Promoting Validity in the Assessment of English Learners

Across the globe, educational tests are being used at a rapidly increasing rate. Traditional uses of educational assessments include measuring students' achievement with respect to subject matter taught during a particular period in time, gauging students' academic strengths and weaknesses, and providing information to admissions officers and others who make decisions about admissions, scholarships, or eligibility to special programs. More recently, educational tests are being used to inform educational policy and for holding educators accountable for student learning. In fact, achievement testing for accountability purposes is one of the defining characteristics of 21st-century educational assessments. One reason educational assessments are used for these important purposes is that they are considered to provide reliable and objective information regarding students' achievement. The fact that these tests are standardized, meaning the content, test administration conditions, and scoring are uniform (consistent) across all test takers, supports this perception of objectivity. However, no test is perfectly suited for all students and so educational tests typically do the best they can for the majority of the population tested (Geisinger, 2000). For this reason, it is important to consider identifiable subgroups of the examinee population when considering the fairness and appropriateness of educational tests.