Identifying language impairment in bilingual school-aged children using one word vocabulary tests

Abstract

Many children within the South African education system are learning in English which is not their first language, placing them at risk for academic failure. Differentiating between a typically developing bilingual child and a bilingual child with a language impairment is challenging. There are limited resources and assessment tools available in South Africa that accurately and sensitively assess bilingual children who are at risk for a possible language impairment. Testing a bilingual child’s vocabulary in both their languages typically provides a more accurate picture of language skills and may be sensitive to possible language impairment. Purpose: The main aim of this study was to explore the use of one word vocabulary tests to identify a possible language impairment in bilingual children in the school setting. This study also aimed to compare the vocabulary of monolingual children and bilingual children. Participants who were at risk for possible language impairment were identified and their language skills were further assessed and described. Method: The research design of this study was quantitative, descriptive, cross-sectional and comparative in nature. Eight-year-old learners (n= 60) at an English medium school were approached to participate. These participants were divided into two groups: one group being monolingual English (n= 30) and the other group having isiZulu as the primary home language with English as an additional language (n= 30). The Receptive and Expressive One Word Picture Vocabulary Tests (Martin & Brownell, 2011a and b) were used to assess each learner. The monolingual group was assessed in English and the bilingual group was assessed in English and isiZulu. Raw scores were analysed using descriptive statistics, independent sample t-tests and paired sample t-tests. Participants that were identified as having a possible language impairment, were further assessed using the Diagnostic Evaluation of Language Variation-Criterion Referenced (DELV-CR) (Seymour, Roeper & De Villiers, 2003) to diagnose and describe their language competence. Results and implications: The results of this study showed that EAL learners benefit from being assessed in both languages as well as having results compared to the peer group mean. The EAL learners scored better in the English assessments than in the isiZulu assessments. When scored for conceptual vocabulary using composite scoring, the mean score of the EAL learners did improve but there was still a significant difference between the composite score and the monolingual English score. This may be due to poor performance on the isiZulu assessment. Six of the 30 isiZulu children were flagged as having a possible language impairment based on composite scoring. Once assessed further on the DELV, 4/30 participants were described as having a language impairment. The language profile that the identified participants presented with was consistent with existing literature on EAL learners in South Africa. These findings indicate that with some caution and the use of peer-mean comparisons, bilingual vocabulary tests and composite scoring has potential to be used as part of a battery of assessments to differentiate between a language difference and a language impairment. Appropriate assessment materials and resources are lacking within the South African context and the results of this study suggest a new valuable method and materials for accurate and reliable assessment, especially valuable in the school context. This study also added to the body of research regarding the language profile of bilingual and monolingual children in South Africa