It is the task of descriptive statistics to identify key trends specific to the current data set. More specifically, if the researcher has a data set for analysis that the hypothesis suggests may be causally related, it is critical to understand the number of variables used for such analysis. For example, if one estimates the impact of an individual’s growth on his or her income level, only two quantitative variables are used: “Growth” and “Income.”
If, however, an age metric is added to this data, then the study is conducted for three variables, and specific trends are sought for pairwise comparisons or a three-dimensional data set. In other words, the number of variables is the primary concern that makes sense in statistical analysis.
Thus, when processing data, the researcher can deal with univariate or multivariate types of analyses. If a set contains a distribution for only one variable (e.g., “Age”), this type of analysis is called univariate. According to RGS, essential descriptive statistics functions can be applied to an array: median, mode, mean, maximum and minimum, and IQR.
It is easy to see that no relationships or causes are explored in this case: it is an ordinal fixation of the data to describe the sample. If two or more variables were used in the data set, this analysis is commonly referred to as bivariate or, in general, multivariate.
In this case, the analysis touches on the relationship between the variables and seeks to find the relationship between them: the effect of one on the other, correlation, and regression. Multivariate analysis with three or more variables is similar to bivariate analysis, but the data set includes at least one more variable, and thus the test is more complex.