The DataHero Blog

Survey Analysis: A Beginner’s Guide

April 3rd, 2014

Survey Analysis_ A Beginner_s Guide-1

Surveys provide incredible insight into your customers; via satisfaction surveys, industry data or user-volunteered information. DataHero helps you take your survey data and make sense of this data creating visualizations in a simple drag and drop motion. In this post I’ll show you how to compare demographics and psychographics in your survey data, tricks for finding relationships between survey responses, and easy filtering.

To begin, connect to SurveyMonkey and import a survey or upload your own survey data from a different service. Analysis, of course, depends on the nature of your own survey data but this post will outline a good guide to getting you started with your survey data. In this example I’m using a sample of the General Social Survey from 2008, a 90 minute survey reflecting responses to a variety of social issues, broken down by other demographics like age, sex, ethnicity, etc.

Demographics

In most surveys, you’ll likely have some demographic data that will give you an idea of who your audience is. In order to determine how best to develop certain personas, experiment with different combinations of demographic attributes. For example, the chart below represents age and marital status.

This chart reveals trends in your survey respondents, like that the category “never married” decreases as age increases, or that respondents who were “widowed” increases with age. However, notice also that the overall raw number of respondents decreases as age increases. This will be important in the next chart.

Personas by Percentage

Determine how significantly each demographic contributes to your customer base using a percentage chart. Simply toggle the percentage button in the top left corner of your chart to do this. The chart below represents the same age and marital status data as in the previous graph, but broken down by percentage.

Here we see very similar information as before, but the raw numbers are obviously not visible, making the “widowed” category mentioned above seem much more significant than in the previous chart. This is why it can be helpful to look at both raw and percentage numbers in determining personas.

Demographics and Psychographics

Psychographic data  answers questions involving attitudes, values, and opinions rather than factual information. In the General Survey we can use the demographic data on age and the psychographic information of opinion on marijuana legalization to generate the following chart:

Thus, we can see that the older a survey participant, the less likely he/she is to favor legalization of marijuana.  Using both demographic and psychographic data together can uncover potential improvements in your product and business. For example, are you surprised that a certain product skews more to men than women? You may need to reassess advertising initiatives to appeal to this different demographic. Or perhaps users who spend more time on your site are less satisfied, which may indicate that you’re catering to less loyal customers.

Relationships Between Variables

Compare two numerical responses with a scatter plot to uncover relationships between different variables. The example below compares number of years of school completed and age when first child is born for both males and females.

You can see a positive correlation between the two, meaning that as the number of years of schools completed increases, the age when a first child is born increases.

Easy Filters

Easily filter survey responses in DataHero to discover your customer personas in a couple different ways. I’ll demonstrate this using age and highest degree of education.

To view the filter broken down by category:

Drag an attribute (in this case highest degree of education) onto the top portion of the chart canvas and in the left side navigation you’ll see the breakdown of education degrees as represented by age.

DataHero then creates this chart:

To view an aggregate of a filter (or one category specifically):

Drag an attribute onto the bottom portion of the chart where it says “Create a Filter”

You’ll then see a box asking you which categories you’d like to view. You’ll also notice a category called “No Value” which means some respondents did not answer this question. Easily filter out the “No Value” category when survey participants skip a question, which you can see below:

In this case I’d like to see how Bachelor and Graduate degrees change by age, so I’d select those two categories and see the following chart of Bachelor and Graduate degrees by age:

Next, grab one of the teardrops in the mini map at the bottom of your chart and you can immediately zoom into your chart, in this case a specific age group.

The aim of DataHero is to allow you to move through your data with the questions you have in mind, without focusing on learning statistical analysis. With the drag and drop motion and easy filtering, you can answer questions about your survey participants without having to interrupt your train of thought for statistical analysis or programming.


DataHero helps you unmask the answers in your data. There’s nothing to download or install. Simply create an account and connect to the data services you use everyday (like Salesforce, Stripe, MailChimp, Dropbox and Box). DataHero automatically decodes your data and shows you the answers you need through dynamic visualizations.

Sign up free for DataHero and import my survey data

 

By Kelli Simpson

Create my Free DataHero Account

Get the fastest, easiest way to understand your data today.