This study investigated use of the Renaissance Partnership Teacher Work Sample (RTWS) as an accountability measure for demonstrating teacher candidates' ability to meet targeted teaching standards. Teacher work samples were collected from nine universities participating in the Renaissance Partnership to Improve Teacher Quality. Trained raters assessed the teacher work samples. Results supported the generalizability of RTWS ratings made using an analytic scoring rubric. Results revealed high dependability coefficients for panels of three or more trained and experienced raters. Support for the content representativeness of the RTWS was obtained using criteria suggested by Crocker (1997), including the frequency, criticality, necessity, and representativeness of the targeted teaching behaviors to actual teaching practice. Results also indicated direct correspondence between the targeted RTWS tasks and seven of the ten INTASC standards. Positive correlations between the RTWS performances and independent ratings of the quality of learning assessments indicated that teacher candidates who scored well on the RTWS provided better evidence of their impact on student learning than those who scored less well. The Quality of Learning Assessment Scale is appended. (Contains 11 tables and 26 references.) (SM)