What a new all-day recording study tells us about measuring early stuttering

If your preschooler stuttered a lot at breakfast, hardly at all by lunch, and then struggled again at the park, you are not imagining it, and you are not doing anything wrong. Stuttering in young children is famously changeable. Until recently, though, no one had tracked exactly how changeable it is across a single day, or asked what that means for the way parents and speech-language pathologists (SLPs) rate it.

A 2026 study by Mark Onslow and colleagues at the Australian Stuttering Research Centre set out to answer that question.

Why this matters

In the Lidcombe Program, an early stuttering treatment for children under six, parents give a daily severity rating (SR) on a simple scale from 0 (no stuttering) to 10 (extremely severe). SLPs use these ratings to decide what to do next, including when a child is ready to move to Stage 2. The scale is free, needs no equipment, and children don’t know they’re being assessed.

But the ratings are meant to reflect a whole day, and the research behind them had only looked at short speech samples of two to ten minutes. No one had checked how well one number captures a full day of speaking.

What the researchers did

Ten children aged 3 to 4 who stuttered wore a small recorder in a custom T-shirt pocket for about seven hours at home. An SLP then rated every 15-minute segment of each child’s day. Parents also gave a “typical for the day” rating, once before and once after brief training on the scale. A random 10-minute phone conversation was recorded during each day to see how well a short sample represented the whole.

What they found

Stuttering varied a lot, hour to hour. Every child’s severity changed across the day. In seven of ten children the range was 4 to 6 points on the scale, enough to span mild, moderate, and severe. Ratings sometimes jumped 3 to 6 points from one 15-minute block to the next.

There was no “good time of day.” Four children trended slightly up and six slightly down, but none by more than about one point. Mornings were not reliably better or worse than afternoons.

Parents rated well, even before training. Parent ratings were typically within one or two points of the SLP’s. Training changed the rating for only four of ten parents, by one or two points. Most parents were already using the scale the way an SLP would. Parents tended to rate slightly higher than the SLP, and their “typical” score behaved more like an average of the child’s best and worst moments than the most common score of the day.

A 10-minute sample was a decent snapshot, but only a snapshot. For most children it landed within one point of the day’s most typical score, yet it can’t show the swings, and it could look misleadingly good if recorded during a low-stuttering stretch.

What this means in practice

For parents: Variability is normal, not a sign of regression or of something you caused. If your SLP asks for ratings, expect to give more than one, such as a typical score and a highest score for the day. Noting when the hard moments happen (a certain activity, listener, or level of tiredness) is genuinely useful information.

For SLPs: The authors suggest that a “typical” plus a “highest” daily SR should be routine rather than optional, particularly before progressing a child to Stage 2, since a single score can hide the peaks. Keep calibrating with parents in session, and resolve discrepancies by consensus. For research, a parent-reported typical and highest SR may better reflect outcomes than a single 10-minute sample.

An honest caveat

This was a small study of ten children, each recorded for one day. All were recruited from clinic waiting lists, none had received treatment, and the children knew they were being recorded, which may have influenced their speech. The study cannot tell us why stuttering varied, whether through activity, listener, fatigue, or chance, nor whether variability looks different once a child is in treatment. Those questions remain open.

What it does show clearly is that one number rarely tells the whole story of a preschooler’s day.

Source

Onslow, M., Thambyaiyah, J., O’Brian, S., Packman, A., & Lowe, R. (2026). Variability of preschool stuttering during a day. American Journal of Speech-Language Pathology, 35, 2107–2114. https://doi.org/10.1044/2026_AJSLP-25-00246

Onslow, M., Webber, M., Harrison, E., Arnott, S., Bridgman, K., Carey, B., Sheedy, S., O’Brian, S., MacMillan, V., Lloyd, W., Santayana, G., Hodes, G., Smith, K., & Hearne, A. (2025). The Lidcombe Program treatment guide. https://www.uts.edu.au/research/centres/australian-stuttering-research-centre/resources/lidcombe-program