In recent years, many areas of instruction and evaluation have undergone reforms in response to changing theories of learning, teaching and testing. One such reform is that the era of teaching has changed into an era of testing in Taiwan. These days it is not enough to say our students have college-level proficiency in English reading, listening, speaking, and writing. With accountability required from the government as well as the public, colleges and universities will be asked to provide evidence of English proficiency and progress like never before. The need for measurable results to prove student-learning outcome turns out to be the importance of English proficiency testing, which is designed to assess students' general English ability outcomes to guarantee students' English ability for future employment and improve the quality of instruction and learning (the so-called washback effect). This study aims to examine and discuss Krashen's (1982) The Affective-Filter Hypothesis and the theories of Washback Effect to find a balanced way to promote students' learning outcome. The researchers strongly suggest except English proficiency tests there be a better way to demonstrate technology college students' learning outcomes, by overcoming students' affective filter in learning, designing reliable, valid, and helpful testing, and employing a positive washback effect of English proficiency testing in teaching.