An Investigation of the Clinical Use of a Single-Word Procedure to Assess Intelligibility (Swedish Test of Intelligibility for Children) and an Evaluation of the Validity and Reliability of the Intelligibility in Context Scale

Measuring speech intelligibility with deaf and hard-of-hearing children: A systematic review

There is great variability in the ways in which the speech intelligibility of d/Deaf and hard-of-hearing (DHH) children who use spoken language as part, or all, of their communication system is measured. This systematic review examined the measures and methods that have been used when examining the speech intelligibility of children who are DHH and the characteristics of these measures and methods. A systematic database search was conducted of CENTRAL; CINAHL; Cochrane; ERIC; Joanna Briggs; Linguistics, Language and Behavior Abstracts; Medline; Scopus; and Web of Science databases, as well as supplemental searches. A total of 204 included studies reported the use of many different measures/methods which measured segmental aspects of speech, with the most common being Allen et al.'s (2001, The reliability of a rating scale for measuring speech intelligibility following pediatric cochlear implantation. Otology and Neurotology, 22(5), 631-633. https://doi.org/10.1097/00129492-200109000-00012) Speech Intelligibility Rating scale. Many studies included insufficient details to determine the measure that was used. Future research should utilize methods/measures with known psychometric validity, provide clear descriptions of the methods/measures used, and consider using more than one measure to account for limitations inherent in different methods of measuring the speech intelligibility of children who are DHH, and consider and discuss the rationale for the measure/method chosen.

Intelligibility in Context Scale: Growth Curves for Typically Developing English-Speaking Children Between Ages 2;6 and 9;11

PURPOSE

The aim of this study was to quantify the clinical utility of the Intelligibility in Context Scale (ICS) English version by characterizing the growth patterns of the ICS composite scores and seven ICS individual item scores of typically developing American English-speaking children.

METHOD

Parents of 545 typically developing children aged 2;6-9;11 (years;months) completed the ICS. Using a proportional odds model, we regressed ICS composite scores on age and computed for model-estimated mean and lower quantile ICS composite scores. Logistic regression and proportional odds modeling were utilized to quantify the relationship of individual ICS items and age.

RESULTS

ICS composite scores of typically developing children changed with age, but change was small and incremental, with scores compressed between 3 and 5 across the range of ages. An average child (i.e., on the 50th percentile) is expected to have an ICS composite score of 4 beginning at 3;0 and an ICS composite score of 5 by 6;6. On average, parents gave different intelligibility ratings based on communicative partners, and the rating differences between communicative partners decreased with age.

CONCLUSIONS

Given that ICS scores increase with age, the expected score for average children also increases. A child's age is a main factor for interpreting ICS scores.

Translation and validation of the Intelligibility in Context Scale into Sinhala for adolescents in Sri Lanka with cleft lip and palate

The Intelligibility in Context Scale (ICS) is a parent-report screening tool used to measure parents' perceptions of children's functional intelligibility. This tool has been translated into over 60 languages and found to have a good reliability and validity. The purpose of the current study was to translate the ICS into Sinhala (the ICS-SIN), the main language spoken by the Sinhalese people in Sri Lanka, and to validate it with both typically developing (TD) children and children with repaired cleft lip and/or palate (CLP). The translation process followed the forward-backward-forward method. A total of 88 parents of TD children and children with CLP aged 12-15 years old (TD n = 50, CLP n = 38) completed the ICS-SIN questionnaire. Parents of TD and CLP children reported their children's speech as most intelligible to parents and least intelligible to strangers. The ICS-SIN had high internal consistency for both groups (TD α = 0.87, p < 0.05, CLP α = 0.97, p < 0.05). The ICS-SIN total scores and item scores showed significant correlations, indicating a good construct validity. TD participant group's ICS-SIN average mean scores (M = 4.88, SD = 0.29) were significantly higher compared to the CLP ICS-SIN average mean scores (M = 4.64, SD = 0.67) and varied according to gender in both groups, suggesting good discriminant validity. The ICS-SIN has overall good psychometric properties. Therefore, this tool has the potential to be used as a valid parent-rating screening tool for clinical and research purposes in Sri Lanka.

Parent rating of intelligibility: A discussion of the construct validity of the Intelligibility in Context Scale (ICS) and normative data of the Swedish version of the ICS

BACKGROUND

Intelligibility can be defined as the speakers' ability to convey a message to the listener and it is considered the key functional measure of speech. The Intelligibility in Context Scale (ICS) is a parent rating scale used to assess intelligibility in children.

AIMS

To describe normative and validation data on the ICS in Swedish and to investigate how these are related to age, gender and multilingualism.

METHODS & PROCEDURES

Two studies were included. Study 1 included ICS forms from 319 Swedish-speaking children (3:2-9:2 years:months). Study 2 included video recordings and ICS forms from 14 children with speech sound disorder (SSD) and two with typical speech. The video recordings were transcribed in the validation process, resulting in intelligibility reference scores to which ICS scores were correlated.

OUTCOMES & RESULTS

Study 1: The mean value of the ICS for the 319 children was 4.73. There were no differences in ICS score related to age or gender. The children in the multilingual group were significantly older than the monolingual group and had significantly lower ICS scores than the group of monolinguals. Study 2: There was a moderate correlation between the ICS score and the transcription-based intelligibility score, with the two children with typical speech excluded; however, this correlation was not significant.

CONCLUSIONS & IMPLICATIONS

We contribute mean scores and percentiles on the ICS for Swedish-speaking children. The finding that the ICS does not provide valid measures of intelligibility for the included children with SSD suggests that the instrument measures a different construct.

WHAT THIS PAPER ADDS

What is already known on the subject The ICS has been translated to numerous languages and validated against articulation measures in several previous studies. The validity of the Swedish version has been investigated against intelligibility based on transcription of single words. What this paper adds to existing knowledge The study provides normative values of the Swedish version of the ICS for children aged 3-9 years. This is the first study to use a gold standard measure of intelligibility in continuous speech to validate the ICS. The results show a somewhat dubious validity regarding ICS for the group of children with SSD included in the study. What are the potential or actual clinical implications of this work? The ICS's suitability as a measure of intelligibility is questionable; however, it might be of use for speech and language pathologists to give an overview of the parents' view of their child's ability to communicate, in order to make a decision on possible further assessment and intervention. The normative values of the Swedish version of the ICS could be of use in this decision process.

Validity of Parent Ratings of Speech Intelligibility for Children with Cerebral Palsy

Aim: To examine the relationship between subjective parent ratings of intelligibility and objectively measured intelligibility scores for children with cerebral palsy (CP) with differing levels of speech severity. Method: Fifty children (84-96 months) with CP were classified into groups based on intelligibility scores during a speech elicitation task - high intelligibility (90% or higher), mild-moderate intelligibility reduction (61-89%), and severe intelligibility reduction (60% or lower). Parent ratings of understandability (on a 7-point scale) were compared to intelligibility scores gathered from 100 naïve listeners. Results: For children with mild-moderate and severe intelligibility reduction, there was a large range of variability in parent ratings. For children with high intelligibility, ratings were consistent with intelligibility scores. There was a range of intelligibility scores within each rating, especially in the middle of the scale. Conclusions: For children with mild-moderate intelligibility deficits, parent ratings may best be used in conjunction with objective measurement of intelligibility.

Audience Response System-Based Evaluation of Intelligibility of Children's Connected Speech - Validity, Reliability and Listener Differences

PURPOSE

We assessed audience response systems (ARS)-based evaluation of intelligibility, with a view to find a valid and reliable intelligibility measure that is accessible to non-trained participants. In addition, we investigated potential listener differences between pediatric speech and language pathologists (SLPs) and untrained adults.

METHOD

Sixteen one-minute samples of connected speech were compiled, collected from 14 children with a speech sound disorder (SSD) and two children with typical speech. 16 SLPs and 13 untrained adults participated in a series of ARS listening sessions, where they were fitted with headphones and hand controls, and instructed to click a button whenever they did not understand the child speaking. Listeners' button clicks were registered and, for each speech sample, totaled into an (un)intelligibility index. The proportion of syllables perceived correctly - based on orthographic listener transcripts - was used as a reference score of intelligibility.

RESULTS

The ARS-based intelligibility scores correlated strongly with the intelligibility reference score. Reliability was high across listener groups and weaker for single listeners. No significant difference was found between the evaluations of SLPs and untrained adults.

CONCLUSIONS

ARS-based evaluation offers a valid and reliable measure of intelligibility of particular value in research as a practical tool for collecting input from listeners without experience or knowledge of SSDs. We stress that the ARS design presupposes a listener panel, and that evaluations obtained from individual listeners are predictably inadequate in terms of reliability.