Network analysis of Korean happiness: a comparative study by sex, generation and income level

South Korea has one of the highest suicide rates among Organisation for Economic Co-operation and Development countries, and lower happiness has been associated with suicidal ideation and suicide attempts in Korean populations. We aimed to estimate a nation-level item network of happiness, affect, personality and well-being evaluations in South Korea and to test whether this network differed by sex, income and generation.

We analysed 2020-2023 Korean Happiness Survey data from 64,531 respondents in South Korea; the network analyses used 61,675 complete cases. A Gaussian graphical model with Least Absolute Shrinkage and Selection Operator regularization estimated edges among happiness, affect, personality and well-being evaluations. Subgroup differences by sex, income and generation were assessed with permutation-based Network Comparison Tests.

The network contained 199 of 435 possible edges (45.7%). The strongest affective and happiness-related pairwise connections included happiness-enjoyment, stress-tiredness, sadness-depression and depression-anger. Global strength did not differ significantly across the planned subgroup comparisons, but network structure differed by sex, income, Millennials and Generation Z (MZ) versus Generation X, and MZ versus Baby Boomers. In the income comparison, happiness was more negatively linked to the agreeableness item 'generally trusting' in the low-income network, whereas its link with time for hobbies was stronger in the high-income network. In the generational comparison, the happiness-stress edge differed between the MZ generation and Baby Boomers, whereas no happiness-connected edge survived false discovery rate correction for MZ versus Generation X. No happiness-connected edge differed significantly by sex.

In South Korea, happiness was most closely linked to enjoyment and embedded within tightly connected affective and evaluative domains. The subgroup differences were selective rather than global, suggesting that future work relevant to population well-being and suicide prevention may benefit from focusing on specific happiness-linked domains such as positive affect, stress-related ties, discretionary time and perceived future security.