Author Response

• Published in: Physical therapy Vol. 70; no. 9; p. 560
• Main Authors: Levangie, Pamela K, Waagfjord, Jonina, Certo, Catherine ME
• Format: Journal Article
• Language: English
• Published: 01.09.1990
• ISSN: 0031-9023; 1538-6724
• DOI: 10.1093/ptj/70.9.560

This excerpt was created in the absence of an abstract. When using single-subject designs, one must understand that accepted analyses and cautious interpretation of results can still be challenged on the basis of alternative analysis and interpretation. In our study, the indisputably marginal nature of the results renders all assessments vulnerable. We and Mr Delitto have each fired our respective statistical guns; our response now will be aimed at a closer qualitative assessment of the intervention process. As Ottenbacher 1 has emphasized, statistical analysis of single-system data should not be conducted to the exclusion of visual, and therefore more subjective, analysis. The admitted flaw in our study is the apparent trend toward improvement seen in the baseline phase. We did not dismiss this trend by noting substantial intrarater reliability values. The intraclass correlation coefficients only rule out data reduction as a possible source of the variability in the data... CiteULike     Complore     Connotea     Del.icio.us     Digg     Reddit     Technorati     What's this? Related Articles Effects of Treadmill Training on Gait in a Hemiparetic Patient Jónína Waagfjörd, Pamela K Levangie, and Catherine ME Certo Physical Therapy 1990 70: 549-558. [Abstract] [PDF] Commentary Anthony Delitto Physical Therapy 1990 70: 558-559. [Abstract] [PDF]