Statistics on the relative pay of public school teachers are routinely cited by plaintiffs in school finance ("adequacy") lawsuits. However, comparisons of pay and benefits for public school teachers to those of professional employees in other sectors are complicated by the fact that most teachers work under contracts that are nine or ten months in length rather than a full year. The authors show that this makes household survey data on weekly earnings in the widely used Current Population Survey (CPS-ORG) unreliable. In general, employer-reported data on salaries and benefits such as the National Compensation Survey (NCS) or state administrative data are preferred for this type of comparison. NCS data on weekly earnings in metropolitan labormarkets suggest that pay of public school teachers compares much more favorably to that of nonteachers than CPS-ORG data suggest.