A primer on common statistical errors in clinical ophthalmology

Department of Ophthalmology, New York University School of Medicine, 462 First Avenue, New York, NY 10016, USA.
Documenta Ophthalmologica (Impact Factor: 1.63). 10/2010; 121(3):215-22. DOI: 10.1007/s10633-010-9249-7
Source: PubMed


Although biomedical statistics is part of any scientific curriculum, a review of the current scientific literature indicates that statistical data analysis is an area that frequently needs improvement. To address this, we here cover some of the most common problems in statistical analysis, with an emphasis on an intuitive, tutorial approach rather than a rigorous, proof-based one. The topics covered in this manuscript are whether to enter eyes or patients into the analysis, issues related to multiple testing, pitfalls surrounding the correlation coefficient (causation, insensitivity to patterns, range confounding, unsuitability for method comparisons), and when to use standard deviation (SD) versus standard error of the mean (SEM) "antennas" on graphs.

Download full-text


Available from: Michael Bach,
23 Reads
    • "According to the first approach, all eyes will be analyzed together without taking into consideration that there are partly correlated measurements, often with high correlation between eyes (Murdoch et al. 1998). Clearly, this is a statistically incorrect way, potentially gravely distorting results (Rosner 1982; Newcombe and Duff 1987; Altman and Bland 1997; Murdoch et al. 1998; Menz 2004; Holopigian and Bach 2010), and it is " so ingrained in statisticians that this is a bad idea that it never occurs to them that anyone would do it " (Bland and Altman 1994). Nevertheless, recent reviews of articles published in different ophthalmology and optometry journals revealed that considering correlation in ophthalmic and ocular outcome data is still not at all widespread today: Karakosta et al. reviewed ophthalmology articles and found that out of 42 articles using statistical inferential techniques and including measurements of both eyes, 31 did not mention possible correlation (Karakosta et al. 2012). "
    Statistical Papers 05/2015; DOI:10.1007/s00362-015-0686-y · 0.81 Impact Factor
  • Source
    Ophthalmic and Physiological Optics 03/2011; 31(2):109-10. DOI:10.1111/j.1475-1313.2011.00823.x · 2.18 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In the decade and a half since Biswal's fortuitous discovery of spontaneous correlations in functional imaging data, the field of functional connectivity (FC) has seen exponential growth resulting in the identification of widely-replicated intrinsic networks and the innovation of novel analytic methods with the promise of diagnostic application. As such a young field undergoing rapid change, we have yet to converge upon a desired and needed set of standards. In this issue, Habeck and Moeller begin a dialogue for developing best practices by providing four criticisms with respect to FC estimation methods, interpretation of FC networks, assessment of FC network features in classifying sub-populations, and network visualization. Here, we respond to Habeck and Moeller and provide our own perspective on the concerns raised in the hope that the neuroimaging field will benefit from this discussion.
    Brain Connectivity 06/2011; 1(2):1-19. DOI:10.1089/brain.2011.0022
Show more