How and how not to prepare students for the new tests

Shanahan, T. (2014). How and how not to prepare students for the new tests. The Reading Teacher, 68(3), 184-188.


A few years ago I sat in a faculty meeting where my colleagues and I were told that we needed to build “a culture of assessment” at our university. That phrase bothered me. I am not anti-assessment; as a teacher educator, I believe that well-designed, varied, and responsibly implemented assessments can tell us a lot about students and what we can do to help them learn. However, that whole “culture of assessment” thing seemed backwards. Shouldn’t we be building a “culture of learning”? Shouldn’t assessments be tools to help us build learning rather than being the focus and center of the whole teaching and learning process?

The example that opens Timothy Shanahan’s essay on the uses and misuses of reading assessments sounds as if Shanahan had encountered one of those administrators who have bought into the “culture of assessment” concept, also known as “data-driven assessment” (p. 184). Shanahan describes an interaction with a school district curriculum director who was interested in raising test scores by aligning test items with the reading curriculum in her district. Students would have actual instruction and practice in answering typical test items. That approach would perhaps raise the number of correct answers on some kinds of test items, but would those test scores tell us that students are comprehending what they read? Shanahan would say no. Here, he takes us through a discussion of how complex the reading process is, and how tests are limited in how well they can possibly capture that process.

The curriculum director in Shanahan’s example thought that she could simply outline the subskills of reading comprehension and then align test items with those. Shanahan points out that reading just does not work that way. Readers are employing multiple skills simultaneously when they work with texts. There also are other factors in play, particularly the difficulty and complexity of the actual texts being read. Anyone who has worked with readers, and in fact, anyone who is a reader her/himself, must realize that the choice of text often makes a big difference. There may even be situational factors that affect how well a reader comprehends a given text on a given day. It’s just not so simple that we can line reading subskills with reading test items and then teach to the test as this curriculum director was proposing to do.

In short, when it comes to reading comprehension, text, and context, are everything. Shanahan’s account is a cautionary tale about what can happen when we focus on assessments that are helpful but limited tools rather than focusing on actual reading and writing and having students do a lot of those two things. Shanahan’s article ends with recommendations that we engage students in a lot of real reading, writing, and talking about texts. Students need to develop stamina to read progressively longer and more challenging texts. We need to engage students in practicing reading skills, and according to Shanahan, the test scores will follow.