Browsing by Subject "online test"

Speechreading is a technique of understanding speech by interpreting visual information from the speaker’s face. Speechreading is a skill that can be trained and to determine the need for rehabilitation and monitor the progress a reliable assessment tool is needed. In Finland, there does not exist any online assessment tool that can be used to assess sentence-level speechreading skills in adults. This study aimed to find out how well the 100 CID Everyday Speech sentences used in the assessment of speechreading skills are suitable for the assessment of speechreading skills when translated into Finnish. This thesis examined how Finnish-speaking normal-hearing adults aged 18–65 perform sentence-level speechreading tasks, whether sentence length affects performance, and whether CID sentences differentiate speechreading skills when translated into Finnish. Study participants (n = 38) were normal-hearing adults aged 18–65 with no reported learning disabilities. The data was collected from an online speechreading test responses and analyzed by statistical methods with the R programming language. The translated CID sentences were divided into ten collections of ten sentences and comparisons were then made between those collections. Participants’ performance was examined using the Friedman test and post-hoc testing was done with the Durbin-Conover test. The consistency in the responses between collections was assessed using Spearman’s correlation coefficient. Sentences were also divided into groups by word count and both the Friedman test and the Durbin-Conover test were used to examine these groups. The translated CID Everyday Speech sentences differentiate speechreading skills well. Statistically significant, moderate, or strong positive correlations were found in the responses between the sentence collections. The word count of the sentences was found to be connected to the percentage of words recognized correctly. Based on the results the translated CID sentences are suitable for assessing the speechreading skills in normal-hearing adults. Based on the results it could also be possible to divide the collection of 100 sentences into shorter collections, which could for example be utilized in monitoring speechreading rehabilitation.