The objective of this paper is to test and correct for systematic differences in reporting of pain severity among older adults by age, gender, ethnic group and socio-economic status using anchoring vignettes. Data from a national survey of community-dwelling older Singaporeans (aged 60 years and over) conducted in 2009 was used. Respondents were asked to rate the severity of their own pain as well as that of others described in the vignettes on a five-point scale ranging from none to extreme. An ordered probit model was used to estimate the coefficients of the independent variables (age, gender, ethnic group, education, housing type) on self-reported pain. Reporting heterogeneity in pain severity was then corrected using a Hierarchical Ordered Probit model. The results showed that before correcting for reporting heterogeneity, women, those older, and those of Malay ethnicity reported greater severity of pain, while there was no association of reported pain severity with housing type and education. However, after correcting for reporting heterogeneity, while women and those older were found to have an even greater severity of pain than what they had reported, Malays were found to have a lower severity of pain than what they had reported. We conclude that there are systematic differences in reporting pain severity by age, gender and ethnic group. We propose that pain management may be improved if medical professionals take into account reporting heterogeneity for pain severity among various population sub-groups in Singapore.