Open the Fox database. Create a new column in the matrix: Data > New > Blank Variable. Name the column Group. For illustrative purposes, we will assign codes (1, 2, or 3) to the 31 samples. Assume that some process has recognized these three groups of samples. Open the sandstone icon and place it next to the Group window. Assign one of the three codes to the first sample by typing it in. Do the same for all 31 samples making sure that you have subequal numbers of each.

Shift-Click on Sandstone and on Group - Calculate - Summaries - Reports by Groups. You should have a window with the summary statistics of sandstone thickness for the three groups. Drag the Evaporite ion over the lable Evaporite on top ofthe label Sandstone - the Evaporite summary statistics for the three groups.

Click on Groups and select Plot - Bar Chart. Bar charts are used for categorical variables - categories. This illustrates how many cases are in each group. The same holds for Pie Charts. Click on Total and Group. Select Plot - Dot Plot Y by X for another represention of the distribution of total thickness in each group.Shift-Click on Sandstone, Carbonate and Evaporite. Select Plot - Dot Plot side by side.

Shift-Click on Evaporite and Sandstone and select Plot - Box Plot side by side. See page 8-13 for a discussion of box and wisker plots. The shaded interval is the 95% confidence interval about the median. If the shaded intervals do not overlap, the two medians are statistically different at the 95% confidence level.

Shift-Click on Sand and Shale. Calc - Correlations- Pearson gives you the correlation coefficient matrix. Drag Evaporite and Carbonate into the Correlation Window. You should have a 4 by 4 matrix of correlations. Select the variable pair with the largest correlation coefficient and produce a scatter plot. Click on the small triangle in the upper left hand corner of the plot. Plot the regression line. Look at the summaries of the Evaporite - Shale regression. Select Shale and Evaporite and obtain the scatter diagram and the regression information. Compare the two regression lines.