How to Use the Change Explorer

When looking at data across time, the standard analysis is to create crosstabs to compare data across different time periods (in the columns). Our Date Aggregation menu makes it easy to create various aggregations and ranges of data to compare, see How to Conduct Significance Tests by Comparing to Previous Time Periods. There are also other tips on how to work with tracking studies, which contain timeseries data, here: Tracking Study Best Practices.

Sometimes you may notice a change in a metric from one time frame to another, and you want to dig in more to see what subgroups of respondents impacted that change the most. The Change Explorer can help you identify the most impactful subgroups of respondents causing change between two periods. This is more of an ad hoc analysis that digs into a specific change and can help you, as the analyst, find the "story" hidden in your data. For example, if NPS is down, did a particular subgroup, say Millennials, have a much lower score this quarter? Or, alternatively, has a particular subgroup that typically has a low score grown relative to the sample overall? This article walks through an example of how to use the Change Explorer to answer these questions.

Examples

The following table shows the Net Promoter Score over time from a survey of mobile phone owners:

The second period shows a significant decline in the NPS. The Change Explorer can be used to visualize the contributions to this change in terms of key subgroups.

Requirements

A numeric outcome variable that you measure over time.

Method

In this example, we will look at the change in the NPS in terms of four key variables from the survey:

Gender
Age
Main phone company
Population density, or whether the respondents live in large cities, small cities, towns, or small towns / rural areas.

Here are the steps:

Select Anything > Visualization > Exotic > Change Explorer Prototype from the toolbar.
In the Outcome box in the object inspector , select the numeric variable you want to focus on.
In the Date box, select a variable in Date format
In the Profiling variables box, select the variables you want to use to profile the change in your outcome variable.
In the Contribution box, select the type you want to use. The choices are:

In this example, we will select Change in Average.
In the Output box, select either Bubble Chart or Table.

In this example, we will select Bubble Chart.
In the Specify periods box, select either Typing dates or Choosing from a Date control and specify the beginning and ending date for each period. In this example, we typed the dates.
In the Significance Levels section, specify the upper and lower z-Score threshold values you want to use to indicate the degree of increase/decrease in your metric of interest. You can also specify the Colors you want to use to identify the categories that showed significant increase or decrease over time, or no significant change at all.
Click Calculate.