Because research has shown that principals' practices can significantly impact teaching and learning in their schools through multiple avenues, it is critical to understand the validity of tools that measure principal practice. While the field has relied heavily on self-report surveys, little is known about their validity. This study compares survey measures of principal leadership practice with comparable daily log measures. We construct a multitrait-multimethod (MTMM) matrix showing intercorrelations among 4 dimensions of principal leadership captured with the 2 instruments. Correlations were evaluated against criteria used for MTMM analyses. Considerable evidence attested to the principal survey's validity, with correlations between log and survey measures of all 4 dimensions exceeding .50. Results suggest that surveys may be better at measuring principals' management of the school building and instructional leadership than measuring more irregular activities like planning and personal professional development. We discuss steps for improving the validity of principal surveys.