Not on broad anxiety scales, and yes on specific situations. The largest randomized trial to date found no statistically significant change in anxiety, depression, or parental stress, while individual socio-emotional goals did improve significantly. That distinction should change how you word your IEP goal and what you tell a parent. In a 2024 Emoquest survey of 16 practitioners and parents, 94% spent 30 or more minutes writing one story, so it is worth aiming it correctly.
What did the largest trial actually find?
Situational goals moved. Global anxiety did not. The 2025 ASSSIST-2 pragmatic cluster randomized trial randomized 87 UK primary and special schools covering 249 autistic students aged 4 to 11. Interventionists were existing school staff, not researchers, which makes the setting close to yours.
| Outcome measured at 6 months | Result | Statistically significant? |
|---|---|---|
| Social Responsiveness Scale-2, whole group | 1.61 point reduction (95% CI -4.18 to 0.96, p = .220) | No |
| Social Responsiveness Scale-2, students with 6 or more sessions | 3.37 point reduction (95% CI -6.65 to -0.10, p = .043) | Yes |
| Individual socio-emotional goal achievement | Met more often than usual care | Yes |
| Anxiety and depression (parent-reported scale) | No difference from usual care | No |
| Parental stress and general health | No difference from usual care | No |
The authors concluded that social stories are a low-cost, low-burden intervention whose benefit shows up in specific socio-emotional goals rather than in global measures. That is not a disappointing result. It is a precise one.
Why does a story calm the fire drill but not the anxiety score?
Because the instrument and the intervention are aimed at different targets. A social story describes one situation. A scale like the Revised Children's Anxiety and Depression Scale samples worry across many contexts the story never mentioned. A student can walk into the cafeteria without freezing and still score identically, because 60 of the 65 questions were about something else.
This is a measurement mismatch, not evidence that nothing happened. It is the same reason the trial's goal-based outcome, which is written per student, detected a change that the standardized scale missed.
From the 2024 Emoquest survey (n=16 parents, school SLPs, OTs, and SPED teachers): 94% spend 30 or more minutes making a single social story, and 25% spend one to two hours. If you are going to spend an hour, spend it on the one situation you can name and observe, not on a general story about feelings.
How should you word the IEP goal?
Write the observable behavior, not the internal state. Anxiety is not measurable from across a classroom. What you can count is latency, refusals, prompts required, and duration.
- Weak: "Student will feel less anxious during fire drills."
- Strong: "Given a social story read before the drill, student will exit the classroom with the class within 60 seconds of the alarm, with no more than one verbal prompt, in 4 of 5 opportunities."
- Weak: "Student will reduce anxiety about the cafeteria."
- Strong: "Student will enter the cafeteria and sit at their assigned table within 2 minutes of arrival, in 4 of 5 days across 3 weeks."
The strong versions are also what the trial measured. Aligning your goal to the outcome the evidence supports is the cheapest credibility you can buy at an IEP meeting.
How many readings does it take?
Six sessions is the floor worth planning for. In ASSSIST-2, story-reading sessions were delivered at least six times across a four-week period, and the students who actually got that dose showed the larger effect. A story read once the morning of the event has not been tested and should not be expected to do much.
Practically: schedule the readings in your session plan the way you would schedule any other trial count. Two readings a week for three weeks, plus one reading immediately before the situation, clears the bar.
What does the wider evidence base say?
Moderate effects, strongest for school-aged children, with older studies more optimistic about anxiety than newer trials. A 2026 Frontiers in Psychology meta-analysis of single-case studies reported a moderate overall effect, with the strongest results for students aged 7 to 12 and for social and safety skills.
Social narratives are rated an evidence-based practice by AFIRM on the strength of single-case design studies covering ages 3 to 22. Single-case designs are good at detecting change in one student in one situation, which is precisely the effect the randomized trial also isolated. The two literatures agree more than they look like they do. If you want the fuller picture, we covered it in what the evidence says about social stories.
When is a social story the wrong tool for anxiety?
When the anxiety is diagnosed, generalized, or driven by something the story cannot change. Three cases to hand off rather than absorb:
- Diagnosed anxiety disorder. The story is a support alongside counseling or cognitive behavioral therapy, not a substitute. Loop in your school psychologist.
- Sensory-driven distress. If the alarm is genuinely painful, no amount of explaining changes the decibels. Pair with ear defenders and an advance warning, and treat the story as the priming layer.
- Unsafe or unpredictable environment. A story that describes a situation inaccurately teaches the student that your stories are not reliable. Fix the environment first.
One more guardrail. Never use a story as a consequence after an incident. It is a priming tool, delivered when the student is calm and before the situation, which is also how the trial delivered it.
Frequently Asked Questions
Did the ASSSIST-2 trial show social stories reduce anxiety?
No. The 2025 ASSSIST-2 cluster randomized trial across 87 UK primary and special schools found no statistically significant difference in anxiety, depression, general health, or parental stress between the social story group and usual care. What did reach significance was individual socio-emotional goal achievement, which favored the social story group.
Why would a social story lower situational anxiety but not test scores?
A social story targets one situation. Broad anxiety scales like the Revised Children's Anxiety and Depression Scale measure anxiety across many situations, most of which the story never touched. A student can be far calmer at the fire drill and score the same overall, because the instrument is not built to detect a single-context change.
Can I write an IEP goal that says a social story will reduce anxiety?
Write an observable, situation-specific goal instead. Anxiety is an internal state you cannot measure directly in a classroom. Write what you can count, such as entering the cafeteria within two minutes of the bell across four of five days, and note the social story as the support.
How many readings before a social story helps a nervous student?
In ASSSIST-2, students who received at least six story-reading sessions across about four weeks showed a larger effect than the group overall. Six exposures before the event is a reasonable floor, and reading the story right before the situation is the part most often skipped.
Should a social story about anxiety name the feeling?
Yes, and name it neutrally. A perspective sentence such as "Some students feel nervous when the alarm is loud" normalizes the feeling without telling the student they are wrong to have it. Avoid directive sentences like "I will not be scared," which set the student up to fail at something they do not control.
Are social stories enough for a student with a diagnosed anxiety disorder?
No. For diagnosed anxiety, a social story is a supporting piece, not the intervention. It works alongside cognitive behavioral therapy, environmental adjustments, and school counseling. Loop in your school psychologist rather than carrying it on a speech caseload alone.
What is a realistic promise to make to a parent?
Say the story may help their student know what will happen next in one specific situation, and that you will track a specific observable behavior for four weeks. Do not promise fewer meltdowns or lower anxiety overall. The trial evidence supports situational goal progress, not global mood change.
One approach for school SLPs short on time is to keep a 5-tool stack: a methodology checklist for the Gray sentence ratio, a slide template you reuse, a folder of stock photos sorted by scenario, an AI text drafter (ChatGPT, Claude, MagicSchool, or Emoquest for one-sentence-in story output), and a delivery format your district already accepts such as Google Slides or PDF. Then put the saved minutes into the six readings, because dosage is the variable the trial actually detected.