Online Surveys = Ongoing Insights

How to get to know your survey data (really, really well): Part Two

 → 
 → 
How to get to know your survey data (really, really well): Part Two

Let Me WorkAs promised, we’re back with more survey data know-how for you. Quick reminder, we’d reviewed data on HBO’s hit show, Game of Thrones, and found that female viewers would prefer to see Daenerys Targaryan on the throne over Jon Snow.

We also learned that women were far more likely to want a woman on the throne in general. So, how do we know if gender is what’s driving the difference?

One way to answer this question is to use the Filter feature in tandem with the Compare feature—here’s a step-by-step guide.

Step 1: On the left side of your screen in Analyze, click +FILTER.

Step 2: Select Filter by Question and Answer option and then find the question you want to filter by.

Step 3: Click on the response option you want to filer by. In this example, we want to look at people who want to see a queen on the throne—now all the results we see in Analyze are only from people who answered “Queen” to the selected question.

1- Screenshot 2

To answer our question, we’ll still leave the Compare feature applied so we can look at gender differences among those who would like a Queen on the Iron Throne.

Now when we examine the data among viewers who’d prefer a Queen on the Iron Throne, the gender difference has gone away and is no longer significant: the majority of men and women in this group would prefer Daenerys as a ruler.

 

2- Graph 6

Similarly, if we look just at those who wanted a king (by creating a new Filter the same way but selecting “King” this time), there is no significant difference based on gender—the clear majority of both men and women picked Jon Snow.

 

3- Screenshot 3

This helps show that the gender difference in support of Jon or Daenerys as ruler is related to whether they want a queen or king to rule the Seven Kingdoms.

Filter and Compare can help you find even more interesting and relevant patterns that may not have popped out at from the onset. These analysis features can also apply to a wide range of important contexts.

For example, when you look at job satisfaction by level of autonomy you don’t see any big differences. However when you filter and look at different industries separately? You might find that autonomy matters more for job satisfaction in some industries than others.

If you want to seek these types of patterns out in your data, be sure to collect enough respondents so that you’re able to split the data up into smaller groups. When you have a smaller respondent sample to work with, there just won’t be enough valuable data there to get truly meaningful and actionable results.

We hope these tips help and as always, let us know in the Comments section below if you have any questions!

Inspired? Create your own survey.

Inspired? Create your own survey.

PRO Sign Up Sign Up FREE or Sign in
Write Surveys Like a Pro

Ever wonder what SurveyMonkey’s really made of?

Ever wonder what SurveyMonkey's really made of?

Read our engineering blog »