The Problematic Nature of Standardized Tests (Part 2)

In Part 1 of this series I mention the various biases within standardized tests. For Part 2, I want to dive into the questions themselves.

Many of these standardized tests include questions that are highly impacted by other factors impacting the child. For example, if the child has difficulty producing the /s/ sound, and is being asked to produce plural words on an expressive language test, they may score poorly on that test item/question because they have a speech sound impairment that is impacting them. So the question must be asked, does that child actually have difficulty with this grammar marker or is the child actually impacted by something unrelated.

Moreover, let’s say you are administering a questionnaire to a caregiver such as the Receptive-Expressive Emergent Language Test (REEL-4). These questions often give way to reporter bias as the questions themselves may be confusing. In my practice as an Early On provider, this was the primary test we would use to determine the child’s eligibility for services. Oftentimes, I would ask a test item and the caregiver would be confused on the wording. As a result, I would provide examples as to what I felt the test item meant. However, I learned from my colleagues that they interpreted the test item differently resulting in different scoring. Additionally, after revealing the results of the testing at a later date, I would have caregivers either report that they did not understand the meaning of the question or realized that their child was actually performing it, but did not realize it at the time.

All this is to say that there is a large amount of bias and confusion when it comes the various test items on a standardized test. Therefore, it is up to us as Speech-Language Pathologists to determine appropriate measures to assess the child and to be sure that when we are evaluating them, that child is not being impacted by something else that is resulting in a lower score.