The Best Reading Test Prep Is Better Reading Instruction

Every spring, something strange happens in schools. As the reading test gets closer, students sometimes spend less time doing the very thing the test is supposed to measure: reading.

Books give way to passages. Discussions give way to questions. Knowledge building gives way to test-taking strategies. Teachers start talking about distractors, stems, constructed responses, and finding evidence. Students practice main idea on Monday, inference on Tuesday, author's purpose on Wednesday, and by Thursday everyone is circling key words and eliminating obviously wrong answers.

The logic makes sense. If students are going to take a reading test, we should prepare them for the reading test; but, what if too much test preparation actually slows the development of the thing we are trying to improve?

That claim requires some caution. We do not have a neat experimental study showing that ten weeks of test preparation causes students to lose a particular number of months of reading growth. However, we do have a substantial body of research and theory that should make us question what happens when test preparation begins replacing reading instruction. The concern is not that students should never see the test format. They should. The concern is what we give up when practicing the test becomes the curriculum.

The Opportunity Cost of Test Prep

There is a simple concept from economics that matters enormously here called opportunity cost. Time spent doing one thing is time that cannot be spent doing something else. In reading instruction, that matters because students need an enormous amount of experience with language and text. They need to read. They need to listen to complex texts. They need to develop fluency. They need to learn words, acquire knowledge, discuss ideas, write about what they have read, and make sense of increasingly complex texts.

The fact is that test preparation takes time from somewhere.

Research on high-stakes testing has documented this problem for years. Au's (2007) metasynthesis of 49 qualitative studies found that high-stakes testing frequently narrowed curriculum, fragmented knowledge into test-related pieces, and increased teacher-centered instruction. Berliner (2011) similarly argued that curriculum narrowing can have long-term consequences because students lose opportunities to acquire the knowledge they will need for later learning.

In other words, test preparation has an opportunity cost, and in reading, the opportunity cost may be reading. There is another problem. Reading comprehension is not a collection of question types.

Comprehension Isn't a Muscle You Isolate

A student does not become a better comprehender simply by answering more "main idea" questions. Main idea is not a muscle that becomes stronger every time we exercise it with another four-paragraph passage. The same is true of inference, author's purpose, theme, and many of the other labels that appear on reading assessments.

Comprehension is the product of a remarkably complicated system. Readers have to recognize words, understand language, know the meanings of words, connect ideas across sentences, make appropriate inferences, monitor their understanding, understand how texts work, and bring relevant knowledge to what they are reading. Motivation and engagement matter too. A child who knows a great deal about a topic can often understand a challenging text about that topic better than a child who knows very little about it.

Yet test preparation can accidentally teach a very different model of reading:

Davis and Vehabovic (2017) called attention to exactly this problem in The Reading Teacher. They described the dangers of "test-centric literacy instruction," including making the test itself the text, allowing item types to dictate instruction, requiring annotation routines regardless of whether they actually support comprehension, and replacing research-supported comprehension instruction with practices designed primarily around the assessment.

Test-Taking Skills Aren't Reading Development

Students can get better at those routines, but that does not necessarily mean they are becoming substantially better readers. This is partly a problem of transfer. Learning tends to transfer most readily when the new task resembles the task that was practiced. If students repeatedly practice a particular test format, they may become more efficient at navigating that format. That is useful. They may learn what certain questions are asking. They may become more comfortable with the testing interface. They may learn to manage their time or avoid common mistakes.

Those are test-taking skills. They are not the same thing as reading development.

This distinction is especially important because a reading test samples reading. It cannot possibly measure everything a student knows or every text a student might encounter. Instead, an assessment selects a relatively small collection of passages and questions and uses performance on those items to estimate a much larger construct: reading ability.

If instruction becomes increasingly focused on the predictable features of that sample, we risk teaching students to demonstrate reading on a particular assessment rather than developing the broader capacities the assessment is supposed to represent.

The Bigger Concern: Knowledge

There is an even bigger concern: knowledge. Reading development is cumulative. Reading helps students learn words and ideas. Those words and ideas help them understand what they read next. That new reading gives them even more knowledge, which makes still more texts accessible. Over years, those differences compound.

Imagine two classrooms.

Both classrooms are technically "doing reading." But they are not necessarily building the same readers.

The first classroom is accumulating intellectual capital. Every new word, concept, text structure, historical event, scientific principle, and literary experience becomes something students can bring to the next text they encounter. The second classroom may be accumulating experience with reading tests.

That difference matters because background knowledge is not an optional extra added after students learn how to comprehend. Knowledge is part of comprehension itself. When students know more about the world, they have more resources available for understanding what texts leave unstated, resolving ambiguity, connecting ideas, and making inferences.

This creates one of the great ironies of test preparation. In our effort to raise reading scores quickly, we may reduce time spent developing some of the things that would raise reading achievement over the long term.

Familiarization Is Not Curriculum

None of this means students should walk into an important assessment having never seen its format. That would be ridiculous. Test familiarization is sensible. Students should know how the assessment works. They should understand the kinds of responses they will be asked to provide. If the test is digital, they should know how to navigate it. They should have opportunities to practice so that unfamiliar procedures do not interfere with their ability to demonstrate what they know.

Showing students how the test works is preparation. Organizing months of reading instruction around the test is curriculum.

That distinction is important. If we want students to perform better on reading assessments, the most powerful long-term strategy may also be the least exciting test-preparation strategy imaginable: develop better readers.

Teach children to decode accurately and efficiently. Develop fluency so that word reading requires less attention. Build oral language. Teach vocabulary. Build knowledge systematically. Read aloud texts students cannot yet access independently. Give students lots of opportunities to read texts they can access. Teach comprehension strategies when those strategies help students make meaning. Talk about books. Write about ideas. Read science. Read history. Read stories. Read poems. Read things worth knowing.

Then, when the test approaches, show students how the test works. Because the goal of reading instruction was never to create children who are really good at answering reading questions. The goal is to create readers. And, somewhat inconveniently, creating better readers may still be the best test preparation we have.