DEVELOPING A PSYCHOLOGICAL DATASET FROM RAW SURVEY RESPONSES INTO LATENT FACTORS: AN EFA-BASED FRAMEWORK FOR MODELLING THE HAPPINESS SCORE OF STUDENTS

Abstract

Nowadays, the psychological health and happiness of students in Indian universities and colleges have been significantly affected by the COVID-19 pandemic, causing mental health problems, such as depression, anxiety, and stress. This research study presents a systematic framework for developing a psychological dataset from raw survey responses of students, combining the OHQ and the PWB scale. A total of 10,197 survey responses were gathered from students across various universities and colleges in Haryana through both online Google Form and offline modes.  After cleaning the data, calculations are performed to determine the happiness score and well-being. Factors are derived from PWB that affect well-being using predefined subclasses. After that, EFA was applied to OHQ to uncover latent psychological constructs underlying the responses. KMO value of 0.677 and a significant Bartlett’s test (p < 0.001) indicated that the sampling drawn from the population is appropriate for factor analysis. Four latent features, i.e. “Optimism and Life Satisfaction”, “Social Connectedness”, “Resilience and Personal Growth”, and “Confidence and Emotional Strength”, were identified, collectively explaining 9.3% of the total variance. Correlation analysis revealed that self-acceptance, purpose in life, and personal growth are key factors of well-being, whereas both income and charity indicated low correlations. For happiness, social connectedness, confidence, emotional strength, and life satisfaction, these factors play a crucial role. The outcomes suggest that emotional stability and psychological factors play a more influential role in predicting happiness than socio-economic factors. The Socio-demographic profile of students is also discussed in this paper. The resulting systematic dataset and analytical methodology become a platform for future research and machine learning applications in predicting mental health and measuring student happiness.