Biologists often use statistics as a tool to describe, compare and predict outcomes of experiments. Statistics help them understand data and test hypotheses. Biologists commonly use several types of statistical tests, but the most common include descriptive statistics, correlations, linear regressions and the analysis of variance test, or ANOVA.

Descriptives

As the name implies, descriptive statistics describe data from experiments. Commonly used values include the mean, median and variance. The mean is the sum of the values divided by the number of values. The median is the middle value. The variance represents how dispersed individual values are relative to the mean. Other common descriptive statistics include the minimum and maximum values, which also provide information about variability. Unlike other statistical tests, descriptive statistics do not determine whether groups within an experiment are distinct.

Correlation

Correlation analyses are used to determine the strength of linear relationships between two or more variables. This statistic ranges from -1.0 to 1.0, with -1.0 indicating an inverse relationship, in which one variable increases as the other decreases. Conversely, a 1.0 value indicates that as one variable increases, so does the other. Values around zero suggest that no linear relationship exists. While correlations provide information about the relationship between variables, they does not demonstrate that one variable influences another. However a zero correlation would suggest that no causal relationship exists between the variables.

Linear Regression

Linear regression takes correlation analyses one step further by determining whether one variable influences another, and how strongly. In linear regression, there are one or more explanatory variables and one dependent variable. The explanatory variables influence the value of the dependent variable and the primary components of linear regression are the intercept and the slope. The slope indicates how much the dependent variable changes as the explanatory variable changes. The intercept is not always meaningful in experiments, but tells the value of the dependent variable when the explanatory variable is zero.

Analysis of Variance

The analysis of variance test, commonly called ANOVA, is used to determine whether differences exist between experimental groups or treatments. For example, a biologist might be interested in determining whether different quantities of fertilizer lead to larger crop yields. Analysis of variance also can determine whether certain types of cancer rates differ between men and women, different pesticides are lethal to several types of birds or predator species have preferences in their prey choices.