ABSTRACT
This chapter describes validation processes used in the design, scoring, and reporting decisions of a newly developed summative English language proficiency assessment implemented in the nation’s largest English learner enrolling state. The authors posit that setting performance standards for an English language proficiency assessment should include validity checks and consider the multiple uses of the assessment’s performance levels. These include measuring and reporting on students’ English language development progress for state and federal accountability and setting policy on both the classification of students as English learners by English language proficiency level and their reclassification to fluent English proficient status, both of which impact school funding and students’ instruction. The validation study methods described and illustrated include a Contrasting Groups standard-setting approach and descriptive box plot, logistic regression, and decision consistency analyses. The results obtained allow for corroboration by triangulating multiple sources of evidence, including expert judgment and empirical data. These methods provide an example of a multistep process that can be used for initial and ongoing validation research, which is important as the population of English learners in United States K–12 public schools continues to grow.
