As computer science (CS) becomes integral to K–12 education, it is important to examine how foundational skills such as reading influence student engagement and performance in introductory programming. This study investigates the relationship between reading proficiency and CS learning outcomes in an urban high school in the Northeastern United States. Grounded in Cognitive Load Theory (CLT) and the Lexile Framework for Reading, the research also explores how the readability of CS assessment items contributes to cognitive demands, particularly for students reading below grade level. A mixed-methods design was employed using i-Ready diagnostic data, Lexile-coded CS assessments, and student interviews. Assessment questions were assigned to Lexile levels to capture textual complexity, and student performance was analyzed through correlation and regression modeling. Results show a weak but positive correlation between i-Ready scores and CS outcomes. Performance declined as item Lexile levels increased, especially for students in lower proficiency bands. Visual analyses indicated that higher-proficiency readers performed consistently, while struggling readers showed evidence of increased cognitive load on reading-demanding tasks. Thematic analysis of student reflections revealed patterns of intrinsic and extraneous load during code tracing activities. These findings highlight the importance of considering readability in CS assessment and instructional design to promote equity in computing education.