gms | German Medical Science

Many children with cochlear hearing impairment demonstrate deficits in phonetic discrimination, phonological perception and auditory short-term memory [Ref. 1], [Ref. 2]. Speech perception deficits are positively correlated with degree of hearing loss and had a negative impact on the acquisition of spelling abilities [Ref. 3]. However, most studies on the development of writing skills use children with a hearing impairment of 70 dB or more.

The objective of our study was to evaluate the writing skills of hearing impaired children. In addition we also wanted to compare spelling skills of hearing impaired children attending grammar school with hearing impaired children attending special schools. We wanted to find out whether one could predict the magnitude of hearing impairment by the kind of spelling errors the children made.

The sample comprised 38 children (19 boys/19 girls) with cochlear hearing impairment. The mean age of the children was 10 years. Primarily all children were fitted for hearing aids in our clinic. None of them had acute progressive hearing loss. The degree of hearing impairment on the better hearing ear was determined by the mean of the hearing loss at 500 Hz, 1 kHz and 2 kHz. The hearing loss varied between 26 dB to 90 dB with a mean hearing loss of 41.7 dB (see [Fig. 1]).

At the time of the study 22 children attended grammar schools and 16 children special schools for the hearing impaired. As expected, results demonstrated that the mean hearing loss of the children attending special schools was significantly higher than that of the grammar school children. The children were given standardized word fill-in dictation tests at the end of 2^nd/beginning of 3^rd grade [Ref. 4]), at the end of 3^rd/beginning of 4^th grade [Ref. 5], or at the end of 4^th/beginning of 5^th grade [Ref. 6]. The fill-in words were read out loud and the children had direct face-to-face interaction with the experimenter.

Figure 2 [Fig. 2] presents the over-all results of the spelling tests. T-values under 40 represent spelling performances below average and T-values of 60 or more above average spelling performances. Moreover, the results were extremely diverse whereby half of the participants (n=20) displayed below average and some extremely poor spelling skills. Four children performed very well on the spelling tests.

The results of the group comparison of grammar school children versus special school children was highly significant (p<0.0002). Grammar school children had a mean T-value of 50.1, whereas special school children attained a mean T-value of 37.1. Performance differences became even more apparent when we compared the type of errors made by each group.

Table 2 [Fig. 3] presents the most important error types made by each group. The respective error rate in the DRT 3A was calculated with a standardized error analysis and we used the error frequencies to determine T-values. Most children attending special schools demonstrated a significantly higher error rate. Twelve grammar school children did not exhibit any particular error classifications in comparison to the special school children who made significantly more errors in "sequential processing", "capitalization", as well as the incorrect application of the spelling rule for the "st"-grapheme

The research with standardized spelling tests on hearing impaired children attending grammar school shows great differences in the spelling performances. The percentage of below average performance of the children in our study was quite high - more than 50 percent reached T-values under 40. Children attending grammar school performed significantly better on spelling tests than the children attending special schools. However, the special school children had greater hearing impairment. We were surprised by one result of the qualitative error analysis which showed that our grammar school children did not make any special kinds of errors which we expected due to their perceptional speech deficits. Maybe these children have learned to compensate their receptive speech deficits with the utilisation of efficient complex verbal and cognitive information processing. Children in this study with greater hearing impairment (special school children) were unable to compensate in a similar manner. These particular children made significantly more sequential processing errors (e.g. omitting graphemes especially when important morphological verbal attributes were available). The greater error frequency can be explained with peripheral hearing impairment.

The results of this study indicate that the development of writing skills in hearing impaired children demonstrate an interesting research area in which findings can contribute answers to pedagogical problems as well as facilitate further development of even more accurate developmental spelling skill models (such as the importance of auditory perception).