Reading comprehension and language skills of deaf students in Croatia

Iva Hrastinski, Purdue University

Abstract

The primary purposes of this study were to assess reading comprehension skills of deaf middle school and high school students in Croatia as well as to investigate the relationship between language skills and higher-level reading comprehension skills (inferencing) in this population. Fifty four deaf students without any additional impairments took part in the study and were grouped based on their linguistic status as either monolingual (Croatian only) or bilingual (Croatian and Croatian Sign Language - HZJ) participants. Croatian receptive vocabulary knowledge (assessed by PPVT-III HR), receptive grammatical knowledge (TROG-2:HR) and reading comprehension skills were determined for all participants. In addition, bilingual students received HZJ receptive proficiency tests designed for this study: HZJ vocabulary assessment, HZJ morphosyntactic knowledge tests and HZJ narrative comprehension assessment. Results indicated overall poor performance on Croatian tests: low level of lexical knowledge, deficient comprehension of grammatical structures and inadequate reading comprehension scores. Monolinguals outperformed bilinguals on all Croatian measures; the interpretation of these results is tempered by educational placement differences between the two groups. As expected, bilinguals scored better in the HZJ narrative comprehension task than in the Croatian reading comprehension assessment. With respect to explicit and indirect (inferencing) questions in the Croatian reading comprehension test, monolinguals achieved better scores on explicit questions than on the inference-requiring implicit questions. In contrast, bilinguals'sperformance on the two question types was statistically indistinguishable. The bilinguals repeated their performance on the two types of questions in HZJ as well. Further examination of the bilingual performance pattern on the two question types in Croatian and HZJ revealed that there were no differences between average group performance on direct questions in Croatian and HZJ, suggesting that bilingual students are equally able to use information from the text/signed story to answer explicit questions in each language. However, their performance was significantly poorer on the implicit questions in the Croatian assessment than in the HZJ assessment, suggesting a relationship between language proficiency (better in HZJ) and higher-order comprehension skills (inferencing skills). Factors contributing to reading comprehension skills differed between the two groups. Multiple regression results indicate that one Croatian receptive grammar test performance explained 64.5% of the Croatian reading comprehension variance in monolingual students. While bilingual students' understanding of Croatian texts was also predicted by Croatian grammatical test scores, their HZJ narrative comprehension was significantly predicted by HZJ vocabulary knowledge. These findings might provide a context for exploration of crosslinguistic transfer of higher-order inferencing skills. Additional research is needed regarding the question of crosslinguistic transfer to enhance performance in a less accessible language (e.g. Croatian) in deaf bilinguals. Future research with a larger sample size and clearly defined bilingual criteria is warranted to explore if the trend noted in current work might inform deaf educational practices.

Degree

Ph.D.

Advisors

Wilbur, Purdue University.

Subject Area

Bilingual education|Linguistics|Audiology|Reading instruction|Language

COinS