Using statistical process control charts to study stuttering frequency variability during a single day

Hamid Karimi, Sue O'Brian, Mark Onslow, Mark Jones, Ross Menzies, Ann Packman

Research output: Contribution to journalArticlepeer-review

14 Citations (Scopus)


Purpose: Stuttering varies between and within speaking situations. In this study, the authors used statistical process control charts with 10 case studies to investigate variability of stuttering frequency. 

Method: Participants were 10 adults who stutter. The authors counted the percentage of syllables stuttered (%SS) for segments of their speech during different speaking activities over a 12-hr day. Results for each participant were plotted on control charts. 

Results: All participants showed marked variation around mean stuttering frequency. However, there was no pattern of that variation consistent across the 10 participants. According to control charts, the %SS scores of half the participants were indicative of unpredictable, out-of-control systems, and the %SS scores of the other half of the participants were not. Self-rated stuttering severity and communication satisfaction correlated significantly and intuitively with the number of times participants exceeded their upper control chart limits. 

Conclusions: Control charts are a useful method to study how %SS scores might be applied to the study of stuttering variability during research and clinical practice. However, the method presents some practical problems, and the authors discuss how those problems can be solved. Solving those problems would enable researchers and clinicians to better plan, conduct, and evaluate stuttering treatments.

Original languageEnglish
Pages (from-to)1789-1799
Number of pages11
JournalJournal of Speech, Language, and Hearing Research
Issue number6
Publication statusPublished - Dec 2013
Externally publishedYes


Dive into the research topics of 'Using statistical process control charts to study stuttering frequency variability during a single day'. Together they form a unique fingerprint.

Cite this