Comparing the Informativeness of Single-Word Samples and Connected Speech Samples in Assessing Speech Sound Disorders

Area of Science:

• Speech-language pathology
• Developmental pediatrics
• Linguistics

Background:

• Accurate identification of speech sound disorders (SSD) in children is crucial for timely intervention.
• Speech-language pathologists (SLPs) face challenges in selecting efficient and sensitive speech sampling methods.
• Understanding the representativeness of single-word versus connected speech samples is key for accurate diagnosis.

Purpose of the Study:

• To compare the sensitivity of single-word and connected speech samples in detecting atypical speech sound development.
• To determine if single-word samples adequately represent a child's overall speech sound performance.
• To inform SLPs on optimal speech sampling techniques for preschool children with SSD.

Main Methods:

• Compared speech sound performance in 37 preschool children with SSD and 37 typically developing (TD) children.
• Elicited speech using a picture-naming task (single words) and a story-retelling task (connected speech).
• Analyzed intelligibility, speech accuracy, phonemic inventory, and phonological patterns across sample types and groups.

Main Results:

• Single-word samples were more sensitive than connected speech samples in differentiating SSD from TD groups.
• Effect sizes for single-word samples were higher for intelligibility, speech accuracy, and phonemic inventory.
• The number of phonological patterns showed no performance gap between sample types.

Conclusions:

• Single-word samples are more informative for differentiating children with SSD from TD children on key measures.
• Findings guide SLPs in selecting efficient sampling methods, especially under time constraints.
• Performance differences between sample types can be influenced by salience and avoidance effects, impacting interpretation.