The present study aimed to explore the relationship among vocabulary size, P honological A wareness (PA), and reading comprehension in English learners with low proficiency in Taiwan’s higher education. Forty-one university students who had taken the Test of English for International Communication (TOEIC) were recruited, 30 of whom were at a proficiency level much lower than B1 Threshold of the Common European Framework of Reference (CEFR) for Languages. Three PA subtests and a vocabulary size test were administered to all participants individually. Pearson’s correlations show that their TOEIC reading scores were correlated with the four measures when all 41 participants were included; however, among the 30 low-proficiency learners, their reading scores were correlated with Elision—one PA... measure—and vocabulary size only. When parallel regression analyses were computed against all participants and the low-proficiency subgroup, the four measures altogether explained nearly 64% of the variance in their TOEIC reading scores in the former but the explained variance dropped drastically to around 40% in the latter. Among the four measures, vocabulary size was the only significant predictor of reading ability and accounted for the largest variance. Meanwhile, phonological awareness explained additional variance in reading comprehension. While different PA measures did not seem to make a difference to the whole sample, Elision seemed to have explained more variance and served as a better task to assess phonological awareness of the low-proficiency subgroup.