specificity=0.56), substantially outperforming conventional total weekly gaming hours (AUC<0.70). Key findings include: (1) A Red Box threshold of ≥9.5 hours yielded 94% likelihood of indicating ICD-11 GD; (2) Total gaming time measures closely approximated Green Box values, suggesting conventional approaches primarily capture leisure gaming; (3) The method effectively distinguishes problematic displacement gaming from recreational gaming; (4) Red Box hours showed stronger correlations with GD status (r=.19) than other gaming time measures when controlling for demographics, impulsivity, and psychological distress. The dataset includes responses from 1,149 male gamers aged 18-35 years reporting ≥12 hours weekly gaming, recruited via Prolific from Australia, Canada, USA, UK, and Asia. Variables include demographic data, gaming time measures (conventional weekly hours, Red/Green Box hours, proportions), gaming disorder assessments (IGDT-10 with both ICD-11 and DSM-5-TR scoring), psychological distress (DASS-21), and impulsivity (BIS-15). This data enables researchers to: validate the Red Box, Green Box method in other populations; develop gaming disorder screening protocols; investigate behavioral patterns in problematic gaming; and advance understanding of functional impairment in behavioral addictions. The method's focus on contextual gaming behavior (displacement vs. recreational) provides a more nuanced approach to assessing gaming-related harm than traditional frequency-only measures.

Files

Steps to reproduce

This prospective, cross-sectional observational study used an online crowd-sourcing approach to gather psychometric validation data for gaming behavior measurement tools. Participant Recruitment: Participants were recruited through Prolific, an online research platform, from a pool of 6,237 eligible candidates across Australia, Canada, USA, UK, and East Asia. Inclusion criteria: male gender, age 18-35 years, self-reported gaming ≥12 hours weekly. Standard compensation of 8 GBP/hour was provided. Final sample: N=1,149. Data Collection Platform: All surveys administered via Qualtrics online platform. Participants completed a comprehensive battery including demographics, gaming behavior measures, clinical assessments, and individual difference measures. Key Measures Implementation: Conventional gaming time: "During the past 6 months, in a typical week, how often do you spend playing video games?" (dropdown selection) Green Box: "Total time (hours) in a typical week gaming during free/hobby time or when passing time" Red Box: "Total time (hours) in a typical week gaming when you feel you should be doing something else (studying, working, sleeping, exercising, etc.)" Gaming Disorder: Internet Gaming Disorder Test (IGDT-10) with dual scoring (ICD-11: 4+ criteria including 3 essential items; DSM-5-TR: ≥5 criteria) Psychological Distress: DASS-21 (depression, anxiety, stress subscales) Impulsivity: BIS-15 (attentional, motor, non-planning subtypes) Statistical Analysis Protocol: Conducted in R version 4.4.2/RStudio 2024.10.31. Methods included: Independent samples t-tests for group differences (GD vs. non-GD) Fisher's exact tests for categorical variables Zero-order and partial correlations (controlling demographics, distress, impulsivity) ROC curve analyses using Youden's index for optimal cut-offs Diagnostic accuracy indices: sensitivity, specificity, PPV, NPV, likelihood ratios, clinical utility indices Power Analysis: Conducted using pROC package, targeting AUC difference detection between correlated ROC curves. Assumed conventional gaming hours AUC=0.70, Red Box hours AUC=0.80, correlation=0.50, α=0.05, power=0.90. Required minimum 25 GD cases; based on 3% population prevalence, targeted sample size ≥834 participants. Data Processing: Green/Red Box responses presented sequentially (fixed order) on same survey page with modification option before proceeding. Calculated derived variables: total red+green hours, Red Box proportion. Applied ICD-11/DSM-5-TR scoring algorithms to IGDT-10 responses. Reproducibility: De-identified dataset and complete R analysis script are available here. All analyses exploratory (study was not pre-registered). University ethical approval obtained (ID: HEG7386-5). Quality Control: English proficiency not assessed; order effects not evaluated; limitations acknowledged in manuscript.

Institutions

Categories

Addiction, Psychometrics, Computer Gaming, Behavioral Addiction, Addictive Behavior

Funders

Licence