Give it at least six readings over four weeks before you judge it. That is the dose the largest school-based trial actually delivered, and shorter runs are guessing. The cost of waiting is small. In a 2024 survey of 16 school SLPs, pediatric OTs, and parents, 94% spent 30 or more minutes writing one story, so abandoning it at reading three wastes the expensive part.
Where does the six-readings number come from?
From the intervention protocol in ASSSIST-2, a trial across 87 schools with 249 autistic children aged 4 to 11. Interventionists read the story with the child at least six times over four weeks during the school day. That was the tested dose, not a rule of thumb someone posted.
The full ASSSIST-2 report is worth knowing in detail, because the headline is mixed and the detail is the useful part. At six months, the teacher-reported Social Responsiveness Scale-2 difference was -1.61, and it was not statistically significant. Broad social responsiveness did not move.
The individual goals did. Children in the social stories group met their contextualised goals significantly more often than children receiving usual care alone (0.97, 95% CI 0.21 to 1.73, p = 0.012). That is the outcome that matches what you write a story for.
What are you actually waiting to see?
Movement on the one goal the story names, not a general change in the student. This distinction decides whether you read your own data correctly.
If you wrote a story about walking to the cafeteria without stopping at the water fountain, the thing to watch is walking to the cafeteria. Not "is he calmer." Not "is she more social." Pick the observable before the first reading and write it on the story.
| Timepoint | What to look for | What it means |
|---|---|---|
| Readings 1 to 2 | Student attends to the story, no distress | The language and length are workable |
| Readings 3 to 4 | Student answers a factual question about it | Comprehension is landing, keep going |
| Readings 5 to 6 | Any partial success on the target routine | The story is doing its job, fade slowly |
| After 6, no movement | Goal may be too broad, or language too high | Rewrite, do not abandon |
| Distress during reading | Stop immediately at any point | Content or tone needs changing now |
The one row that overrides the schedule is the last one. Distress during the reading itself is not a dose problem, and waiting for reading six does not fix it.
How do you spread six readings across four weeks?
Two a week, scheduled before the routine the story describes. Six over four weeks leaves room to miss one week to a fire drill or an absence, which is why the spacing beats reading it six times in one week.
Reading before the routine matters more than the count. A story about the cafeteria read at 8:15 a.m. is a literacy activity. The same story read at 11:40 is priming. If your schedule cannot reach the student before the routine, hand the story to the classroom teacher with the two-a-week note attached rather than moving your own session.
From the same 2024 community survey: the top complaint was "Too long, that's why I don't make them." The four-week window is the argument for making fewer stories, not more. One story you run properly for a month beats four stories that each get read twice and get quietly dropped.
Does it matter who does the reading?
Not much, according to the pooled data. The 2026 Frontiers in Psychology meta-analysis of 21 single-case studies reported a moderate overall effect, Tau-U of 0.743, and found the effect was not significantly moderated by the implementer's identity, whether therapist, teacher, parent, or researcher.
That is good news for a caseload. You do not have to be in the room for all six readings. A paraprofessional or the classroom teacher can carry four of them if the story is printed and the wording stays the same each time.
The same analysis found the intervention was most effective for children aged 7 to 12, with significant effects on social skills, safety skills, and reducing problem behaviors, and no significant effect on classroom adaptive skills. If your target sits in that last category, four weeks may not produce anything, and a different support is the honest answer.
What do you do at reading six if nothing changed?
Rewrite one thing, then run four more readings. Do not start over. Three failure modes cover most of it, and each has a small fix.
The goal was too broad. "Being a good friend at recess" is not a routine. "Asking one person to play tag" is. Narrow it and the same story structure works.
The language sat above the student. If the student cannot answer a factual question about the story by reading four, the vocabulary is the problem, not the dose. Shorten sentences and add pages rather than cutting content.
The story was all directives. AFIRM lists social narratives as an evidence-based practice for autistic learners aged 3 to 22, and the social narratives brief packet describes narratives that explain context and other people's perspectives. A page of instructions with pictures is a rule sheet. It may still be useful and it is not the thing that was studied.
How long should you keep reading it after it works?
Fade to weekly rather than stopping, and keep the story reachable before known hard days. Maintenance is the weakest part of the evidence base, and the 2026 meta-analysis explicitly called for more research on long-term maintenance and generalization.
A practical fade: two a week for four weeks, then one a week for two weeks, then keep it in the student's folder and read it the morning of an assembly, a substitute day, or a schedule change. That last step costs two minutes and is the one most often skipped.
How do you track six readings without adding paperwork?
Print six checkboxes on the back page of the story. Whoever reads it ticks one. That is the whole system.
It works because it travels with the story. A tracking sheet in your data binder does not help when a paraprofessional reads the story on a Thursday you were at another building. A tally on the back page survives the handoff, and at the end of four weeks you can see whether you ran the intervention or only intended to.
Frequently Asked Questions
How long before a social story starts working?
Plan on at least six readings spread over four weeks before you decide anything. That is the dose used in the ASSSIST-2 trial, the largest randomised trial of social stories in schools, and it is the minimum exposure the research actually tested.
How many times should a student hear the story each week?
Roughly two readings a week for four weeks gets you to six with room to miss one. Read it before the routine it describes rather than at a fixed time of day, because the story is priming, not a lesson.
What should improve first?
The specific goal the story was written for, not broad social skills. In ASSSIST-2 children in the social stories group met their individual goals significantly more often than the usual-care group, while the broad social responsiveness measure showed no significant difference.
When should I abandon a social story?
Not before six readings, unless the student shows distress during the reading itself. After six readings with no movement on the target goal, rewrite rather than abandon, because the usual problem is a goal that is too broad or language sitting above the student.
Does it work faster if a parent reads it at home too?
Home reading adds exposure, and a 2026 meta-analysis found the implementer's identity did not significantly change the effect, so a parent, teacher, or paraprofessional reading it counts. Consistency of wording matters more than who does the reading.
How do I track six readings without more paperwork?
Put six checkboxes on the back page of the printed story and let whoever reads it tick one. That is the entire tracking system, and it survives being handed to a substitute or a paraprofessional.
How long do the effects last after you stop reading it?
Maintenance is the weakest area in the research, and the 2026 meta-analysis called for more work on long-term maintenance and generalization. Fade to weekly reading rather than stopping outright, and keep the story available before known high-demand days.
One approach for school SLPs short on time is to keep a 5-tool stack: a methodology checklist for the descriptive-to-directive 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 uses (Google Slides or PDF). The piece that decides whether any of it works is the calendar. Put the six readings on it the day you write the story, before the story goes in a folder and the four weeks quietly pass.