In all likelihood, you have used a Likert scale (or something you’ve called a Likert scale) in a survey before. It might surprise you to learn that Likert scales are a very specific format and what you have been calling Likert may not be. Not to worry — researchers that have been doing surveys for… Read More »
Cross-Tab Reports
Available to Enterprise and Dedicated Accounts.
First of all, What is a cross tab report?
The full Cross-Tab report allows you to compare as little as 2 questions against each other, one on the horizontal and one on the vertical, or multiple questions against each other. The report also displays the relationship based upon question sequence of questions being compared as well as independent of their order (index). The report is displayed in a large table that is familiar to market researchers and the analysis is also referred to as row, column, index instead of horizontal, vertical, index. This is available to all Enterprise and Dedicated account levels.
What it can do for you?
By analyzing your survey data with cross tab reporting you can isolate patterns as they pertain to certain questions or responses
———————————————————————————————————————————————————————–
How do I create a Cross-Tab?
1. Go to your Current Project–>Reports
2. Select “Cross-Tab” from the list.
3. A window will pop up asking to you name your Report Title
4. Click “Create Report”
5. Once inside, you will want to click on the small pencil icon to the right of the “Cross Tab” element so you can choose which questions you’d like to cross-tab.
6. Inside you can add questions to your vertical and horizontal axis to compare multiple questions to one another. The Horizontal axis only allows for a maximum number of 3 questions, but the vertical can handle a lot more. Generally if you have more than two questions in your survey, the first option in the Horizontal axis will be checked when you click into this, and most of the options on the Vertical axis will be checked off. You can add and remove the checks as you need to. Just remember to hit “Save” once you’re done.
7. Once you’ve done so, this should bring you back to the edit tab. Go ahead and click on the “Run/View Report” tab on the left.
8. Under the Run/View Report Tab– Select Run Report Now (or Refresh Data if it’s been run before).
This is how your cross tab report will look with no data in it. In the following section you’ll see how to analyze this data. On this screen you can do a number of things, including Export to Excel, Word, and PDF. You can also see when the last time the report was run.
If you’d like to get the most recent data, you can click on the “Refresh Data” button.
What do these numbers mean?
Step 1 for Analyzing a Cross-tab Report: Knowing your Totals
Above, you’ll see that there were a total of 61 people who answered both “What is your favorite color” and “Do you like cats”. The horizontal line to the right of 61 will give you the Total# of respondents who chose each answer option from the horizontal access question. You’ll notice this will add up to the Total# of respondents:
In the example above: Total # of resp Horizontal Axis = Total# of responses: 12+16+9+12+12 = 61
You’ll also see that there are vertical options that show you the Total# of respondents who chose each answer option from the vertical axis question.
In the example above that is Total # of resp Vertical Axis = Total# of responses: 30+31= 61.
Step 2 for Analyzing a Cross-Tab report: Calculating v% and h%
In order to calculate the v% and h%, you’ll first need to know what each of them does. The best way to think of them would be in sentence form.
When calculating v% the highlighted section above= (Total# of Horiz + Vert): Total# of Horizontal Options who Also selected the Vertical Option
When calculating h% the highlighted section above= (Total# Vert+ Horiz): Total# of Vertical Answer Options who Also selected Horizontal options.
- So for example, the best way to think about v% is “Of the people who answered ‘Yes’ to ‘Do you like cats?‘, 6 respondents also selected ‘Yellow’ for ‘What is your favorite color?‘”
- Similarly “Of the people who answered ‘Yes’ for ‘Do you like cats’, 10 respondents they also liked ‘Blue’ to ‘What is your favorite color?‘”
- The best way to think about h% is: “Of the people who selected “Yellow” for ‘What is your favorite color?‘, 6 respondents also selected ‘Yes’ to ‘Do you like Cats?’”
- Similarly “Of the people who selected “Blue” for ‘What is your favorite color?’, 10 respondents also selected ‘Yes’ to ‘Do you like Cats?’”
Step 2 for Analyzing a Cross-Tab report: Calculating v% and h% part 2:
You may want to confirm that the h% and v% are actually properly working. To do so, this is how we calculate the cross-tab report.
For h%: Of the people who answered Yes, 20% of them said their favorite color was yellow.
The calculation for this is as follows:
Total# of Vert +Horiz (Yellow and Yes) Divided by the Total# of resp Horizontal Axis(Yes) =20%
Total# of Vert +Horiz (6) Divided by the Total# of resp Horizontal Axis(30) =20%
It should come out like this (6 Divided by 30 = .2 which times 100 =20%)
For v%: Of the people who answered Yellow, 50% of them also answered ‘Yes’.
The calculation for this is as follows:
Total# of Horiz + Vert (Yes and Yellow) Divided by the Total# of resp Vertical Axis(Yellow) =20%
Total# of Horiz + Vert (6) Divided by the Total# of resp Vertical Axis(12) =50%
It should come out like this (6 Divided by 12 = .5 which times 100 =50%)
Step 3 for analyzing Cross-Tab Reports: How to Calculate the Index
The Index represents the likelihood for the respondent to have selected both of the crosstabbed responses. (The median of less likely versus more likely is 100, so anything over 100 is more likely, and anything under is less likely).
In the example above, respondents are 27% percent more likely to favor the color Blue and say Yes to liking cats. If the Index is less than 100, then the respondents are seen as less likely. For example, with the color Blue and the answer “No”, respondents are 26 percent less likely to favor both of these items.
(You get those numbers by looking at 127 (that’s 27% more likely because it’s 27 over 100) and 74 (that’s 26% less likely because it’s 26 under 100)
To calculate the index 127 listed above, the only way at this time to do so is using the Vertical% on the Vertical Axis:
For the example above: 127 is for those who answered Blue and Yes:
You would divide Vertical% for the Blue column v% box (62.5%) by the Vertical% for the rows total column box (49.2%) Times 100 =127
62.5/49.2=1.27 X 100 = 127.
For Additional explanations of cross tabs, please see the following article: http://www.surveygizmo.com/survey-blog/cross-tab-report/
Note: At this time there are certain things that are not supported in Cross tab reports:
- Hidden Values
- Custom Groups/Tables
Lastly, Are there any question types not compatible with Cross tabs?
Yes there are. Cross tab reports were built to be able to handle single-select questions only such as Radio Button, likert scale, etc.
At this time we do not have the ability to cross-tabulate multi-select questions such as custom tables, custom groups, ranking tables, etc.
We are however planning on trying to implement these in the future.
Apr. 17 
Apr. 5