Intensive English programs (IEPs) strive to make certain that international students have sufficient levels of speaking ability, which is typically assessed through a combination of different tasks. One drawback of including multiple tasks is that the development, administration, and scoring might not be practical. Therefore, it is important to investigate how well the tasks account for examinees' speaking ability, as using fewer tasks could help in minimizing resources. Using quantitative methods of analysis, this study evaluates how well four types of speaking tasks on proficiency and achievement tests account for students' speaking ability in an IEP. The findings indicate that several tasks uniquely contribute to the speaking construct. This study has implications for the importance of balancing practicality with construct representativeness and presents a model of how IEPs might approach this issue.