The “Explore” command in IBM SPSS produces an output that includes several statistics for one variable either across the whole sample or across the subsets of the sample (Kent State University, n.d.). To divide the sample into subsets while utilizing the “Explore” command, it is needed to move the categorical variable using which the sample will be split into subsets to the “Groups based on” field.
Therefore, employing a variable as a factor allows for calculating the statistics separately for each group into which the sample is divided by that variable. It often might be useful to use such variables as “gender” or “race” as factors. For instance, the variable “gender” may be used when it is needed to compare the average income that males and females earn, for which aim it is required to calculate the means, standard deviations, and other statistics separately. The variable “race” might allow for comparing the mean of some other variable across the representatives of different races in the sample. In short, variables that provide a meaningful division into groups can be used as factors.
On the other hand, variables that do not split the sample into meaningful groups are not usually used as factors. For instance, it makes no sense to use a variable such as an ID number of a participant, because ID carries no internal meaning; in addition, there would be as many groups as participants (George & Mallery, 2016).
It also makes no sense to use certain continuous variables as factors, because there would be too many groups. For instance, using an overall course grade on a 100-point scale as a factor would produce too many groups, and such a division would be meaningful. On the other hand, using the 6-point scale (A-F) would allow for a meaningful division into a manageable number of groups.
References
George, D., & Mallery, P. (2016). IBM SPSS Statistics 23 step by step: A simple guide and reference (14th ed.). New York, NY: Routledge.
Kent State University. (n.d.). SPSS tutorials: Descriptive stats for one numeric variable (Explore). Web.