Gold standard test of child’s speech abilities may not capture the whole picture, new study reveals
By Charlene N. Rivera-Bonet | Waisman Science Writer
At a Glance:
- Orthographic transcription intelligibility measures are commonly used in research settings to measure speech intelligibility in children, while the Intelligibility in Context Scale is more commonly used clinically.
- This study, from the lab of Katie Hustad, PhD, looked into how well these two measures were correlated.
- Accounting for age, orthographic transcription intelligibility and the Intelligibility in Context Scale did not have a strong correlation with each other. This suggests that they may be complementary measures, as opposed to redundant or a proxy for each other.
- The research manuscript won an Editor’s Award for meeting the highest quality standards in research design and presentation.
Two well-established measures of how well a child is understood by others, previously believed to capture the same thing, are not well correlated in typically developing children when accounting for age, a new Waisman Center study shows. This suggests that they may be more complementary than redundant, and both may be needed for a fuller picture of a child’s speech.
The study, from the lab of Katie Hustad, PhD, Oros Bascom Professor of Communication Sciences and Disorders and Vilas Distinguished Achievement Professor, examined orthographic transcription intelligibility and the Intelligibility in Context Scale measures from over 500 children and their parents in Wisconsin, and over 1000 listeners who heard those children, to assess if the tools are capturing the same or different aspects of intelligibility. The research, published in the American Journal of Speech-Language Pathology won an Editor’s Award for meeting the highest quality standards in research design and presentation.
Reduced intelligibility may lead to communication challenges that can negatively impact academic performance and social engagement. Because of this, speech intelligibility measures are important to identify children who need support and intervention early on.
Orthographic transcription intelligibility, used mainly in research, involves an unfamiliar person listening to recorded speech samples and typing the words they hear. It is an objective measure where the listener has no context clues and relies purely on the quality of the acoustics. The downside is that requiring an unfamiliar listener to evaluate the recordings can be impractical and time-consuming.
On the other hand, the Intelligibility in Context Scale, mostly used in clinical settings, asks parents seven questions related to how well they think their child is understood by others including family members, teachers, or strangers. This provides a quick way to estimate a child’s speech intelligibility. “It basically asks parents to make a judgment of a child’s understandability,” Hustad explains. “So, it’s very much through the lens of the parents, and parents bring a different perspective to the table than people who know a child less well,” she adds.
In order to more accurately understand the correlation between the objective and subjective measures in typically developing children, they included age as a variable. Children’s speech and intelligibility is expected to become clearer as they grow older. “If we take out that variable of age, how correlated are [the measures]?” explains Jenn Soriano, PhD, former graduate student in Hustad’s lab and first author of the study. “If we just look at it without taking out the effect of age in this picture, [the correlation] will look good because children are much more well understood when they’re nine as compared to when they were two.”
Accounting for the effect of age, the results showed that the orthographic transcription intelligibility and the Intelligibility in Context Scale did not have a strong correlation with each other, suggesting that they may be complementary measures, as opposed to redundant or a proxy for each other.
“From a clinical perspective, it suggests that the Intelligibility in Context Scale tool that’s so widely used doesn’t capture the whole picture,” Hustad says. “And that if you want a more precise estimate of how well a child’s speech is working to convey words that people understand, you need to measure that quantitatively.” She hopes these results will help clinicians make more informed decisions about intelligibility measurements and what information can be interpolated from each one.
In the future, they would like to understand the relationship between these measures in different speech disorders. “With typically developing children there is that trajectory where their speech is getting better as they get older, but for some of the children who have developmental conditions, it may not be the same,” Soriano says.
This research study was recognized with an Editor’s Award from The American Speech-Language-Hearing Association. “It was a very lovely surprise because I’m in my early career as a researcher, and so this is a very meaningful affirmation that I can be a researcher, and that the work I’m doing can contribute to the field, and it’s actually recognized,” Soriano shares.
“It’s a really significant achievement for Jenn in her early career to have her work recognized with this most prestigious honor in the field,” Hustad says. “As a PhD student, to win an Editor’s Award is a really big deal.”