CONCLUSION
The present study has considered the correlation between reading and spelling proficiency. A range of positive correlations in line with other studies was found; the range is explained by differences in tests of reading proficiency and individual student differences. Misspellings were analysed as evidence of direct, visually mediated and indirect, phonologically mediated word recognition strategies. It was found that both strategies were employed, and that, although not quantifiable, direct access is significant and possibly preponderant and indirect access is associated with vocalic errors. This was expected in view of the formal nature and status of Arabic orthography and the transfer of reading strategies across languages and their orthographies reported in the literature.
The misspellings demonstrate that, although English orthography invites flexible use of direct and indirect access of words and that both are employed by the present students, these strategies are often unsuccessfully applied. The methodological implication is that two problems must be addressed. First, explicit teaching of vocalic grapheme to phoneme correspondences, accompanied by extensive practice in vocalic phoneme discrimination, might be considered. Second, the visual concepts of �word� in English and Arabic orthography should be maximally separated, with emphasis placed on spacing between letters and words in English. To this end, I strongly recommend the simple expedient of never allowing an Arabic student of English, or his teachers, to read or write cursive script, and where feasible use should be made of word-processors.
The study suggests that misspellings, which are convenient to collect, can be used as evidence of reading strategies employed by less proficient readers at the word level. Further research could study the extent to which individuals produce misspellings inspired by one rather than the other of the two word recognition processes so that individualized learning programmes could be designed.