This study examined the extent to which ethnic-based differences exist in teacher ratings of African American students and White students on the language and literacy domain of a curriculum-embedded performance assessment for students in grade 1. It extended previous research on performance assessments to focus on issues related to equity in teacher ratings on a curriculum-embedded performance assessment of young children in an urban school district. The major findings of this study are discussed in terms of the: (1) consistency in teacher ratings, (2) disparity in teacher ratings of African American and White students, and (3) absence of indicator bias on the language and literacy domain. The first major finding of the study indicated that across indicators on the language and literacy domain, teachers demonstrated a high level of consistency in rating students' performance. The second major study finding was the discrepancy in teacher ratings between African American and White students. Across indicators, a greater proportion of African American students received ratings in the "not yet" category compared to White students. Furthermore, a smaller proportion of African American students received ratings in the "proficient" category compared to White students. The final major finding of this study was the absence of differential item functioning (DIF) in teacher ratings on the indicators. The results showed that the language and literacy domain score was the major contributor to teachers' predicted ratings, accounting for between 70% and 86% of the total variation in teacher ratings. The findings in this study are to be viewed with caution in terms of generalizability. This study explored ethnic-based differences in a particular state-wide assessment program in a specific urban school district. The curriculum standards exhibited in this state's assessment program are not necessarily reflected in another state's assessment program. Furthermore, this study's focus was a single school district's performance on a state-wide assessment program. The lack of representation of first-grade students from across the state makes the findings of this study not representative of the diverse geographical and economic make up of the state. Implications of this study are discussed. (Contains 3 tables.)