Send to

Choose Destination
Int J Lang Commun Disord. 2016 Nov;51(6):715-731. doi: 10.1111/1460-6984.12242. Epub 2016 Mar 15.

Diagnostic accuracy of repetition tasks for the identification of specific language impairment (SLI) in bilingual children: evidence from Russian and Hebrew.

Author information

Bar Ilan University, Ramat Gan, Israel.
Bar Ilan University, Ramat Gan, Israel.



Previous research demonstrates that repetition tasks are valuable tools for diagnosing specific language impairment (SLI) in monolingual children in English and a variety of other languages, with non-word repetition (NWR) and sentence repetition (SRep) yielding high levels of sensitivity and specificity. Yet, only a few studies have addressed the diagnostic accuracy of repetition tasks in bilingual children, and most available research focuses on English-Spanish sequential bilinguals.


To evaluate the efficacy of three repetition tasks (forward digit span (FWD), NWR and SRep) in order to distinguish mono- and bilingual children with and without SLI in Russian and Hebrew.


A total of 230 mono- and bilingual children aged 5;5-6;8 participated in the study: 144 bilingual Russian-Hebrew-speaking children (27 with SLI); and 52 monolingual Hebrew-speaking children (14 with SLI) and 34 monolingual Russian-speaking children (14 with SLI). Parallel repetition tasks were designed in both Russian and Hebrew. Bilingual children were tested in both languages.


The findings confirmed that NWR and SRep are valuable tools in distinguishing monolingual children with and without SLI in Russian and Hebrew, while the results for FWD were mixed. Yet, testing of bilingual children with the same tools using monolingual cut-off points resulted in inadequate diagnostic accuracy. We demonstrate, however, that the use of bilingual cut-off points yielded acceptable levels of diagnostic accuracy. The combination of SRep tasks in L1/Russian and L2/Hebrew yielded the highest overall accuracy (i.e., 94%), but even SRep alone in L2/Hebrew showed excellent levels of sensitivity (i.e., 100%) and specificity (i.e., 89%), reaching 91% of total diagnostic accuracy.


The results are very promising for identifying SLI in bilingual children and for showing that testing in the majority language with bilingual cut-off points can provide an accurate classification.


bilingual children; diagnostic accuracy; forward digit span; non-word repetition; sentence repetition; specific language impairment (SLI)

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center