Exploring Methodological Decisions for Calculating the Minimally Detectable Change in Dysarthria: Reliability, Statistics, and Standard Error of Measurement

Purpose: The minimally detectable change (MDC), widely used in rehabilitatiosciences to interpret changes in outcome measures, is calculated using a reability method, reliability statistic, and standard error of measurement (SEMThis study examined how different methodological choices affect MDC thresh-olds of speech intelligibility in speakers with dysarthria. The goals of this stuwere to compare MDCs calculated using (a) three different reliability method(b) two different reliability statistics, and (c) three different SEM calculations. Method: Recordings of the Speech Intelligibility Test from 200 speakers incluing speakers with amyotrophic lateral sclerosis (n = 16), Huntington’s diseas(n = 44), multiple sclerosis (n = 60), and Parkinson’s disease (n = 40), along healthy controls (n = 40), were drawn from two databases. Thirty inexperienclisteners completed two sessions, providing orthographic transcriptions of 20 speakers. MDCs of intelligibility were calculated using (a) three reliabilitmethods (i.e., test–retest, split-half, and intrarater), (b) two reliability statistic(i.e., Pearson r and intraclass correlation coefficients [ICCs]), and (c) three diffeent formulas for calculating the SEM. Kruskal–Wallis tests were used to assethe effects of reliability methods, statistics, and SEM calculations. Results: Significant differences were found between the MDCs when usinsplit-half and test–retest reliability, when using Pearson r and ICC, and whusing two of the three SEM calculations. Conclusions: Results demonstrate that methodological decisions can impact MDCs of speech intelligibility in speakers with dysarthria, highlighting the neefor specific, detailed reporting of methodology used to calculate MDCs in futuwork. Findings can provide methodological guidance for future studies and cotextualize existing research on intelligibility changes.