|Whenever possible, establish evaluation conditions producing data suitable for statistical as well as visual analysis.|
|Avoid response-guided experimentation methods.|
|Plot residuals to demonstrate how well your data fit a linear model; consider a nonlinear model when the residual mean deviates from zero, shows inconsistent variance, or contains autocorrelation.|
|Provide regression-based estimates of reliability in addition to agreement percentages.|
|When data are aggregated, present median rather than mean values under conditions of unusual variability or in the presence of suspected outliers.|
|Include range bars and trended ranges when possible.|
We view visual inspection and statistical analysis as complimentary tools in the development and verification of hypotheses in research and clinical practice. Visual inspection provides techniques for hypothesis generation, increased understanding of the problem, and increased ability to explore alternative explanations. Statistical analysis refines and verifies the hypotheses. Statistical models may be useful in predictions of future events and making probabilistic attributions of causality.
Visual inspection is descriptive in nature. Statistical analysis is inferential and predictive. Panaceas exist in neither, and their combined use may increase validity but never ensure it.
Allison D. B. ( 1992). "When cyclicity is a concern: A caveat regarding phase change criteria in single-case designs". Comprehensive Mental Health Care, 2( 2), 131-149.
Allison D. B., Franklin R. D., & Heshka S. ( 1992). "Reflections on visual inspection, response guided experimentation, and Type I error rate in single-case designs". Journal of Experimental Education, 61( 1), 45-51.
Armitage P., McPherson C. K., & Rowe B. C. ( 1969). "Repeated significance tests on accumulating data". Journal of the Royal Statistical Society, Series A, 132, 235-244.
Baer D. M. ( 1977a). "Reviewer's comment: Just because it's reliable doesn't mean that you can use it". Journal of Applied Behavior Analysis, 10, 117-119.
Baer D. M. ( 1977b). "Perhaps it would be better not to know everything". Journal of Applied Behavior Analysis, 10, 167-172.
Baird J. C. ( 1970). Psychophysical analysis of visual space. New York: Pergamon.
Barlow D. H., & Hersen M. ( 1984). Single case experimental designs: Strategies for studying behavior change. New York: Pergamon.
Barrios B. A., & Hartmann D. P. ( 1988). "Recent developments in single-subject methodology: Methods for analyzing generalization, maintenance, and multicomponent treatments"