Anonymizing data

It’s a good idea to transform sensitive private data before you share it. There are three possible methods: sequential, random and mapping. All three can be applied to both quantitative and qualitative data.

XLSTAT options for anonymizing data

XLSTAT allows to transform your quantitative and qualitative data according to three methods. 

Sequential

The modalities of all selected variables are replaced by a sequentially selected integer starting at 1. The number associated with a modality appears as many times as the modality appears in the dataset, so the resulting file a numerical.

Random

This method varies depending on the variable type. For a quantitative variable, the values are randomly mixed. Thus they remain in the same scale as the initial data. For qualitative variables, the modalities are replaced by a string of randomly selected characters. This string is used as many times as the modality appears in the dataset.

Mapping

Here the data is replaced by new values provided by a mapping table. This specifies the original value of the variables to be replaced in the left column and the new values in the right column.

To complete the tool, you have the option of anonymizing variable labels.

XLSTAT results for anonymizing data

Original data: this table groups together all the selected data as displayed in the datasheet.

Anonymized data: this table groups together all the data that has been anonymized. They are displayed in the same order as the original data. If the option "Anonymized variables" has been chosen, then the variable labels are presented in their anonymous form.

Mapping table: this table lists the modalities of the variables that have been transformed. In the left column the initial values are listed, and in the right column the new ones.