FluencyBank Derived Warner Corpus

Haley Warner Communicative Sciences and Disorders New York University haley.warner@nyu.edu

Eric Jackson Communicative Sciences and Disorders New York University eric.s.jackson@nyu.edu

Users of this dataset should cite: Warner, HJ, Shroff, R, Zuanazzi, A, Arenas, RA, Jackson, ES (2023). Linguistic features of stuttering during spontaneous speech. Journal of Fluency Disorders, 78, 106016.

Words from the Voices of Adults Who Stutter interviews were coded at the word level using a binary coding system (i.e., stuttered or non-stuttered).

Our reliability process: Stuttering events were coded within two research teams across two sites: New York University (site 1) and The University of New Mexico (site 2). During the first round of coding, two independent raters from each site coded every word. At site 1, coding disagreements were reconciled by an additional researcher who had not participated in the initial coding round (Haley J. Warner). At site 2, coding disagreements were reconciled between the two raters. Interrater reliability between sites was assessed using Cohen’s kappa statistic (McHugh, 2012). Reliability was calculated on a random 30% sample of the dataset (10,452 words) and indicated substantial agreement (κ = .732). All words lacking agreement between the two sites (n = 1,749) were independently re-assessed by three additional researchers from site 1. Words that were still lacking agreement across raters (2.79% of the dataset) were then independently reviewed and reconciled until agreement was reached by the laboratory directors at both sites, Dr. Eric S. Jackson and Dr. Richard M. Arenas.