3.4.2 Contingency analysis

../../_images/ContingencyAnalysis1.png

Fig. 3.15 Contingency Analysis Widget

The contingency analysis estimates association between categorized (discrete) raster datasets. The contingency analysis computes Pearson’s contingency coefficient C and Cramer’s V.

3.4.2.1 Usage

  1. Add at least two discrete raster datasets. You can either drag and drop the rasters from your
    file manager or select them from your PC (1).

  2. If necessary change the mask raster by unchecking (5) and selecting a new one (4).

  3. Name the output file (6).

  4. Start the Analysis (7).

  5. View the results by double-clicking on the output file in the
    catalog.

If you want to remove a raster from inputs, select it and press the minus button (2).

You can view the Raster Attribute Table (RAT) of an Input raster by selecting it and pressing the RAT button (3).

3.4.2.2 Input and Output

Input

Raster datasets (.tif) with limited discrete classes.

Mask raster (.tif) providing the spatial reference and the cell size. (Default: project mask)

Output

Analysis file (.npz) Path: /results/statistics/*output name*_ctg.npz

3.4.2.3 Results

../../_images/ContingencyAnalysisResults1.png

Fig. 3.16 Contingency Analysis Results Widget

This window shows the pairwise raster comparison.

Choose between displaying Pearson’s C or Cramer’s V in the Table (2) with the combo box (1).

Clicking the button (3) opens a dialog to export the current Matrix in the Table (2) as an excel file.

Double-clicking a cell in the Table (2) opens the details for that comparison.

../../_images/ContingencyAnalysisResults2.png

Fig. 3.17 Contingency Analysis Results Widget (details)

This window shows the pairwise raster value comparison.

Choose between displaying the frequency table, Chi-squared or Phi (2x2) in the Table (3) by selecting the corresponding tab (2).

Clicking the button (1) opens a dialog to export the current Matrix in the Table (3) as an excel file.