Examining the Accuracy of a Conversation‐Based Assessment in Interpreting English Learners' Written Responses