Listener ratings of severity for specific disfluency types in children

Absolute and Relative Reliability of Percentage of Syllables Stuttered and Severity Rating Scales

“…These results are consistent with previous relative reliability reports for the SR scale, which used either ICC (e.g., Eve et al, 1995;Sherman & McDermott, 1958) or Pearson correlation (e.g., Aron, 1967;Curran & Hood, 1977;Lewis & Sherman, 1951). Analysis of ICC showed acceptable intrajudge relative reliability for both measures but more favorable intrajudge relative reliability for %SS scores than for SR scores.…”

The effects of duration and frequency of occurrence of voiceless fricatives on listeners’ perceptions of sound prolongations

“…The findings of the current study are also consistent with past studies that have tested changes in the perception of stuttering when manipulations of the duration and form of samples presented to listeners (Amir & Yairi, 2002;Curran & Hood, 1977;Jones et al, 2005;Kawai et al, 2005Kawai et al, , 2007Sander, 1963). Jones et al (2005) found that the minimum sound duration perceived as abnormally long or stuttered was 235 ms for the duration when sounds began with /z/ and 279 ms when sounds began with /ae/.…”

“…Research on listeners' perceptual labels of normally fluent and stuttered speech began several decades ago. Sander (1963) and Curran and Hood (1977) demonstrated that listeners' perceptions could be manipulated through small changes in the speech sample that would create a shift in the perception of words from fluent to stuttered. Specifically, Sander changed part-word repetitions from one-to two-units per repetition (e.g, Sa-Saturday to Sa-Sa-Saturday) and found that it took only half the instances of double unit than single unit repetitions in the speech sample for listeners to judge the speech sample as stuttered.…”

Self-Rating of Stuttering Severity as a Clinical Tool

“…Early studies by Sherman and colleagues (Lewis and Sherman, 1951;Sherman, 1952Sherman, , 1955Sherman & Trotter, 1956) evaluated a 9-point scale and found this to be a reliable tool for assessing stuttering severity. Later studies highlighted the fact that whether a scale had 5, 7, 9, or 15 points seemed to make little difference to mean scale values or reliability (Cullinan, Prather, & Williams, 1963;Curran & Hood, 1977). Further, there is little difference in scores when points on the scale are clearly defined (Cullinan et al, 1963), when participants are given repeated exposure to the task or feedback about group mean scores (Young, 1969a(Young, , 1969b, or when live as opposed to recorded samples are used (Cullinan et al, 1963).…”