Phonological predictors of nonword repetition performance in bilingual children

Previous research (Gibson et al., 2015; Summers, Bohman, Gillam, Peña & Bedore, 2010) has suggested an advantage in Spanish over English on nonword repetition tasks with Spanish-English bilingual children. However, comparing nonwords of equal syllable lengths across languages may not sufficiently account for phonological differences across languages. We compared Spanish-language nonword sets of different lengths to select a set that would be equivalent with respect to difficulty in English. We considered language-specific phonological structure and level of difficulty in evaluating nonword performance in 126 first- and second-grade, Spanish-English bilingual children. We predicted that adding 5-syllable words to the Spanish nonword set and monosyllabic words to the English set would result in comparable difficulty.

Participants repeated nonwords of increasing lengths in English of 1, 2, 3, and 4 syllables and in Spanish of 2, 3, 4, and 5 syllables. Percent phonemes correct was calculated for total sounds. A repeated measures ANOVA was used to compare accuracy across word lengths.

Results indicated significant differences in overall means between languages when nonwords of equal word lengths were compared, but no significant differences in overall means when lengths of 1 - 4 syllables in English and 2 - 5 syllables in Spanish were used. Differences between languages varied at individual levels of complexity.

The finding that level of word length affects performance may have implications for understanding the factors in clinical test development for bilingual children. Further research may benefit from assessing additional language variables and additional language combinations to extend findings.