| Summary: | Understanding of inferences and non-literal language, such as idioms, is critical for successful communication and academic learning. Assessment of inferential and idiom comprehension is essential for children who are showing difficulties in these areas so that appropriate intervention and support can be provided. There is very little information on the typical development of these specific areas of verbal comprehension in the literature and there are very few current assessments of inferential and idiom comprehension for British school-aged children. While many assessments that do exist have face validity, very few are standardised. Some children with comprehension difficulties do well on existing picture-based assessments of verbal comprehension but they demonstrate significant difficulties with more abstract language comprehension. There is a gap in the current battery of assessments available to paediatric speech and language therapists for assessing inferential and idiom comprehension in detail. The primary aims of this thesis were to develop a robust standardised British assessment of inferential and idiom comprehension for 5:00 to 9:11 year-old children, to provide supporting validity and reliability data for the newly devised assessment, to provide normative and statistically significant data for inferential and idiom comprehension in typically developing children aged 5:00 to 9:11, to provide qualitative information on the typical development of these areas of verbal comprehension and to carry out exploratory studies using the new assessment with children with communication impairments. Secondary aims of the study were to examine if there was any relationship between gender and test performance and between socio-economic status and test performance. A new assessment, the Hewitt Inferential Comprehension and Idioms Test (HICIT) was created following a review of the literature and of the existing assessments in these areas of verbal comprehension. A pilot study was carried out with sixty-two 4:06 to 10:06 year-old children in two primary schools in the North West of England. The fourteen section, 210 item pilot test was reduced to the twelve section, 140 item final version of the HICIT. This was then trialled with a further 200 children, aged 5:00 to 9:11, making an overall standardisation sample of 250 children. Normative data were provided from the application of descriptive and inferential statistics to the results. A two-way ANOVA examined the effects of age group and gender on test scores and a one-way ANOVA and post-hoc independent T tests looked at the relationship between socio-economic status and test performance. The construct validity of the test was examined with exploratory and confirmatory factor analyses. These demonstrated a single factor loading and good model fit measurements. The reliability of the test, as assessed by Cronbach’s Alpha was moderate, similar to an existing published British standardised verbal language assessment. Rasch analysis indicated that the internal consistency of the test was good. The inter-rater reliability of 98.6% was excellent. The descriptive and inferential statistics demonstrated that there was a developmental progression between the age groups for inferential and idiom comprehension but that there was no effect for gender. The idioms sub-section was the only section not to reach ceiling scores by 9:11. The results for the relationship between socio-economic status and test scores were inconclusive. Quantitative analysis of the HICIT data demonstrated that the test is a robust assessment of inferential and idiom comprehension. Some sub-tests of the test are more robust with different age groups so different versions of the test could be used with different age groups. Qualitative analysis of the test responses, exploratory case studies using the HICIT to assess four children with communication impairments and feedback from practising speech and language therapists produced useful information on the possible applications of the test and suggested that it could be very useful to assist in differential diagnosis of different types of communication impairments.
|
|---|
