Presented at the “Language Disorders in Greek” conference, Patra, Greece, 10–11 June 2006

Greek is a relatively free-stress language in which the location of stress on a phonological word is restricted to the last three syllables. As a core feature of the phonological specification of words, stress assignment is essential in word production. Because of the existence of stress minimal pairs, such as γέρος-γερός, in which stress alone disambiguates a word pair, correct stress assignment from printed text is necessary for high reading performance. In spelling, stress is indicated by a diacritic over the vowel of the stressed syllable, and is mandatory on every word with two or more syllables. Hence any omission or misplacement of the stress diacritic is a spelling error.

Despite the importance of stress assignment in reading and spelling, that is, in both receptive and expressive written language, little attention has been paid to the cognitive and linguistic processes involved in stress assignment, to the vulnerability of these processes in the case of specific learning disabilities affecting written language, or to the potential usefulness of assessing stress assignment performance for the evaluation of reading and writing competence. In a study of word and pseudoword reading, Protopapas (2006) found that children make frequent stress assignment errors in pseudowords but not in words, and that such errors are somewhat related to reading ability, suggesting that decoding the diacritic may be a demanding process and that stress assignment in reading may be largely lexical.

Spelling studies have determined that Greek children use phonological strategies at the phoneme and syllable level from the earliest stages (Porpodas, 2001), gradually augmented with morphological strategies to allow spelling of grammatical morphemes (Nunes, Aidinis, & Bryant, 2006), and are sensitive also to morpheme frequency in applying such strategies (Diakogiorgi, Baris, & Valmas 2006). If children are guided by phonology then they should use the stress diacritic correctly, since stress is a phonological property. However, Kotoulas & Padeliadu (1999) have examined stress assignment in spelling and found it a highly problematic domain of performance, both for reading-impaired and nonimpaired children. Thus it appears that children’s phonological strategy may not encompass suprasegmental features, at least not beyond the single-syllable level.

Here we report results from 123 7^th-grade children and 114 children from primary grades 3 and 4, based on a battery of psychoeducational assessment tasks including reading and spelling of nonwords (reading only), words, and passages (Protopapas & Skaloumbakas, in press). In reading, mispronunciations were classified as phonetic (segmental) or stress errors. Spelling errors were categorized into phonological (producing an incorrect segmental sequence), grammatical (on grammatical morphemes, determined by rule), historical (dependent on etymology), and stress (omission or misplacement of stress diacritics).

Analysis of the intercorrelations among measures and examination of their structure with factor analyses separately for each age group revealed four factors in each case, interpreted as fluency, decoding, intelligence, and orthography. Loadings indicated that (a) stress assignment errors in reading both words and nonwords were most strongly related to phonetic reading errors; and (b) stress assignment errors in spelling were most strongly related to phonological, grammatical, and historical spelling errors. Therefore stress assignment closely follows other domains of reading and spelling performance.

We then selected subpopulations from each sample on the basis of their poor performance (25^th percentile) on text reading speed and spelling accuracy (excluding stress assignment for the criterion), because these measures have been found to be the most reliable indicators of reading disability in previous comparisons between the general population and clinical samples including children diagnosed with dyslexia (Protopapas & Skaloumbakas, in press), a developmental disorder of language affecting phonological representations and processes and manifesting itself primarily at the word level as slow and effortful reading. 20 primary and 21 secondary education children met the low performance criterion in both measures. These children did not differ significantly from the rest of the respective samples in nonverbal intelligence or text comprehension, but they did differ significantly in almost every measure of reading and spelling performance, including stress assignment accuracy.

The following comparisons were made on the basis of effect size estimates (partial η²) for statistically significant differences (p<.003 adjusted for multiple comparisons) using multivariate analyses of variance comparing the selected low-performing children to the rest of the sample, separately for each age group, with sex as an additional factor.

For both age groups, effect size estimates indicated that stress assignment errors in spelling were not more substantial than other types of spelling errors, and therefore should not be considered a particularly strong indicator of reading disability. This finding is not in agreement with Kotoulas & Padeliadu (1999), perhaps due to different testing materials. It seems to reflect the observed predominance of grammatical and historical spelling errors, because phonological errors (included in the group selection criterion as part of spelling performance) also produced small effect sizes.

With respect to stress assignment errors in reading, for the older children the effect sizes were very small, less than 0.1 (and in the case of pseudoword reading not statistically significant), similar to the effect sizes for phonetic errors, indicating that at this age decoding is at most a secondary indicator of reading problems, compared with the importance of text fluency. For the younger children the effect sizes for stress assignment errors exceeded those for phonetic errors and even for word fluency, suggesting that incorrect stress assignment in reading may be a potential indicator of reading disability in early primary education.

In conclusion, analysis of stress assignment in reading and writing has revealed stronger relationships within tasks (reading vs. spelling) than within linguistic domains (stress vs. segmental), and has suggested a potential role for the assessment of stress assignment performance in younger children;rsquo;s reading in the context of reading disability evaluation. Further research is necessary with diagnosed samples and with finer classification of errors.

References:
Diakogiorgi K., Baris Th., & Valmas Th. (2006). Ικανότητα χρήσης μορφολογικών στρατηγικών στην ορθογραφημένη γραφή από μαθητές της Α’ τάξης του δημοτικού [Ability to use morphological strategies in spelling by 1st grade pupils]. Psychologia: The Journal of the Hellenic Psychological Society, 12, 568–586.
Kotoulas V. & Padeliadu, S. (1999, April). The nature of spelling errors in the Greek language: The case of students with reading disabilities. In K. Nikolaidis & M. Mattheoudakis (Eds.), Proceedings of the 13th International Symposium of Theoretical and Applied Linguistics (pp. 330-339). Aristotle University of Thessaloniki.
Nunes T., Adinis A., & Bryant P. (2006). The acquisition of written morphology in Greek. In R. M. Joshi & P. G. Aaron (Eds.), Handbook of orthography and literacy (pp. 201-218). Mahwah, New Jersey: Lawrence Erlbaum.
Porpodas C. D. (2001). Cognitive processes in first grade reading and spelling of Greek. Psychologia: The Journal of the Hellenic Psychological Society, 8, 384–400.
Protopapas A. (2006). On the use and usefulness of stress diacritics in reading Greek. Reading & Writing: An Interdisciplinary Journal, 19, 171–198.
Protopapas A. & Skaloumbakas C. (in press). Computer-based and traditional screening and diagnosis of reading disabilities in Greek. Journal of Learning Disabilities.