This is a brief tutorial on the use of all of the statistical routines included in this version of STATsimple. Data for this tutorial are located in the folder labelled "Samples".
Open the STATsimple application and follow the sections below.
Click the "OPEN" icon in the Toolbar Area. A standard Open File dialog box will be presented. Locate and open the "Samples" folder. You will see two sub-folders with the names "Heating Study" and "Temperature Measurement". Open the "Heating Study" folder. You will then see four data files listed. Open the file "Apr–Sep Temp. (°C)". Click the "OPEN" icon again and open the file "Apr–Sep Heating ($)". Repeat this action and open the files "Oct–Mar Temp. (°C)" and "Oct–Mar Heating ($)".
From the "Format" menu, select "0.00" to set the number precision to two decimal places. The data shown in the Viewing Area will now be displayed with a precision of two decimal places.
If the "EDIT" icon in the Toolbar Area is not already selected, then click it now. The Viewing Area shows the data contained in the currently selected sample. The currently selected sample is displayed in the Samples Area with a small flag to its immediate right. Click on the other samples in the Samples Area to see their data.
Click on the "DESC" icon in the Toolbar Area. The Viewing Area shows some descriptive statistics of the currently selected sample including the Mean, Standard Deviation and Standard Error of the Mean (SEM). Click on other samples to see their descriptive statistics.
Click on the "HIST" icon in the Toolbar Area to view a histogram plot. Initially, STATsimple will estimate the lower and upper plotting limits as well as the number of classes. To set your own plotting parameters, double-click anywhere on the plot and you will be presented with a dialog box that allows you to set the Number of Classes, Lower Limit, Upper Limit, and Bar Pattern. Click on other samples to see their histogram plots.
Click on the "T-TEST" icon in the Toolbar Area. The Viewing Area shows a framework for t-Test results. Click on the sample labelled "Apr–Sep Heating ($)". The Viewing Area now shows the t-Test framework with "Apr–Sep Heating ($)" added. Now click on the sample labelled "Oct–Mar Heating ($)". The t-Test framework shows the results of a t-Test comparing "Apr–Sep Heating ($)" with "Oct–Mar Heating ($)". It shows the t- and p-values as well as the critical t-values for the 90%, 95% and 99% levels. Other t-Tests can be immediately conducted by clicking on other pairs of samples.
Click on the "REG" icon in the Toolbar Area. The Viewing Area shows a framework for a regression plot. Click on the sample labelled "Apr–Sep Temp. (°C)". At this point, STATsimple adds the selected sample to the X-axis of the plot. Now click on the sample labelled "Apr–Sep Heating ($)". The Viewing Area now shows a linear regression plot. The upper right section of the plot shows the regression equation in the form "Y = aX + b, (r)", where "r" is the regression coefficient.
Initially, STATsimple will estimate the lower and upper plotting limits for both axes. To set your own plotting limits, double-click anywhere on the plot and you will be presented with a dialog box that allows you to set the plotting limits and increments for both axes. Other Linear Regressions can be immediately performed by clicking on other pairs of samples.
All of the Toolbar buttons have keyboard shortcuts in the form Command-ƒ, where ƒ is the first letter of the button's name. (The Command button is the button on your keyboard with the Apple symbol). Try these:
Command-T Shows the last t-test performed.
Command-H Shows a histogram of the currently flagged sample.
Command-D Shows descriptive statistics of the currently flagged sample.
Now select any sample and type Command-W. This will close the sample and remove it from the Samples Area. Close the other samples using this technique.
Now use the keyboard shortcut Command-O to open some more samples. The standard Open File dialog box will be presented. Locate and open the "Samples" folder again. This time, select and open the "Temperature Measurement" sub-folder. You will see four data files listed. Open the file "Alcohol". Use Command-O to open the other three files; "Mercury", "Platinum RTD" and "Thermocouple".
Click on the "ANOVA" icon in the Toolbar Area. The Viewing Area shows a framework for a one-way analysis of variance. Click on the sample labelled "Alcohol". The Viewing Area now shows the framework with one sample added. Now click on the sample labelled "Mercury" and you will see the complete ANOVA results for the two selected samples. Add "Platinum RTD" and "Thermocouple" to the analysis by clicking on them. The framework now shows the complete ANOVA results for the four samples. It shows the F- and p-values as well as the critical F-values for the 90%, 95% and 99% levels. Note that the Samples Area now shows the four sample containers as open since they are all involved in the results displayed in the Viewing Area. As an exercise, remove "Thermocouple" by clicking on it. Note the changes to the ANOVA results. Now click on it again to put it back into the analysis.
With all four samples selected for the ANOVA, click the "BON-T" button in the Toolbar Area. The Viewing Area now shows a grid of all the possible t-test comparisons using the Bonferroni inequality. Comparisons with a p-value less than or equal to 0.01 are shown with a grey pattern. Comparisons with a p-value less than or equal to 0.05 are shown with a hatched pattern, and comparisons with a p-value less than or equal to 0.10 are shown with a dotted pattern.
The Bonferroni t-tests are linked to the ANOVA routine. To demonstrate this, remove "Thermocouple" by clicking on it. Note the changes in the results. Now click the "ANOVA" button and you will see that "Thermocouple" is also excluded from the Analysis of Variance.
