This exploratory study investigated how reading comprehension was conceptualized on the new high-stakes test, the 2011-2012 State of Texas Assessment of Academic Readiness (STAAR). Specifically, comprehension, rate, and accuracy scores on the Gray Oral Reading Test 4 (GORT-4) from a group of struggling, low-SES, Hispanic middle school students (n = 59) were set as predictor variables to examine possible relationships with the STAAR. Initial bivariate correlations showed a weak relationship between GORT-4 predictor variables (comprehension, rate, accuracy) and STAAR ELA scores. Moreover, the overall regression model was not a good fit, with the linear combination of the GORT-4 components of comprehension, rate, and accuracy accounting for only 3.5 % of the variance in STAAR scores. The weak relationship between STAAR test results and the GORT-4 is examined in light of the current research on high-stakes testing, particularly for the at-risk population studied.