How does data analysis work? Extracting meaning and insight from your survey data.

How does data analysis work? Extracting meaning and insight from your survey data.

Perhaps one of the most exciting stages of survey research occurs after your data is collected, processed and ready for data analysis. It’s at this stage when you take a look at the results and see how people answered your survey. Your goal should be to build a story or narrative behind the data and to extract the insight that addresses the research objective or business problem that was the very reason for the research in the first place. This process is as much of an art as it is science, but there are a few foundational concepts you can follow to get you going.

Start with the research objective

The starting place for data analysis should be the original research objectives and hypotheses. You can investigate how the data answers these objectives and whether it supports or rejects your hypotheses. From there, a deeper examination of the data should look to answer why the expected outcome did or did not occur. A narrative behind the data would then begin to emerge as answers to some questions opens exploration and examination into others. This is best illustrated with a case example.

Imagine a study to understand career opportunities in the finance industry. What might you want to learn about?

  • Salary and compensation
  • Attitudes about job satisfaction
  • Work – life balance
  • Responsibilities / roles
  • Demographic makeup

Summarizing Data

An analysis of the data would begin by reporting the findings along these key areas of your research goals. You are essentially summarizing the characteristics or opinions of the people you surveyed. This is done with descriptive statistics like using averages and percentages to directly report how people answered your survey.

How does data analysis work? Extracting meaning and insight from your survey data.
Figure 1: Descriptive statistics are used to profile or ‘describes’ the sample

Descriptive statistics frames the story behind your data. Now you have a profile of what a career in the finance industry looks like. People tend to earn high salaries. The work-life balance can be challenging. The industry tends to skew male. Having the basic foundation of ‘who’ and ‘what’ the finance industry is important to know, and this sets the table for diving deeper into your data for further insights.

Crosstab analysis

Researchers often tackle the data analysis process by looking at whether the summary statistics differ between subgroups of the sample. This involves using a technique called ‘crosstab’ analysis. Here, the survey data is ‘split’ or ‘segmented’ in order to compare the opinions and behaviors between one subgroup and another. Demographic splits are common, such as looking at gender or age differences. Attitudinal segments are also widely used. For example, comparing those that are satisfied with their finance job versus those that are not. It’s a simple technique yet looking for segment differences yields a wealth of insights. In fact, crosstab analysis is one of the most widely used techniques for analyzing survey data.

Geographical Differences

As an example, splitting the data set by geographic region may show the following characteristics differ depending on a person’s job location:

  • Salary
  • Job level / Seniority
  • Attitudes about job satisfaction
How does data analysis work? Extracting meaning and insight from your survey data.
Figure 2. Looking a data differences between subgroups can yield a wealth of insight

Once you know these differences exist, that then begs the question why? Perhaps being close to an urban center or the corporate headquarters drive these differences. Or maybe the geographic location reflects a unique culture for how work/life balance is approached and that’s causing a difference in compensation. You can check whether the data supports or rejects any of these hypotheses. It’s this process of questioning what’s going on, and why, which will provide the guidance on where and how to look at your data. In this way, you build out a narrative for your original research objective to understand career opportunities in the finance industry.

This just scrapes the surface of how a researcher analyzes a data set. There are other analytical methods like gap analysis, multivariate techniques, and looking for associations in the data. But the fundamental approach remains the same. The analysis begins with investigating the original research objective. Hypotheses are formulated and tested. You want to question and understand the who, what, when, where, how and why behind the behaviors and opinions of consumers. These investigative questions are your guide for how to explore your data set and piece together a story with meaning and insight.

Share this article:

Want to talk? Have a question? We’d love to hear from you. E-mail

See for yourself: Begin your journey to better customer insights

Request a demo

Sign up to receive news from FocusVision