This study explored the relationship among working memory (WM), speaking accuracy and length of utterance of Iranian Intermediate EFL learners. The data were collected from 38 female EFL learners whose age range was between 12 and 15 studying English at a language institute in Tehran. First, an Oxford Placement Test (OPT) was administrated to ensure the homogeneity of the participants and based on the results of the test thirty homogenous learners were selected as the main participants of the study. Next, a working memory capacity test developed by Daneman and Carpenter (1980) was administered to the participants. Later, the researcher administered a speaking test on a topic appropriate to the level of the participants which was taken from Top Notch 1 (Saslow & Ascher, 2011). Then, the researcher recorded their voices and transcribed them in order to calculate the number of lexical words the students could articulate based on a formula developed by Gilmore (2004). Speaking accuracy was also measured using a formula developed by Gilabert (2004); In fact, the students' performance was rated by two experienced teachers. Then, the Pearson correlation formula was utilized to analyze the obtained data. The results revealed a significant correlation between working memory capacity and speaking accuracy. Based on the findings, no significant correlation was shown between working memory and length of utterance. And finally, no significant correlation was depicted between length of utterance and speaking accuracy.
