The present study investigated a washback effect regarding vocabulary test format and students' learning outcomes. Ninety-three Korean university students in English reading courses were given a vocabulary pre-test and pre-treatment survey. Then they were divided into three groups. A control group was taught how to write a summary in English and a receptive vocabulary knowledge group was taught how to study vocabulary, then both of these groups were instructed that they would have a receptive vocabulary knowledge test. A productive vocabulary knowledge group was taught how to study vocabulary and told that they would have a productive vocabulary knowledge test. After 15 weeks, they all were given a post-vocabulary test consisting of both receptive and productive vocabulary knowledge items, and post-treatment survey. Results indicated that the productive vocabulary knowledge group showed the best outcomes overall. Its students earned significantly higher scores on the productive vocabulary knowledge post-test than the other groups. Regarding the receptive vocabulary knowledge post-test, it featured scores similar to the receptive vocabulary knowledge group, which had the highest scores. Moreover, the productive vocabulary knowledge group's pre- and post-treatment survey data revealed that its students had improved their ability to use the vocabulary learning strategies that were taught.