Our World Statistics Day conversations have been a great reminder of how much statistics can inform our lives. Do you have an example of how statistics has made a difference in your life? Share your story with the Community!
Understanding your customers has always been important, but it's even more so when marketing products and services related to winter snow sports. Winter sports can have a short season and a limited number of consumers. How do you find them, identify their needs, create new promotions and assess whether the promotions are having an impact? With JMP 11 and the new consumer research platform, you can easily handle these types of questions and tasks.
In this post, we will import survey data and explore the relationships among the categorical variables using the new Categorical platform.
The data was originally stored in the Triple-S format. According to the Triple-S.org website, Triple-S is a language used for describing survey metadata. The add-in for importing from Triple-S version 2 is posted on the JMP File Exchange, which is part of the JMP User Community.
JMP now includes a full suite of tools for performing customer and consumer research. In the past, you might have had to use one product for consumer research work and JMP for design of experiments (DOE). Now for the first time, you can do both types of analyses using a single product, for a more efficient use of your most precious resource: your time. Tools for performing these statistical analyses are now located in one convenient place under the Consumer Research menu.
You already collect information about how customers use merchandise or services or how satisfied they are with your offerings. The subsequent insight lets you generate better products and services, happier customers and more revenue for your organization.
We prepared a hypothetical survey to understand the winter sports buying preferences of consumers; the survey had 448 respondents. The survey asked about Winter Olympics watching preferences, skiing and snowboarding frequencies, whether or not they take family ski vacations, and more.
After importing the survey data using the Triple-S XML add-in, we analyzed the data using the new categorical platform. The new categorical platform in JMP Pro enables you to discover the needs of your customers by performing categorical response analysis. The Categorical platform tabulates and summarizes categorical response data, including multiple response data, and calculates test statistics. The strength of the Categorical platform is that it can handle responses in a wide variety of formats without needing to reshape the data.
The Structured tab enables us to construct complex tables of descriptive statistics simply by dragging and dropping column names into green icon drop zones to create side-by-side and nested results. We have chosen to include responses not in data by checking the option in the bottom left under the Grouping options. This displays rows and columns for all data values, even those with zero cell counts.
The platform allowed us to determine whether families with school-age children are planning to watch the 2014 Winter Olympics. Of the 448 respondents, 64.1% were planning to watch. Of those responding positively, the viewers with school-age children were more likely to watch, with 69.2% versus 60.5% with no school-age children. We quickly were able to create reports based on respondent’s age group, preferences for skiing, snowboarding and winter sports.
Different survey tools report responses in different formats. The Categorical platform supports many common formats without requiring the user to reshape the data. Multiple-response questions, for example, instruct you to “select all that apply.” In JMP, the columns that store these data are called “Multiple Response” questions, as they may store multiple answers per respondent.
One of our "select all that apply" questions asked respondents when they liked to ski: On Vacation, On Business Trips, On Weekends, or On Weekdays. The responses were stored as a comma-delimited string in the data table.
Using the Multiple Response Delimited response type on the Multiple tab, we were able to output the results into crosstab report tables and display frequency charts.
The completed analysis tables can be output and shared with others on our team.
JMP produces a great many default statistics for the resulting crosstabs, including mean scores and chi-squared tests for association. These statistics can be accessed using the Categorical platform’s red triangle menu.
Different survey tools can use different formats to store multiple responses, and JMP can handle many of these through selections on the Multiple Response tab. See the Additional Examples of the Categorical Platform in the JMP Consumer Research reference for more examples of multiple response columns.
Free text responses are used for respondent comment fields and were analyzed for frequency counts of each word used. We then used a tree map to visually explore the most frequent word counts collected in the comments.
In future blog posts, we will use the other analysis platforms used for Consumer Research, including Factor Analysis, Choice, and Uplift platforms to discover simple arrangements in the pattern of relationships among variables, discover the preference structure of consumers, and maximize the impact of our marketing budget by sending offers only to individuals who are likely to respond favorably, even when we have large data sets and many possible behavioral or demographic predictors.
Do you analyze consumer research data currently? Leave us a comment and let us know what types of analyses you routinely do with your survey data or other categorical response data such as defect records or side effects?
Note: Melinda Thielbar, Research Statistician Developer for JMP, co-wrote this blog post with Stan Koprowski.