Numerous studies have observed that lower SES is associated with increased risk of incident cancer for some sites and reduced survival for many cancer sites. One exception is for female breast cancer as higher SES is associated with increased risk and better survival. However, evidence of the impact of SES is limited in the fact that SES is often measured at an aggregate or area level (e.g., mean household income for a zip code or census tract, proportion poor in a block group). This proposal will implement a strategy that permits a direct measurement of SES at the individual level through record linkage to state birth certificate records available in the Utah Population Database. In possible collaboration with other SEER registries, this proposal will assess the feasibility of applying Utah?s linkage methodologies to a California population of pediatric cancer cases born on or after 1997, and compare and contrast results with the comparable pediatric population in Utah. This strategy may provide a better understanding of health and cancer disparities in terms of SES. The understanding of how SES affects these risks remains surprisingly underdeveloped. This deficiency in the literature is partly attributable to the fact that SES is often measured at an aggregate or area level (e.g., mean household income for a zip code or census tract, proportion poor in a block group). While there is no question that spatial indicators of SES are important, they are not substitutes for individual or family level measures of SES. Moreover, spatial indicators are treated as a static measure so that upward or downward changes in SES are not considered. Finally, the spatial data are often based on SES at the time of diagnosis, death, or at the time of a survey when it is likely that the SES value of interest could be from earlier in life. A review of 124 studies examining SES effects on cancer risk shows that while only about 25 percent collected individual SES data, most of these were based on small samples (N<5000) and where SES was measured once at the time of the survey.