A inteligibilidade e a gravidade do desvio fonológico julgadas por três grupos de julgadores

Abstract

Communication problems may happen by a phonological deviation present in the speech of some children during the sounds acquisition and it may cause a limited phonetic inventory besides a simplified phonological system that will lead to a decrease in intelligibility. This study aimed at determining the correlation between intelligibility and the phonological deviation severity since the analysis of three different groups of judges. The research consisted of two samples; one sample to be assessed (subjects with phonological deviation) and another sample that comprised the judges (speech therapists, laypeople and mothers). The first sample was composed of 30 subjects, 10 female (33,3%) and 20 male (66,7%), average age 4:1 to 7:11, who presented phonological deviation determined by speech therapy, phonological tests and complementary assessments developed in the Program of Speech Therapy Assistance at UFSM. The second sample comprised the judges and it was consisted of five laypeople, five subjects with a background in the field (speech therapists and seniors from the Language Therapy graduation course at UFSM) and five subjects selected from the children s families (mothers). The speech of the subjects presenting phonological deviation was transcribed, analyzed and classified after nominalization, imitation and spontaneous speech tests. Then, it was determined the phonological deviation severity according to the Percentage of Correct Consonants (PCC) proposed by Shriberg & Kwiatkowski (1982). It was also carried out a narrative test through which it was obtained a sample of spontaneous speech that was analyzed by the judges. The pieces of narrative were drawn to be presented to the judges randomly, ranging from Speech 1 to Speech 90, accompanied by some questions of identification and description of the judges, together with a chart for the registration of intelligibility and levels of severity. Following it, it was carried out the mode of 90 pieces of narrative, which helped in the statistical analysis of the data based on Concordance Analysis by Kappa and Spearman s Correlation, using the statistics program STATA. It was verified speech intelligibility and the phonological deviation severity assessed by the group of judges, as well as the concordance between these variables. Finally, it was examined the correlation between the severity index according to PCC and the values presented by each group of judges. Regarding intelligibility, it was widely classified as regular by all groups, and it was noticed that the speech therapists were more tolerant in their assessment. It was observed that the group of speech therapists and mothers used more the category average in their assessment of severity while the laypeople group used more the category moderate-severe . There was more concordance among the groups of judges concerning the extremes in the judgment of intelligibility (good and insufficient) and also of severity (average and severe). Considering the comparison among the results by the groups of judges; mothers and laypeople presented almost the same evaluation to intelligibility, and the level of concordance was higher to intelligibility evaluated as good. In terms of severity, the similarities among the groups were stronger in the extremes (average and severe); whereas among speech therapists and mothers it was almost perfect to average and severe levels. The concordance among all the groups was substantial to the assessment of good intelligibility as well as of average and severe levels of phonological deviation severity, being more difficult the judgment and classification of mild-moderate and moderate-severe severity as well as regular intelligibility. It was found a higher correlation between the severity and the severity in PCC by speech therapists and mothers, showing that laypeople face more difficulties in presenting a precise assessment of the phonological deviation severity. The correlations between the intelligibility of the speech and the phonological deviation severity by all the groups were positive and similar to the real indexes for all the groups of judges. Therefore, it was easy to the judges analyzing and correlating the assessments according to intelligibility and severity in the narratives of the subjects. It was observed that the more unintelligible the speech was, the more severe was its classification by the groups of judges.