ACCESS for ELLs^® Overview: Tiers

The solution to making the test appropriate to each individual was to present the test items in 3 tiers for each grade level: A, B, and C. The following chart shows how the different tiers map to the English language proficiency levels.



You can see from this test design that the tiers overlap, a necessity for making sure each tier is measuring to a common proficiency scale.

Each tier, of course, is only able to discriminate performance on its portion of the proficiency scale, so to make sure the whole ACCESS for ELLs^® test works as intended, it is necessary to place each student into the tier that best matches his or her English language proficiency level. The decision as to where the student currently falls on the scale is best made by the student's teachers, based on the information they have about the student's language proficiency, including performance on other language tests. The W-APT™ screener test yields a composite score which indicates which tier a child should be placed in for the ACCESS test, however, that score should be supported by additional criteria for tier selection.

The ACCESS for ELLs^® test battery is a collection of assessment instruments administered to all ELL students across all grades and all proficiencies. Each test form consists of a set of thematic folders, or parts, each of which contains three to six test items. The test is arranged in this way to give students a context for the items they are presented with and to minimize the cognitive leaps they have to make in transitioning from math items to language arts items to science items, and so on.

Each test instrument takes particular aim at a certain grade level cluster and range of proficiencies, but each also has to align with all the other instruments in the battery. That is, they each measure a certain segment of a common academic English proficiency measurement scale. To do this successfully, some of the items on any single test are shared among certain other tests. They are "carried over" between grade level clusters and tiers as complete thematic folders. Each Tier A form, with the exception of the grades 1-2 form, has two thematic folders from the grade level below it. Similarly, each Tier C form contains two thematic folders from the grade level above it. For example, the grade 3-5 Tier A test shares certain items with the 1-2 Tier C and 3-5 Tier B test forms. Likewise, the 6-8 Tier C form borrows items from 6-8 Tier B and 9-12 Tier A. It might be argued that asking a child to respond to questions for which he or she has no preparation on one hand or for which he or she is overqualified makes those items unfair indicators.

However, in fact, this configuration provides just the kind of confirmatory evidence needed for equating all the forms to the common measurement scale. In this way, we can better assure that as a child progresses through the grades and in English proficiency, we get an accurate picture of his or her real advances from year to year.

Continue reading: Origins of WIDA & Development of the Field Test