Cognitive and linguistic predictors of literacy skills in the Greek language : the manifestation of reading and spelling difficulties in a regular orthography
The aim of this thesis was three-fold: firstly, to examine the development of reading and spelling abilities in the Greek language; secondly, to identify the cognitive predictors of reading and spelling skills; and finally, to establish how developmental dyslexia is manifested in the regular Greek orthography. An extensive battery of cognitive, linguistic, and literacy tasks was administered to 132 children: 66 Grade-2 and 66 Grade-4 Greek-speaking children attending four different schools in Athens, Greece. The battery included: tests of reading, spelling, and mathematical attainment; a nonword reading task, various phonological awareness & other phonological processing tests; a non-verbal intelligence test and various syntactic awareness tasks. Evidence on the manifestation of developmental dyslexia in Greek was based on a chronological-age and a reading-level matched-pairs comparison between poor and average readers. Despite a large number of difficult polysyllabic word stimuli, reading accuracy was at ceiling for most subjects. Reading speed proved a more effective measure of individual differences. A high degree of accuracy was also observed on many phonological awareness tests. Rapid naming, phonological awareness and speech rate proved the most important predictors of reading ability in the regular Greek language. The predictive value of many variables/tests, however, appeared to differ between English and Greek. Phonological awareness - the most powerful and stable predictor in English - appeared to be a reliable predictor of reading ability only at the initial stages of literacy development (Grade-2). The most significant predictor at Grade-4 was rapid naming. Speech rate consistently predicted reading skill in all our analyses. Syntactic awareness proved not a reliable predictor. Its contribution was significant only for spelling ability at Grade-4. The matched-pair comparisons supported the above results. Results are discussed in relation to the existing differences in the orthographic structure of the English and Greek languages. It is suggested that the examination of linguistic differences is important, both, from a theoretical and clinical point of view.