This use case examines how to set up your analysis in Advanced Exploration if you want to crosstab 2 variables that are Dimensions, and view this data trended over time period.
Some common examples that this use case can be used for include:
- Task Completion by Purpose of Visit, by quarter
- Visit Frequency by Path to Site, by month
- Task Completion by Browser Type, by week
- Any crosstabs that involve two different demographic data variables
For this use case, we'll address the following request:
I want to view respondents’ monthly Task Completion by their Purpose of Visit.
Step 1: Select your Columns (X-Axis) and Rows (Y-Axis)
Note that the Resp_TimeEnd variable specifies when respondents submitted their responses, and can be used for analyses that require trending.
Since we are evaluating data trended over time, you would place the Resp_TimeEnd variable as a Column (X-Axis):
Then, since we are evaluating how Task Completion scores vary based on respondents' Purpose of Visit, you would "nest" the Task Completion variable within the Purpose of Visit variable. To do so, first you would add the Purpose of Visit variable as a Row, then add the Task Completion variable as another Row to the right of the Purpose of Visit variable:
As you can see in the image above, by "nesting" the Task Completion variable within the Purpose of Visit variable, we are now able to view the "No" / "Yes" Task Completion data for each of the different Purpose of Visit options.
Step 2: Convert the Resp TimeEnd variable from “Year” to “Month Year”
By default, adding the “Resp TimeEnd” variable as a Column / Row provides the data by Year. For this example, right-click the variable, and select the “Month” option for which the year is also provided:
Step 3: Determine the format in which the data will be shown
The “Color”, “Shape” and “Text” options provided in the Marks card allow you to specify how to show this data in the table or chart. For this example, the data will be shown in “Text” format.
Since respondents’ Task Completion is being evaluated, click and drag the Task Completion variable from the Dimensions list to the “Text” option in the Marks card:
Step 4: Determine how the data in the output will be calculated
Now that we have specified the questions that are included in our analysis, and the format in which this analysis will be displayed, you must specify how the data will be calculated (e.g. as a percentage).
For this step, the items below can be performed by right-clicking the Purpose of Visit variable in the “Marks” card:
Convert this variable from a Dimension to a Measure (Measure > Count). This will update the data to provide the sample size counts for each item in the table:
Then, specify the type of calculation that should be performed in the table. Since we are interested in the proportion of respondents who had certain Purposes of Visit, we want to view the data as a percentage (Quick Table Calculation > Percent of Total):
Finally, since we are working with percentages, we must specify whether the Columns or the Rows in the table should add up to 100%. In this example, each of the panes (%Yes + %No) for the individual Purposes of Visit within each of the Columns should add up to 100% (Compute Using > Pane Down).
Step 5: Specify your chart / table type
You can change the chart type for your analysis by using the "Show Me" option in the toolbar. As there may be a considerable amount of data included in your analysis, it would be recommended to select one of the options that makes it easy for you to easily navigate your data.
Additional Use Cases