Variables

 

Working with variables in StatsDirect is simple; there is nothing special to do other than entering data in a worksheet.

 

Each column of numbers that you select from a worksheet is a variable. Other statistical software packages might ask you to define names for variables; StatsDirect uses the label of the column header or the first cell (if it contains non-numeric data) of a selected block as the variable name. See the test worksheet for an example of variable names entered as text-based labels in the first cell of columns in a worksheet.

 

A variable usually contains data from observations such as a score from a questionnaire, age or sex. Variables therefore share a measurement scale. For categorical data with dichotomous (e.g. male/female) or nominal (e.g. eye colour) measurement scales, you may wish to use these data as group identifiers. An example of splitting a response variable by a grouping variable is the use of the Data_Group_Split function in a StatsDirect worksheet to split say "questionnaire score" into two columns (scores for males, scores for females) by "sex". Group identifier variables are also known as classifiers, and the term classifier has a more specific meaning in regression applications.