Abstract
Background: Clinicians rely on parents to provide information regarding the onset and development of stuttering in their own children. The accuracy and reliability of their judgments of stuttering is therefore important and is not well researched.
Aim: To investigate the accuracy of parent judgements of stuttering in their own children's speech when compared with judgements made by the parents of normally fluent children and experienced clinicians.
Methods & Procedures: Twenty parents of children who stutter made judgements of stuttering during repeated presentations of 3‐min audio‐visual samples of their children. Twenty control parents of children (age and gender matched) who do not stutter also assessed the children who stutter speech samples.
Outcomes & Results: The parents of both the children who stutter and those who do not stutter displayed high levels of judgement accuracy (parents, of children who stutter percentage agreement mean (standard deviation) = 90.9 (6.02); parents of children who do not stutter = 86.4 (9.7)) and consistency (90.3–90.6%) in judging stuttering in young children. But the parents of children who stutter showed significantly higher levels of accuracy (t(59) = 4.63, p<0.0001; d = 0.8) in judging stuttered intervals than the children who do not stutter parents.
Conclusions & Implications: Parents of children who stutter may be both accurate and reliable in identifying brief intervals of speech containing stuttering and non‐stuttering in their own children.
Notes
1. For instance, Yairi et al. (Citation1993) report an inconsistent correlation between parent and clinician ratings on an eight‐point scale; the correlation was significant (but far from perfect) at the 3‐ and 6‐month visits (r = 0.85 and 0.46, p<0.05), but not for the initial visit (r = 0.40, p>0.05).
2. In the Lidcombe programme the parents' severity rating on a ten‐point scale is the principal measure of their child's stuttering during the treatment. A comparison between a nine‐point severity scale and %SS scores suggest that these are generally reliable and similar measures of stuttering in adults (O'Brian et al.Citation2004) — they were, however, poorly related when there was a low frequency of stuttering.
3. According to the parents' reports, the time since onset was, on average, over one year (1:1 year, range = zero to 3:2 years). For five children stuttering reappeared after having been absent for between a few weeks and 1.6 years. The parents reported that for six other children their stuttering had been episodic. Only one child had been stuttering consistently for almost three years.
4. Each speech and language therapist viewed the 20 edited videotapes individually and identified at least one unambiguous stuttering event in each child's 3‐min speech sample.
5. One of the initial aims was to investigate the need for training the parents to improve the accuracy with which they identify occurrences of stuttering. But given that the parents already possessed a satisfactory (above 90%) level of judgement accuracy, the proposed training programme became redundant. Nevertheless, that conclusion would be more supportable if training did not produce any additional level of accuracy. Therefore, a test was made to see if this was the case by giving the parents the Icelandic version of the SMAAT training programme (SMAAT‐child) (Einarsdóttir and Ingham, forthcoming Citation2007). Ten randomly selected CWS parents were trained to judge stuttering events using SMAAT‐child and their judgements were compared with those made by ten untrained CWS parents. The training programme proved not to have significant main effects on improving the accuracy with which total Agreed intervals or combined Agreed Stuttered and Non‐stuttered intervals were judged. The effect of the SMAAT training has previously been investigated with adults: here judgement accuracy improved from 80–85% before training to 90–91% after training (Cordes and Ingham Citation1999). In the present study, therefore, the training programme was likely contending with a ‘ceiling effect’: the CWS parents were, on average, as accurate before and after training as the judges in the study by Cordes and Ingham after training. Obviously, the effectiveness of the training programme might have been identified more clearly if the present study had included only parents who were especially poor in recognizing occurrences of stuttering.