Speech tests as an assessment of hearing impairment : a reliability and validity study

Abstract

Introduction: The aim of the study was to investigate the reliability and validity of speech perception tests for young Maltese adults. Additionally, preliminary performance outcome trends were identified for young adults (18-35 years) with a hearing impairment. Methodology: 19 participants participated in the study. 13 participants with typical hearing levels and 6 participants with sensori-neural hearing loss (SNHL) ranging from mild to severe in severity were recruited. The study investigated test-retest, inter-rater and intra-rater reliability of the Maltese Word Test of Hearing (MWTOH), the Arthur Boothroyd (AB), the English non-word repetition test (eNWRT) and the Maltese non-word repetition test (mNWRT) in both quiet and noise. Moreover, correlational analysis was used to determine construct and concurrent validity of the aforementioned tests for both in quiet and noisy environments. Data analysis: Statistical analysis was used to compare the mean scores between T1 and T2, inter-rater and intra-rater scores. Results: Overall, no statistical difference was noted between T1 and T2, inter-rater and intra-rater mean scores. However, higher testretest reliability was noted in the clinical group (ClG; composed of participants with a hearing impairment). There were moderate to strong correlations between T1 and T2, inter-rater and intra-rater mean scores. The control group (CG; composed of participants with typical hearing levels) achieved statistically significant higher scores across all speech tests in both quiet and noisy environments than the ClG. These results suggest good construct validity. Correlational analysis showed that both the mNWRT and the eNWRT demonstrated moderate to strong positive correlation with the AB. Conversely, a weak to moderate correlation of the MWTOH with the AB was noted. Conclusions: These results indicate moderate to good test-retest reliability, inter-rater and intra-rater reliability. Weak to strong correlations were observed between speech tests for concurrent validity. However, further reliability and validity studies on a larger population is needed to strengthen the findings.