groupwisdom™ t-test (Welch’s)

How do I generate a t-test (Welch’s) report in groupwisdom™?

You can generate a t-test on the Reports page of groupwisdom™ after you have created a scenario, point map, and selected a cluster map solution. When creating a new t-test using the New Report button, the option is found under the Show More sub menu.

If you are more of an audio/visual learner, check out the video tutorials.

For more information on Interpreting the Maps and reports based on those maps, read chapter 6 of Concept Mapping for Planning and Evaluation.

Reports – t-test

  1. To perform a t-Test, go to the Reports section  and select the NEW REPORT button on the top right of the screen.
  2. Under GO-ZONE, select Show More to expand the report options available.
  3. Select T-TEST and then scroll down to select the CONTINUE button.
  4. Name your T-TEST and SELECT SCENARIO using the drop-down menu.  Your cluster solution will be the default cluster that is chosen under SELECT CLUSTER MAP.
  5. Next, configure your T-TEST for Group 1 and Group 2. See examples below.
  6. Finally, at the bottom of the screen select the SAVE button
  7. After reviewing your results, they can be exported as a JPG, a PNG, a Word document, or a PDF.

Comparing Different Ratings

After naming my t-test, selecting the scenario and ensuring the correct cluster solution was selected – I am now ready to configure the t-test.  I am comparing ratings on Importance and Feasibility for a specific cluster where I saw a slope in my pattern match.

For Group 1 I configure the same cluster but select the Importance Rating.  I do not adjust my filters.

Next, for Group 2 I configure the same cluster but select a different rating scale, Feasibility.  I also do not edit the filters.

Then I scroll down and select the SAVE button to view my results:

There shows that the level of significance is p<0.005.

Comparing Different Sub-groups

The other common comparison is between 2 sub-groups using the same cluster and the same rating.

First, I configure Group 1 by selecting the cluster and the rating question – then using the Filter Participants to select the Participant Question and desired respondents.

Next, I configure Group 2 by selecting the same cluster and rating scale and then use the filter to select the same Participant Question, but a different option for respondents.

Then I scroll down and select the SAVE button to view my results:

This report shows significance is p<0.001 between my sub-groups.

Buy Now