How to Write STAAR-Style Questions for Your Own Unit Tests

You finish a unit on the American Revolution. Your test has 20 multiple choice questions and students score in the 70s and 80s. Two weeks later, you give a practice test with STAAR-style questions covering the same material and the scores drop 15 points. The content is the same. The format is the difference — and your unit test never prepared them for it.

Writing STAAR-style questions for your own unit tests isn't about teaching to the test. It's about making sure the assessments you give all year are compatible with the question format students will face in May. Every unit test you give is a missed practice opportunity if the questions don't match how STAAR actually asks things.

Why Standard Unit Test Questions Aren't Enough

The typical teacher-written multiple choice question looks like this:

What was the main cause of the American Revolution?
A) Taxation without representation
B) The French and Indian War
C) Slavery
D) The Emancipation Proclamation

One answer is clearly correct. One is a related-but-wrong event from the right era. Two are obviously wrong. Students don't need to reason through this — they eliminate the obvious wrong answers and pick the most familiar option. That's not what STAAR asks them to do.

A STAAR-style question on the same content might embed a primary source excerpt from the Stamp Act debates, ask students to identify which constitutional principle the colonists were applying, and offer four choices that are all plausible but wrong in different ways. The question rewards understanding, not recognition. Those are two different skills, and only one of them is tested in May.

Action step: Pull your last unit test and count how many questions have two obviously wrong answer choices — answers a student could eliminate without knowing anything about the content. Any question with two clearly incorrect options doesn't match STAAR complexity. Flag those questions first. That's your highest-leverage edit.

What Makes a Question "STAAR-Style"

STAAR questions have specific structural features that distinguish them from typical teacher-written questions:

• Stimulus material: Most STAAR questions in RLA, social studies, and science embed a text excerpt, graph, map, data table, or image that students must interpret. The question doesn't stand alone — it requires students to process information before answering.
• Plausible distractors: Every wrong answer should be the kind of mistake a student who partially understands the content would make. No obviously wrong answers. If a student could eliminate a choice without reading the stimulus or thinking about the content, it's a weak distractor.
• Higher-order stems: STAAR uses "which best explains," "what can the reader conclude," "how does X affect Y," "which evidence best supports" — not "what is," "who was," or "when did."
• Application to a specific context: Many STAAR questions ask students to apply knowledge to a specific scenario — "which of the following best demonstrates" — rather than asking for the knowledge itself in the abstract.

Action step: Take one section of your next unit test and rewrite every stem that starts with "What is" or "Which of the following is" to start with "Which best explains" or "What can be concluded about" or "How did X affect Y." This single change increases the cognitive demand without requiring you to rewrite the entire test.

Writing Stems That Match STAAR Complexity

The question stem is the most important factor in whether a question matches STAAR complexity. Weak stems test recognition. Strong stems test reasoning.

Weak stems (avoid these):

• What is the definition of ___?
• Who was the first ___?
• What year did ___ happen?
• Which of the following is an example of ___?

Strong stems (use these):

• Which statement best explains why ___?
• How did ___ contribute to ___?
• What can be concluded about ___ based on [the passage / the data / the map]?
• Which evidence from [the text / the chart] best supports the claim that ___?
• How does the author's use of ___ develop the [theme / argument / central idea]?

The strong stems require students to do something with the content — explain, compare, conclude, evaluate. The weak stems require retrieval of a fact. STAAR is almost entirely strong stems. If your unit tests are mostly weak stems, your students are practicing the wrong cognitive skill all year and encountering the right one cold in May.

Action step: Write your next unit test with a goal of at least 70% strong stems. After grading it, compare the class average to your previous test. If the average drops, that's important information: students know the content but can't apply it yet, and you still have time to build that skill before STAAR. A lower score on a better test is better data than a higher score on an easier one.

Choosing Wrong Answers That Diagnose Misconceptions

This is the part of question-writing most teachers don't think about — and it's where the real instructional value of STAAR-style unit tests comes from. Every wrong answer choice should represent a specific, predictable misconception or error that a student might make.

If you know your students tend to confuse the causes of World War I with the causes of World War II, one of your distractors should use WWI causes in a WWII question. Not to trick students — to diagnose them. Getting that question wrong tells you exactly what to reteach. Wrong answers that are random or obviously wrong give you no diagnostic information at all.

When writing distractors, ask yourself:

• What would a student who partially understands this concept choose?
• What would a student who confused this with a related concept choose?
• What would a student who answered the intermediate question instead of the final question choose?
• What would a student who misread a key word in the stem choose?

One distractor per question for each of those patterns gives you four choices that actively diagnose what went wrong when a student answers incorrectly.

Action step: After your next test, sort wrong answers by which distractor students chose most often. If 40% of your class chose the same wrong answer, that distractor is telling you something specific about a shared misconception. Reteach that misconception directly — don't just show students the correct answer and move on.

How Many STAAR-Style Questions You Need Per Test

You don't need to rewrite your entire assessment practice to get the benefit of STAAR-style questioning. The 70/30 rule works well in practice: aim for 70% of your unit test questions to be STAAR-style (strong stem, plausible distractors, stimulus where appropriate) and leave 30% as straightforward recall questions. The recall questions give students confidence and let you check basic content knowledge. The STAAR-style questions build application skill and generate useful diagnostic data.

For classes that are significantly behind grade level, start at 50/50 and move toward 70/30 over the semester. Students who haven't been exposed to STAAR-style questions at all need a transition period — going straight to 70% STAAR-style can damage confidence before it builds skill, and a demoralized class is harder to move than a confused one.

Action step: Review your most recent unit test and calculate the actual ratio of STAAR-style vs. recall questions. If you're below 50%, pick the five questions that would be easiest to upgrade and rewrite those before your next test. Don't try to overhaul everything at once — progressive improvement over the semester compounds into a meaningfully different assessment practice by spring.

Getting Started Without Starting from Scratch

The most efficient way to start writing STAAR-style questions is to work backward from released STAAR items. Find two or three released questions on the content you're about to test, study the question structure carefully, and use it as a template for your own questions with your own content.

You're not copying questions — you're learning how STAAR engineers complexity and applying that to the specific concepts you just taught. A released Grade 7 science question about ecosystems teaches you the format; you write a question about your current unit using the same structural moves. After you've done this with five or six released items, the pattern becomes intuitive and you won't need the template anymore.

If you want pre-built STAAR-aligned items you can use or modify as models, the TestPrepGrow content library has questions organized by grade, subject, and TEKS. Using them alongside your unit tests isn't test prep — it's building an assessment practice that prepares students for the format all year so they're not encountering it for the first time in May.

The teachers who see the smallest gap between unit test scores and STAAR scores aren't the ones doing the most test prep in April. They're the ones who've been asking STAAR-quality questions since September.