DOI QR Code
Purpose - The purpose of this study was to investigate several household characteristics related to panel attrition, examining how they may have conditioned the panel data in the Korea Welfare Panel Study (KOWEPS). Design/methodology/approach - We studied the cause of the differences in household income between the original panel and the new panel in KOWEPS. Findings - To summarize our findings, whereas it is highly likely that a low-income household or a household without health insurance will remain in the panel, it is highly likely that a high-income household or a household of more than three members will be taken off the panel. Research implications or Originality - The proportion of low-income household tends to decrease over the years, which appears to result from an overall increase in household income. Such changes are reflected in the pattern in which older panels have higher estimates of household income than newer panels.