On this page:
Dr. Shona explains that, even though it’s better that test questions are now written so that a reader cannot fully succeed without understanding the passage as a whole, we shouldn’t be asking students to read these questions first since we are essentially asking them to fill up their memories before they ever get started reading.
The pedigree, or even genealogy, of Texas assessments reveals that some of the strategies that worked previously are still hanging around, like how Texas STAAR test questions are still being handled. And that’s a problem because what we want is a true understanding of a text and the ability to do something important with it. Let me explain.
Psychometricians: These folks study tests and how the population responds. When the tests first came out, teachers didn’t know what to expect. But, over time, they figured out what the tests valued and developed strategies to help students “hack” the test and pass the exam. One such strategy was to look at summary question choices. Students marked the answer choices for BME: Beginning, Middle, and End. That worked because one answer had a summary from all three parts of the text; one had only pieces; one was a detail; and another was just silly or irrelevant. Students didn’t even have to read the text to figure out what the best summary was. And that’s not the point of the assessment.
Ultimately, Texas (and teachers) want to know if readers understand what they read. The strategy was a bypass for meaning and didn’t tell us what we needed to know. As a side note, a summary is more than a sentence or statement that contains ideas from the beginning, middle, and end of the text. A summary subsumes text structure and overall gist and purpose. BME isn’t sufficient.
Another strategy was to read the questions first. Then, readers could skim and scan to look for keywords that matched the questions. Test-takers could answer the questions without reading the text at all. Again, psychometricians could tell that the goal of measuring comprehension was not met. I could tell you about other strategies we used to bypass meaning and help kids pass. But the main point is that TAAS was used between 1991 and 2003. TAKS was used between 2003 and 2013. And, we’ve had new standards. So, those strategies are at least 33 years old, developed for less-rigorous exams, and for standards that no longer exist.
The Partial Problem With How Texas STAAR Test Questions Are Handled Now
The psychometricians today purposefully write questions so that a reader cannot fully succeed without understanding the passage as a whole. Which isn’t a bad thing. We want Texans to understand what they read and make informed decisions from that understanding. We don’t want Texans who think they know it all from a cursory glance at some questions and data.
Furthermore, when you add what we know about working memory and cognitive demand, we realize that, by asking students to read the questions first, we are asking them to fill up their memories before they ever get started reading.
Yes, it is true: the questions used to be written to where you could tell what you were looking for in the stem. It was like an addition problem. (My spellcheck thinks the word should be “addiction.” I laughed to myself. Seems like folks are addicted to telling kids to read questions first.) Now, you have to read the questions as a complete unit as well. Each answer choice must be considered in concert with the stem and preponderance of evidence from the text. Then, the learner must think and reason logically to prove or refute each choice. It’s more like converting units, adding, and then doing long division to find an answer. Questions are more complicated now in terms of cognition.
The old ways need to go. The new ways need to return to what we all want: true understanding of a text and the ability to do something important with it.
Caveat: So, my friend Liza does have her kids read the Texas STAAR test questions first. But, she uses it as a prereading strategy to collect information about the topic, genre, and content of the text. Then, her readers can use the information from the questions as background information to provide schema and purposes for reading that prepare them for the reading act.
Again, this is a completely different purpose for reading. One that is tied to how we comprehend and use texts in the real world. We have questions we need to answer. We research and find what we need to know.
Reading the questions first still bothers me because I worry that kids won’t understand why and how it helps them. I worry that their memories will become overwhelmed before beginning. I’m worried that folks will tell learners to read the Texas STAAR test questions first as a form of compliance and steps they are supposed to use rather than choices the learners purposefully make to guide their comprehension. If those things don’t happen, I guess I’m ok with it as a strategy chosen specifically by the learners because it works for their individual needs.
TEKS Commentary
Foundational Language Skills: Fluency: A
Foundational Language Skills: Self-Sustained Reading: A
Comprehension Skills: A, B, C, H, I
Multiple Genres: Literary Genres: A, C
Multiple Genres: Genres: A, B, C, D, E, F
Research and Inquiry: A, B, C, D
Up Next
Now that we have talked about some of the problems with Texas STAAR test questions, we’ll tackle introducing new text features and scroll bar characteristics.
About the Author
