Search

Perceptual rating scales can be valid, reliable, and convenient tools for evaluating speech outcomes in research and clinical practice. However, they depend on the perceptions of observers. Too few raters may compromise accuracy, whereas too many would be inefficient. There is therefore a need to determine the minimum number of raters required for a reliable result. In this context, the ideas of Generalizability Theory have become increasingly popular in the behavioral sciences; suggestions have been made for their application to the assessment of speech-language disorders. Here we review the concepts involved, which are applied in a companion article dealing with speech naturalness data obtained from clients who recently completed treatment for their stuttering. We pay particular attention to the statistical requirements of the theory, including some cautions about possible inappropriate use of these techniques. We also offer a new interpretation of the results of the analysis that aims to be more meaningful to most speech-language pathologists.

This article reports the development of a tool for measuring how comfortable a person feels when communicating with someone who has undergone treatment for stuttering. The person rates the speaker on a 9-point Listener Comfort Scale (9 = extremely comfortable, 1 = extremely uncomfortable). In a preliminary investigation of the reliability and validity of the scale, 15 unsophisticated listeners rated video recordings of 10 adults before and after a prolonged-speech treatment for stuttering and of 10 matched controls. The results were compared with those of another 15 listeners who rated the same recordings with the widely used 9-point Speech Naturalness Scale (R. R. Martin, S. K. Haroldson, & K. A. Triden, 1984). Results showed that reliability of the Speech Naturalness Scale was superior to the Listener Comfort Scale, although users of both scales were able to distinguish between pretreatment speech, posttreatment speech, and the speech of controls. The results suggest that the Listener Comfort Scale captures information that is somewhat different than the information captured by the Speech Naturalness Scale. The authors concluded that the concept of listener comfort is a potentially useful additional way of investigating the social validity of behavioral treatments for stuttering.

Background: The benefits of treating stuttering close to onset have become obvious in recent years, and the Lidcombe Program has emerged as an effective and safe treatment method for children in their preschool years. The benefits of implementing the programme with young children, however, need to be weighed against the knowledge that many children recover from stuttering without treatment. In light of this, speech-language therapists need to know how long treatment is likely to take and whether treatment time can be predicted. In particular, they need to know if adopting a 'watchful waiting' approach--to see if natural recovery occurs--jeopardizes responsiveness to treatment. A recent Australian study of 250 preschool-age children found that stuttering rate was the only significant predictor of treatment time with the Lidcombe Program. In other words, children whose stuttering was more severe took longer to pass through the programme. There were other trends in the data but they did not reach significance., Aims: The present study, conducted independently in the UK, was designed to replicate the Australian study. Direct replication enabled pooling of the data from the two studies in a meta-analysis., Methods & Procedures: The study included 66 children who began treatment before 6 years of age. They were treated with the Lidcombe Program at a specialist stuttering clinic in Norwich. Logistic regression analyses were conducted on the data. The data from both the British and Australian cohorts were pooled in a meta-analysis., Outcomes & Results: Results indicated that Stage 1 of the Lidcombe Program was completed in a median of 11 clinic visits, which is in line with the findings of the Australian study. Stuttering rate at first clinic visit was again found to be a significant predictor of treatment time. The remaining data trends were similar to those in the Australian data. In the meta-analysis, stuttering rate was once more found to be a predictor of treatment time. Of particular interest, however, was that the increased power provided by the meta-analysis identified an additional predictor, namely onset-to-treatment interval. Contrary to what is known about the responsiveness of children to the Starkweather and Gottwald treatment, and contrary to what might be expected given what we know about natural recovery, children who had been stuttering for more than 12 months took less time to progress through the programme than children who had been stuttering for less than 12 months., Conclusions: These findings indicate that delaying intervention with the Lidcombe Program for 1 year after onset, within the preschool years, is unlikely to jeopardize responsiveness to treatment. The clinical implications of these findings are discussed.

Preliminary Phase I and II trials for the Lidcombe Program of early stuttering intervention have found favorable outcomes and that the treatment is safe. Although speech-language pathologists (SLPs) often need to intervene with pre-schoolers' early stuttering, many of these children will recover at some time in the future without such intervention. Consequently, they need to know whether the Lidcombe Program's effect on stuttering is greater than that of natural recovery. Participants were 23 pre-school children who were randomly assigned to either a control group or a treatment group that received the Lidcombe Program for 12 weeks. A repeated measures ANOVA showed no main effect on stuttering for the group (control/treatment), a significant main effect for the measurement occasion (at the start and at the end of the treatment period), and a significant interaction between group and measurement occasion. Stuttering in the treatment group reduced twice as much as in the control group. These results are interpreted to mean that the introduction of the Lidcombe Program has a positive impact on stuttering rate, which exceeds that attributable to natural recovery.

The capacity to make reliable inductive statements about populations is critical for the advancement of scientific knowledge. An important contribution to that advancement of knowledge is determining that effects are either present or not present in populations. Statistical power is an important methodological qualification for any research that presents statistical results, and particularly so for research that presents null results. In this paper we describe the statistical concept of power, outline parameters of research that influence it, and demonstrate how it is calculated. With reference to selected published research, attention is drawn to the problems associated with a body of underpowered research, one being that population effects may go undetected. One way to prevent this problem is to calculate power a priori in planning research and include confidence intervals when presenting the results of research. However, it is difficult, if not impossible in many cases, to obtain high participant numbers for communication disorders of low prevalence such as stuttering. With this in mind, the paper concludes with an attempt to open discussion about ways to redress the problems associated with statistical power.

It is known that operant treatments can control stuttering in children. However, at present it is unknown why such treatments are effective. Changes in the usual way of speaking are frequently observed after behavioural treatments for adults who stutter, and it is possible that operant treatments for children also invoke such changes. To explore this idea, selected acoustic measures of speech timing were made in eight preschool children before and after receiving the Lidcombe Program, which is an operant treatment for stuttering. No systematic changes were detected after treatment. Considering this finding and a previous report, there is no evidence to suggest that the reductions in stuttering that occur with this treatment are related to systematic changes in speech timing or curtailment of language function.