The costs of comprehensively genotyping human subjects have fallen to the point where major funding bodies, even in the social sciences, are beginning to incorporate genetic and biological markers into major social surveys. How, if at all, should economists use and combine molecular genetic and economic data from these surveys? What challenges arise when analyzing genetically informative data? To illustrate, we present results from a "genome-wide association study" of educational attainment. We use a sample of 7,500 individuals from the Framingham Heart Study; our dataset contains over 360,000 genetic markers per person. We get some initially promising results linking genetic markers to educational attainment, but these fail to replicate in a second large sample of 9,500 people from the Rotterdam Study. Unfortunately such failure is typical in molecular genetic studies of this type, so the example is also cautionary. We discuss a number of methodological challenges that face researchers who use molecular genetics to reliably identify genetic associates of economic traits. Our overall assessment is cautiously optimistic: this new data source has potential in economics. But researchers and consumers of the genoeconomic literature should be wary of the pitfalls, most notably the difficulty of doing reliable inference when faced with multiple hypothesis problems on a scale never before encountered in social science.