Basic Statistical Considerations in Virological Experiments
Year: 2004
Type: article
Abstract: Some experimental results, such as cases where mutations nearly abolish infectivity or enzymatic activity, may be sufficiently clear so that statistical methods are not needed to form a consensus among scientists regarding the conclusions. However, when the effects are more subtle, statistical methods, starting with a study design with sufficient power to detect significant differences, can be critical for drawing conclusions that are generally accepted by the research community as a whole. At as early as the experimental design stage, it is important to have statistical considerations such as sample size, power, significance level, and the appropriate statistical tests in mind so that one has a better chance of truly answering the scientific hypotheses of interest. Once data are collected, a description of the data (e.g., summary statistics) and an accurate presentation of the results of statistical testing are necessary so that readers may objectively assess the study. Finally, a detailed description of how statistical analyses were performed (preferably in the methods section of the paper) is just as important as a description of laboratory methods, because the absence of details on statistical or laboratory methods make it difficult for the reader to evaluate the quality of the study. While the basic considerations are covered here, more-complex data, study design issues, and analyses may require consultation with a statistician. As virologists continue to incorporate more quantitative and/or high-throughput methods into their research, statistical methods will become even more essential for interpreting data and drawing conclusions that others can understand. Clarity in the interpretation of experiments through proper statistical methods is the key to objectively and consistently assessing whether observed biological differences are real or due to random chance.
Source: Journal of Virology
Authors: Barbra A. Richardson, Julie Overbaugh
Institutions: University of Washington, Fred Hutch Cancer Center
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1966178745