Analysis of Variance (ANOVA) is one of the most popular statistical methods used in fields such as psychology, biology and business. ANOVA is a method of separating observed variance data into different components in order to gain further insight and deeper understanding of relationships between variables. Through this method of analysis, it is possible to compare the influence of variables like sex, age or treatment on a dependent variable like height or yield.

When deciding to use ANOVA, the researcher must choose between a one-way ANOVA or a two-way ANOVA, depending on their data set. A one-way ANOVA is used when the researcher wishes to compare three or more groups of data. The researcher specifies one independent variable with different levels which are compared between groups. This type of ANOVA is used to accurately measure the relationship between the dependent variable, to determine whether the groups differ from one another, and the independent variable.

The independent variable is broken down into different levels by the researcher and what remains is the sum of all the variations within the levels. The ANOVA calculates the F-ratio (variance between group mean over variance within group mean) to determine the extend of systematic variability of the independent variable—if the F-ratio is equal to or close to 1, there may not be any true variance between the groups.

In contrast, a two-way ANOVA measures the interaction between two independent variables on say a dependent variable, and is helpful in understanding the effects of one independent variable on the dependent variable when the levels of another variable is held constant.

ANOVA tests are conducted in order to understand the causes of variance and any differences between data points. ANOVA is used to measure the relationship between variables, to test if variability in the dependent variable is due to a particular levels of the independent variables, and to compare the differences between two or more groups.