PURPOSE: to characterize, compare and classify the performance of students from 2nd to 5th grades of private teaching according to the semiology of errors. METHOD: 115 students from the 2nd to 5th grades, 27 from the 2nd grade, 30 students from the 3rd and 4th grades, and 28 from the 5th grade divided into four groups, respectively, GI, GII, GIII and GIV, were evaluated. The tests of Spelling Evaluation Protocol - Pro-Orthography were divided into: collective version (writing letters of the alphabet, randomized dictation of letters, word dictation, non-word dictation, dictation with pictures, thematic writing induced by picture) and individual version (dictation of sentences, purposeful error, spelled dictation, spelling lexical memory). RESULTS: there was a statistically significant difference in inter-group comparison indicating that there was an increase in average accuracy for all tests as for the individual and collective version. With the increase in grade level, the groups decreased the average of writing errors based on the semiology of errors. We found a higher frequency of natural spelling errors. CONCLUSION: data from this study showed that the increase in average accuracy according to grade level may be an indicative for normal development of student's writing in this population. The higher frequency of natural spelling errors found indicates that formal instruction on phoneme-grapheme correspondence may not be occurring, since that they are directly dependent on the learning of the rule of direct phoneme-grapheme correspondence.
Evaluation; Handwriting; Learning; Education; Education Status
