Development and Validation of a Depression Scale for Online Assessment: Cross-Sectional Observational Study

Abstract

BackgroundDespite increased awareness and improved access to care, depression remains underrecognized and undertreated, in part due to limitations in how current assessment tools capture emotional distress. Traditional depression scales often rely on fixed diagnostic language and may overlook the varied and evolving ways in which individuals express depressive symptoms—particularly in digital environments. Social media platforms have emerged as important spaces where people articulate psychological suffering through informal, emotionally charged language. These expressions, while nonclinical in appearance, may hold meaningful diagnostic value. ObjectiveThis study aimed to develop and validate the Depression Scale for Online Assessment (DSO), a tool designed to capture ecologically valid expressions of depressive symptoms as articulated in digital contexts. MethodsA cross-sectional, observational study was conducted with a community sample of 1216 adults, from which 1151 valid responses were retained for analysis. The scale’s items were developed based on expert reviews and social media research. To identify the factor structure, exploratory factor analysis (EFA) was conducted on a randomly selected half of the sample (n=575), followed by confirmatory factor analysis on the remaining half (n=576) to validate the model. Internal consistency was assessed following the EFA, and convergent validity was examined by correlating each DSO factor score with established depression measures, including the Korean version of the Center for Epidemiologic Studies Depression Scale-Revised and the Patient Health Questionnaire-9. ResultsEFA identified a 5-factor structure (ie, social disconnection, suicide risk, depressed mood, negative self-concept, and cognitive and somatic distress) that explained 66.53% of the total variance, indicating an acceptable level of explanatory power for a multidimensional psychological construct. confirmatory factor analysis indicated acceptable model fit (χ²109=403.5, P<.001; comparative fit index=0.96; Tucker-Lewis index=0.95; standardized root-mean-squared residual=0.03; root-mean-square error of approximation=0.07). The scale showed high internal consistency (total Cronbach α=0.95), and subscales were significantly correlated with the Center for Epidemiologic Studies Depression Scale-Revised (r=0.68-0.77) and the Patient Health Questionnaire-9 (r=0.64-0.74), supporting convergent validity. ConclusionsThe DSO is a psychometrically sound and clinically relevant tool that captures both core and emerging expressions of depression. Its digital adaptability makes it especially useful for research and clinical practice in mobile and remote care settings.