When researchers conduct a study into a clinical topic such as a screening test or disease treatment, the raw experimental results are interpreted using statistical analysis. Statistics is the branch of mathematics involved in collecting, analyzing, and presenting data, and making predictions and recommendations using those data.
Statistics are an essential part of evaluating clinical research. Recently, an article in Psychological Science for the Public Interest (Gigerenzer et al., 2008) explored the implication of ignorance of statistics both in the general public and among media and health professionals. Lack of knowledge of this field has a negative impact on public health. This series presents basic information about statistics and explains examples taken from the article.
Basic Principles of Statistics
Given a set of data, a number of measures can be derived that describe the set. These are called descriptive statistics and include values such as the mean and median, which show an average of the data, and the standard deviation, which describes how much variability is in the data.
In a clinical study, the results are never clear-cut. Theoretically, there is always a possibility that the observed data are the result of random chance. Using statistical methods, the researchers calculate the likelihood that the results are the result of chance. They determine in advance a threshold that this probability, abbreviated p, must reach, typically 5%, 1%, or 0.5% (expressed as p = .05, p =. 01, or p = .005, respectively). If p equals, say, 0.04, that means there is only a 4% probability that the results were due to chance, and therefore a 96% probability that a real correlation has been observed. A p value less than the threshold value is called a statistically significant result.
This type of analysis is called applied statistics. The pieces of information being analyzed may be raw data (such as blood test values) or they may be values derived from descriptive statistics.
Benjamin Disraeli said, "There are three kinds of lies: lies, damned lies, and statistics." Statistics are often used, both unintentionally and intentionally, to mislead and confuse. Much of statistical thinking is counter-intuitive, and education is necessary to interpret it properly. Without that education, people can be led to false conclusions via the manipulation of statistics, either intentional or accidental. The phenomenon of most of the population lacking this statistical education is called collective statistical ignorance, and not only patients, but also many doctors and researchers suffer from it.
Gigerenzer, G.; Gaissmaier, W.; Kurz-Milcke, E.; Schwartz, L. M.; and Woloshin, S. 2008. "Helping Doctors and Patients Make Sense of Health Statistics" (PDF). Psychological Science for the Public Interest 8:2 (53-96).
This post is part of the series: Statistics Sense
- Statistics Sense, Part 1: Making Sense of Medical Statistics
- Statistics Sense, Part 2: Confusing and Misleading Statistics
- Statistics Sense, Part 3: Comparisons and Transparency