A DataSet in Evince is used for making multivariate models. The DataSet holds information about which observations and variables to include in the model and which variables belong to data (X) and response (Y). You can also make modifications in the form of new identifiers, categories, new variables / observations or transformations. The DataSet table is by default displayed in the Table Area after importing data.

The DataSet table contains two different views: observations and variables. You can switch between the two views by using the tabs at the bottom of the table. Each view contains columns for identifiers and green bars indicating which actions affect the current observation or variable. Adjustments to the DataSet are made by changing the green bars by double-clicking on a cell to include/exclude it or by dragging the up/down arrows.

The observation view is used for overlooking observations and their properties. Each row in the table corresponds to a unique observation. The columns show properties about the observations. The first column to the left displays the observation identifiers. The columns further to the right show included and excluded observations and if they belong to the the training set or test set. If the cell for a particular observation is green, this means the observation is included. If categories are created, these will show up as yellow columns to the right of the table. If a transformation is added, it will be shown as green bars to the right of created categories.

Observations can be set to belong to the training- or test set or both.

The variable view is used for overlooking variables and their properties. Each row in the table corresponds to a unique variable. The columns show variable properties. The first column shows the variable identifiers. The columns further to the right shows included/excluded variables and whether they belong to the X or Y blocks. The columns farthest to the right show the variable datatype, categories and used scaling techniques in that particular order.

DataSet features:

- Changes are only applied to selected items.

- Double-clicking on any green cell will exclude it from the table. Double-clicking on a cell in the X/Y-column will change the corresponding variable from X to Y or vice versa.

- The green bars can be changed by dragging the up/down arrows at each end. If a single cell is excluded, it is also possible to drag the bars from the adjacent cells.

- Right-clicking anywhere on the table will bring up the table menu, where new modifications can be added.

- Right-clicking on the column index, will bring up a table header menu where it is possible to sort items or include/exclude all items.

- Observations / Variables can easily be included or excluded, either from the table menu or from the table toolbar.

- If two or more identifiers are available, it is possible to scroll between them using the bar at the bottom of the table.