Current Research: Optimal Language Sample Analysis
The lab is currently recruiting bilingual children between 6;0 and 9;11 to participate in a study. This study aims to determine the optimal elicitation technique for Language Sample Analysis (LSA) in school-age children who are learning English as a second language. Speech-language pathologists (SLPs) often examine bilingual children’s language using LSA to determine the presence of a developmental language disorder (DLD; e.g., Ortiz et al., 2024). LSA is a less-biased assessment compared to standardized tests for this population. It involves analyzing language skills from a sample generated through various methods (i.e., conversation, narrative, play-based, expository), depending on the child's age. A language sample is optimal when it showcases the child’s language skills, so an ideal elicitation method will generate a language-rich sample. This study is the next step following a project that compared narrative and conversational samples with bilingual children (Potratz & Haskill, 2025). The results of that project showed that narratives produced greater syntactic complexity (grammar), while conversations showed greater lexical diversity (vocabulary). Further research is needed to examine children’s expository skills (explaining/describing).
To determine the optimal elicitation method for children learning English as a second language, the research question for the proposed project asks if there are differences between LSA measures in conversational and expository (FGST) contexts elicited from bilingual children.It is hypothesized that the expository method will elicit more complex language compared to conversation for LSA measures, like what has been found with monolingual children (Nippold et al., 2005; Westerveld & Vidler, 2016). The overall goal of this line of research is to prevent bilingual children from being misidentified as having speech and language impairments. By identifying an optimal language sample elicitation method for bilingual children, language skills will be showcased, so the true skills can be assessed.
Bilingual and Monolingual Speech-Language Pathologists’ Intelligibility Ratings of Children’s Non-Native Speech
Purpose: This study examined bilingual and monolingual speech-language pathologists (SLPs) speech intelligibility ratings of non-native English-speaking children’s conversational speech to determine whether differences exist between these two perceiver groups. Bilingual perceiver differences have been shown in previous research and could impact SLPs’ intelligibility ratings, which are an important part of an assessment for speech sound disorders (SSD).
Method: Thirty bilingual and 30 monolingual SLPs rated audio recordings of conversational speech samples from 15 school-aged, typically developing Spanish-English and Mam-English bilingual children. Speech samples were presented in a randomized order in an online experiment. The data were analyzed using a linear mixed-effects model to determine the effects of SLPs’ lingualism on intelligibility ratings.
Results: Bilingual and monolingual SLPs’ intelligibility ratings were not significantly different from each other. The intelligibility ratings of the Mam-English speaker group were significantly different from those of the Spanish-English speaker group in both listener groups.
Conclusions: Bilingual and monolingual SLPs rated the intelligibility of spontaneous speech samples from bilingual child speakers similarly. The familiarity of specific non-native accents plays a role in SLPs’ intelligibility ratings. Clinically, when assessing a bilingual child whose first language is unfamiliar, SLPs must incorporate information from resources such as interpreters.
Potratz, J. (in press). Bilingual and monolingual speech-language pathologists’ intelligibility ratings of children’s non-native speech. Perspectives of the ASHA Special Interest Groups.
Purpose: Overall, seeing a speaker results in higher intelligibility than hearing speech alone because of visual speech cues. Intelligibility ratings of conversational speech samples provided by school-age bilingual children in two presentation modalities (i.e., audio-only and audio-visual [AV]) were examined to determine whether there were other impacts of seeing the children.
Method: Forty speech-language pathologists (SLP) rated the intelligibility of conversational speech samples provided by 23 typically developing school-aged children in three groups: bilingual Spanish-English, bilingual Mam-English (Mam is an Indigenous, Mayan language spoken in Guatemala), and monolingual English (control). SLPs heard audio-only and AV recordings, with individual two-minute speech samples in a randomized order, in an online experiment. Data were analyzed using a linear mixed-effects model to determine if there were effects of presentation modality, speaker group, or accentedness on the intelligibility ratings.
Results: No statistical differences were found between intelligibility ratings from the audio-only and AV presentation modalities for any speaker group. Accentedness was a significant factor on intelligibility ratings, whereas speaker group was not.
Conclusions: SLPs rated conversational speech samples of monolingual English and bilingual child speakers as having similar intelligibility regardless of whether they heard audio-only or AV recordings, even though visual speech cues should have caused a difference. Clinically, when SLPs use intelligibility ratings in their assessments of bilingual children’s speech for speech sound disorders, it is vital to consider the influences of non-native accent, language exposure, and linguistic bias.
Potratz, J. R. (2025). Impacts of Seeing Bilingual Children on Intelligibility Ratings. Perspectives of the ASHA Special Interest Groups, 1–9. https://doi.org/10.1044/2025_PERSP-25-00043
Manuscript Under Review
Purpose: This study examined differences between two language sample types, conversation and narrative, to determine whether there is an optimal type for assessing bilingual school-age children’s syntactic complexity and lexical diversity with language sample analysis (LSA).
Method: Conversation and narrative language samples were elicited from 24 typically developing school-age children in three groups: bilingual Spanish-English, bilingual Mam-English, and monolingual English. Two measures of syntactic complexity (mean length of utterance in morphemes [MLUm] and clausal density [CD]) and two measures of lexical diversity (number of different words [NDW] and moving average type token ratio [MATTR]) were calculated for language sample type comparison.
Results: All four LSA measures differed significantly between the two language sample types. The measures of syntactic complexity (MLUm and CD) were significantly greater in narratives compared to conversations. Similarly, the measures of lexical diversity (NDW and MATTR) were significantly different between narrative and conversation contexts; conversational values were higher than narrative values. Group differences were found in both of the language sample types.
Conclusions: The optimal language sample context for bilingual school-age children may best be determined by the suspected area of difficulty for each child. If a syntactic complexity deficit is suspected, a narrative sample is preferable; if a lexical diversity deficit is suspected, a conversational sample is preferable.
Defining and Measuring Intelligibility in the Evaluation of School-Aged Children’s Speech
Coauthor: Melissa Redford, Under review
Purpose: The current study tests whether a measurement focus on listener comprehension versus speech impacts the sensitivity of the intelligibility measures, in particular, the ability to distinguish children with speech sound disorder (SSD) from typically developing (TD) children in the context of age and second language effects on speech.
Method: Conversational speech and controlled sentences were elicited from 30 younger and older school-aged children in four language groups: English-speaking monolinguals with SSD (SSD-English); typically-developing monolinguals (TD-English), and first language Spanish (TD-L1 Spanish) or Mam (TD-L1 Mam) speakers. Forty-two SLPs provided global intelligibility estimates (GIE) on conversational samples—a common clinical practice; they also transcribed and rated controlled sentences for laboratory-based intelligibility scores (LIS) and comprehensibility ratings (CR). Language skill ratings (LSR) and accentedness ratings on conversational speech were also collected.
Results: Although all intelligibility measures (GIE, LIS, CR) were sensitive to effects of age and language group, GIE provided the poorest discrimination between groups and CR the best. No measures distinguished SSD-English from TD-L1 Mam. All measures overlapped with LSR. Perceived accentedness was a significant predictor of by-child intelligibility above and beyond independent language measures, regardless of the measurement type.
Conclusions: Conversational samples introduce confounds to intelligibility measures, including language skills and non-native English features. Orthographic transcription of controlled sentences overlooks prosody that also impacts intelligibility. Comprehensibility ratings of controlled sentences have the fewest confounds while allowing for more holistic listening.