Developmental changes in the spelling of derivational suffixes by typically developing Greek children: effects of transparency, lexicality, letter length and frequency

ABSTRACT

The study aimed to investigate developmental changes in the spelling performance of derivational suffixes by 188 Greek school children (ages: 7–14 ½) of typical development, along with effects of transparency, lexicality, length and frequency. Children were evaluated preliminary by a standardised spelling task followed by three experimental spelling tasks of derivations constructed for the study. In particular, the first task entailed 88 related pairs of derivations and pseudo-derivations matched in terms of phonological transparency of their suffixes (transparent vs. opaque). The second task entailed 40 derivationally related words classified in terms of suffix length (one- vs. two- vs. three-syllables), while the third task comprised of 20 related pairs of derivations equally allocated into high and low frequent items in terms of suffixation. Results showed that developmental differences of the derivational acquisition were significant in the first two years of schooling. Accordingly, the effects of phonological transparency, length and frequency of derivational suffixes were most prominent in the first years of schooling than in the later years. These results are compatible with the experimental literature and have important implications for designing adequate assessments of orthography for typically developing children or those with spelling disabilities or dyslexia.

KEYWORDS:

• Spelling
• derivational morphology
• suffixes
• typical spellers
• shallow orthography

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 According to developmental models of literacy acquisition (Seymour & Duncan, 2001), spelling is acquired in phases, as follows: (i) Preliteracy, where children have not yet started to read but they may be competent of the linguistic structure of words in oral language either implicitly or explicitly; (ii) Foundation, where children begin to acquire the alphabetic principle by which the phonemes of speech may be represented by letters/letter groups; (iii) Othographic, where children start to build the syllabic system of the language based on word-specific memory for the exact letter choices to particular words; (iv) Morphographic is based on children's knowledge of conventions for representing bound morphemes (prefixes, inflections, derivational suffixes) and combining them with word stems.

2 For instance, the derivational suffix <-ωση> /-osi/ attached on verbs builds a large family of 195 words, i.e., <βελτίωση> /veltiosi/.

3 The allocation of students per grade is the following: (i) Grade 1 (n = 17, males: 9, females: 8) with a mean chronological age of 6.27 years (SD = 0.41), (ii) Grade 2 (n = 16, males: 8, females: 8) with a mean chronological age of 7.36 years (SD = 0.45), (iii) Grade 3 (n = 19, males: 10, females: 9) with a mean chronological age of 8.53 years (SD = 0.48), (iv) Grade 4 (n = 14, males: 9, females: 5) with a mean chronological age of 9.34 years (SD = 0.43), (v) Grade 5 (n = 12, males: 6, females: 6) with a mean chronological age of 10.42 years (SD = 0.47), (vi) Grade 6 (n = 13, males: 6, females: 7) with a mean chronological age of 11.36 years (SD = 0.45), (vii) Grade 7 (n = 29, males: 14, females: 15) with a mean chronological age of 12.60 years (SD = 0.47), (viii) Grade 8 (n = 30, males: 17, females:13) with a mean chronological age of 13.56 years (SD = 0.45), (ix) Grade 9 (n = 38, males: 20, females: 18) with a mean chronological age of 14.67 years (SD = 0.46).

4 According to Porpodas (1999, p. 408), ‘orthographical exceptions were considered to be words in which an inconsistency occurred in terms of different letters or letter combinations being used for the same phoneme.’ These inconsistencies are responsible for children’ spelling errors (see, Appendix 5), as for instance the phoneme /o/ could be written as <o >/ <ω>. Thus, items with <ο> categorized as transparent, items with <ω> as non-transparent. The phoneme /u/ is consistently written as <ου>, thus categorized as transparent. The criterion of consistent one-by-one phoneme-grapheme correspondences was chosen to categorize suffixes as transparent/non-transparent as the most sensitive to capture differences for a highly shallow orthography (spelling: 80% consistency), such as Greek.

5 Value of 100 was considered as a suitable criterion of frequency for this set of suffixes ranged from 5.840 to 1 to secure a large difference between the two categories.

6 HF items = 5 consistent/5 inconsistent, mean word length = 3.6, LF items = 4 consistent/6 inconsistent, mean word length = 3.2 (see, Appendix 4).