What Is the New Multi-Stage Adaptive Testing in TOEFL 2026 and Will It Be Harder?

What Is the New Multi-Stage Adaptive Testing in TOEFL 2026 and Will It Be Harder?

When ETS announced that the TOEFL iBT 2026 would use Multi-Stage Adaptive Testing, many test-takers had the same reaction: "Does adaptive mean harder?" The short answer is no. It means different, and in most ways, it means fairer. But understanding how MST works is essential for effective preparation, because it changes the test-taking experience in ways that matter for your strategy, your pacing, and your mindset.

This article explains exactly what MST is, how it works in the TOEFL iBT 2026, and what it means for your preparation.

What Is Adaptive Testing?

Adaptive testing is a method where the test adjusts its difficulty based on your performance. Instead of giving every test-taker the same questions, the system selects questions that are most appropriate for your demonstrated ability level.

The concept is not new. The GRE and GMAT have used adaptive testing for years. Medical licensing exams use it. The principle behind all of them is the same: you learn more about a person's ability by giving them questions that are neither too easy nor too hard for their level.

Think of it this way. If you are learning to play tennis, you do not improve by playing against a 5-year-old or against a professional. You improve by playing against someone close to your level. Adaptive testing applies the same logic to measurement: the test "plays" at your level so it can measure you more precisely.

How MST Works in the TOEFL iBT 2026

The TOEFL iBT 2026 uses a specific type of adaptive testing called Multi-Stage Testing. Here is the structure:

Stage 1: Module 1 (Medium Difficulty)

Every test-taker starts with the same Module 1, which contains questions at a medium difficulty level. This is the baseline measurement. Your performance on these questions gives the system its first estimate of your ability.

You answer all the questions in Module 1 before moving on. You cannot skip ahead or go back to previous questions within the module.

The Routing Decision

After you complete Module 1, the system uses Item Response Theory (IRT) to calculate an ability estimate. IRT is a statistical model that considers not just how many questions you got right, but the difficulty and discrimination power of each question you answered.

Based on this estimate, you are routed to one of two paths:

  • Module 2 Easy: If your Module 1 performance suggests a lower ability level.
  • Module 2 Hard: If your Module 1 performance suggests a higher ability level.

This routing happens automatically and invisibly. You will not see a message saying "You have been routed to the easy module." You simply move to the next set of questions.

Stage 2: Module 2 (Easy or Hard)

Module 2 contains questions calibrated for your estimated level. If you were routed to Module 2 Hard, you will face more challenging questions. If you were routed to Module 2 Easy, you will face more accessible questions.

Your final score for the section is calculated using your performance across both modules, weighted by the difficulty of the questions you received.

How IRT Scoring Differs from Simple Counting

Under the old linear TOEFL, scoring was relatively straightforward: your raw score (number of correct answers) was converted to a scaled score using a conversion table. Everyone answered the same questions, so a simple conversion worked.

With MST and IRT, the scoring is more sophisticated:

The Key Principle

A correct answer on a hard question is worth more than a correct answer on an easy question.

This is the fundamental difference. Two test-takers who both get 15 out of 20 questions correct may receive different scores if one answered harder questions than the other.

How IRT Estimates Ability

IRT assigns each question a set of parameters:

  • Difficulty: How hard the question is (expressed as an ability level where a test-taker has a 50% chance of answering correctly).
  • Discrimination: How well the question distinguishes between test-takers of different ability levels.

When you answer a question correctly, your ability estimate goes up. When you answer incorrectly, it goes down. But the size of the adjustment depends on the question parameters. Getting a very hard question right produces a large upward adjustment. Getting a very easy question wrong produces a large downward adjustment.

After all questions are scored, the system produces a final ability estimate with a confidence interval. This estimate is then mapped to the 0-30 score scale.

Practical Implications

  1. You cannot count correct answers to predict your score. The difficulty of the questions matters as much as your accuracy.
  2. Getting routed to Module 2 Hard is not bad news. It means the system thinks you are strong, and answering those harder questions correctly will be rewarded with higher scores.
  3. Getting routed to Module 2 Easy does not cap your score at a low level. You can still score well if you answer the Module 2 Easy questions correctly, though the ceiling is lower than Module 2 Hard.
  4. Every question matters. There is no strategy of "sacrificing" easy questions to save time for hard ones. Each answer updates your ability estimate.

Is MST Harder Than the Old Format?

This is the most common question, and the answer requires some nuance.

It Is Not Harder in Absolute Terms

The overall difficulty of the test is calibrated to produce the same score distribution as before. A student who would have scored 100 on the old TOEFL should score approximately 100 on the new one. ETS performs extensive equating studies to ensure this.

It Feels Different

What changes is the subjective experience:

  • If you are strong: Module 1 may feel manageable, and then Module 2 Hard will feel genuinely challenging. Under the old format, you would have breezed through easy questions mixed in with hard ones. Now the hard questions are concentrated in Module 2.
  • If you are developing: Module 1 may feel challenging, and Module 2 Easy will feel more appropriate for your level. Under the old format, you would have faced many questions that were too hard, leading to frustration and random guessing.
  • If you are in the middle: The experience is closest to the old format, since Module 1 is medium difficulty and Module 2 could go either way.

The Key Insight: Difficulty Adjusts to You

The test is not harder or easier in a fixed sense. It adjusts to you. A test that adapts to your level will always feel appropriately challenging, which is actually what you want. Questions that are too easy do not help your score. Questions that are far too hard lead to guessing, which does not help your score either. MST gives you questions in your productive zone.

How MST Applies to Each Section

MST is applied independently to each section of the TOEFL iBT 2026:

Reading

Module 1 contains a mix of reading passages (daily-life and academic) at medium difficulty. Based on your Module 1 performance, you are routed to Module 2 Easy or Module 2 Hard for reading. The harder module features more complex passages, more nuanced questions, and trickier vocabulary-in-context items.

Listening

The same structure applies. Module 1 contains listening items (Choose a Response, Conversations, Announcements, Academic Talks) at medium difficulty. Module 2 adapts based on your listening performance. Harder listening modules feature faster speech rates, more complex academic topics, and questions requiring deeper inference.

Speaking

The speaking section also follows the two-module structure. The difficulty adjustment may involve more complex interview topics, more abstract questions, or scenarios requiring more sophisticated language.

Writing

Writing follows the same pattern. Module 2 Hard may feature more complex email scenarios, more nuanced academic discussion prompts, or topics requiring more sophisticated argumentation.

How MST Affects Your Preparation Strategy

Understanding MST should change how you prepare in several specific ways:

1. Module 1 Performance Is Critical

Your Module 1 performance determines your routing. If you underperform in Module 1 due to nervousness, fatigue, or carelessness, you may be routed to Module 2 Easy, which limits your scoring ceiling for that section.

Preparation implication: Practice starting strong. Simulate test conditions and focus on performing well from the first question. Warm up before the test.

2. Do Not Panic if Module 2 Feels Hard

If you are routed to Module 2 Hard, the questions are supposed to feel harder. This is not a sign that you are failing. It is a sign that the system recognized your strong Module 1 performance.

Preparation implication: Practice with difficult materials so that encountering hard questions does not trigger anxiety. Build the mental habit of thinking "hard questions mean I did well in Module 1."

3. Every Question Counts Equally for Routing

In Module 1, there is no strategic benefit to spending disproportionate time on certain questions. Every correct answer contributes to your ability estimate, and every incorrect answer reduces it.

Preparation implication: Practice consistent pacing. Do not rush through some questions to spend extra time on others. Develop a steady rhythm.

4. You Cannot Game the Routing

Some test-takers wonder if they should deliberately underperform in Module 1 to get easier questions in Module 2. This does not work. The IRT scoring model accounts for question difficulty, so answering easy questions correctly yields a lower score than answering hard questions correctly. You always benefit from doing your genuine best.

Preparation implication: Always try your hardest on every question. There is no strategic sandbagging.

5. Practice With Adaptive Tests

Practicing with linear tests (where every question is the same difficulty) does not prepare you for the MST experience. You need to practice with a system that actually routes you to different modules based on your performance.

Preparation implication: Use practice platforms that replicate the MST format, including the Module 1 to Module 2 routing decision.

Common Myths About MST

Myth: "If I get routed to Module 2 Easy, my score is capped."

Reality: Your score ceiling is lower with Module 2 Easy, but you can still achieve a respectable score. The system is designed so that a strong performance on Module 2 Easy still produces a fair score. However, the highest scores do require Module 2 Hard.

Myth: "The computer decides my score before I finish."

Reality: The routing after Module 1 is just a branching point. Your final score is calculated from your performance across all questions in both modules. Module 2 performance matters just as much as Module 1.

Myth: "Adaptive testing is unfair because different people get different questions."

Reality: IRT scoring accounts for question difficulty. A test-taker who correctly answers 8 hard questions and a test-taker who correctly answers 10 easy questions may receive similar scores, because the hard questions provide more information per correct answer. The system is designed to produce equivalent scores regardless of which path you took.

Myth: "I should guess randomly on Module 1 to get easier Module 2 questions."

Reality: This is the worst possible strategy. You would be routed to Module 2 Easy, and even if you answered every Module 2 Easy question correctly, your score would be limited because the system knows those questions were easy. You maximized your score by trying your best on every question.

How Ace120 Replicates MST

Most TOEFL practice platforms still use the old linear format. They give you a fixed set of questions regardless of your performance. This means you are not practicing the actual test experience.

Ace120 implements the full MST adaptive testing system:

  • Module 1 presents questions at medium difficulty across all sections.
  • IRT-based ability estimation calculates your routing after Module 1.
  • Module 2 Easy or Module 2 Hard is selected based on your actual performance.
  • IRT scoring produces your final section score, accounting for question difficulty.

This means your practice tests give you an accurate preview of what test day will feel like, including the experience of being routed to harder or easier questions and understanding how that affects your score.

The Bottom Line

MST is not harder. It is smarter. The test adapts to your level so it can measure you more precisely. Strong test-takers will face harder questions (and be rewarded for answering them correctly). Developing test-takers will face more appropriate questions (and get a fairer measurement of their actual ability).

Your preparation strategy should focus on three things: start strong (Module 1 matters), stay calm if questions get harder (it means you did well), and practice with a system that actually uses adaptive routing.

Practice With Real Adaptive Testing

Ace120 is the only practice platform that replicates the full TOEFL iBT 2026 MST experience with IRT-based routing and scoring. Stop guessing what adaptive testing feels like and start experiencing it. Take your first adaptive practice test at Ace120.