The present study examined listeners' identification and rating of words in passages as stuttered when the duration and frequency of occurrence of sound prolongations were manipulated. Thirty-six participants listened to a 219-word passage containing voiceless fricatives digitally increased from their normal durations to 200, 300, and 420 ms. Listeners heard one of three passages that contained 5%, 10% or 15% altered stimuli within the passage. In Condition 1, listeners identified words considered stuttered. In Condition 2, listeners rated specifically selected words in the passage relative to the extent they considered the words stuttered. The results showed that (1) both the duration and the frequency of occurrence of the altered phonemes within the paragraph length material had an impact on listeners' perception of words identified as a sound prolongation; (2) listeners gave significantly higher ratings in Condition 2 than Condition 1 when determining if a word was stuttered or produced fluently. The implications of these results are discussed. Learning outcomes: After reading this article, the reader will be able to: (1) Describe the past literature on listener perceptions of stuttering. (2) Differentiate between listener's perceptions of sound prolongations that are altered in duration and frequency of occurrence. (3) Describe how paragraph-length speech material compares to past research that has used isolated utterances. (Contains 5 tables and 3 figures.)