Cross-tabulation – your guide to business applicable results

Have you ever wanted to take your survey data and discover patterns and relationships between your audience groups, but aren’t sure how to have those reflected best on paper? Researchers and insights managers need to understand the “numbers” or the data before claiming that the research conducted was successful in providing business-applicable results. The problem is that the raw data you initially get won’t give that for you and can even be daunting and misleading to review.

The first question you ask yourself when initiating a market research study is, “Does my audience want to buy this product or service?” But you also have to consider all of the different types of customers that represent your audience. Brands differentiate between customers based on demographics, psychographics, and various purchase behaviors and trends. 

You know that there can be multiple different themes and outcomes, but how can you translate your data into results, and further into theories? How do you find those relationships? How differently will one audience group feel towards a product or service than another? What specifically are those differences?

Join our webinar on how to use cross-tabulation for actionable business decisions and make the most of your survey data. 

Cross-tabulation in survey research

When running a survey report, you’ll first have your basic frequency analysis – or the “overall” insights from the respondents who completed your survey, which gives you a general impression of what your audience ‘on the whole’ is thinking and feeling. For example, you’ll see what percentages of your total audience selected each response to a question. And you’ll be able to review the demographic representation of the total audience participating in this study. While insightful, this usually does not fulfill the insights or the big-picture objective you, the researcher, were initially seeking out for this study; more data-mining and in-depth analysis needs to be done.

So what is cross-tabulation?! It is a comprehensive breakdown, and a mainframe statistical model displayed in the form of banners or pivot tables, composed of rows and columns. What these tables do is allow you to analyze and measure the interaction between two variables. The rows (or the x-axis) lays out the questions and their difference survey responses. At the same time, the columns (or the y-axis) represent the variable(s) you want to run a comparison on (*sometimes this can also be based on a survey question). This facilitates identifying different patterns, trends, and correlations between parameters within your study – whether they are or aren’t mutually exclusive.

Cross-tabulation allows you to go beyond raw data and will enable you to see how one or more questions (or variables) correlate with each other. Cross-tabulating also takes data that looks like it can have several possible outcomes and helps you ‘zero in’ on a single theory by drawing these trends, these comparisons, and these correlations between different factors in your study.

To know more about effectively using cross-tabulation in your research studies, join our webinar

When and how is Cross-tabulation used in surveys?

When running crosstabs, you’re able to apply to following types of groupings as variables:

  • Demographics – compare your survey results by splitting out your audience by age, gender, household income range, ethnicity or background, and location (country, region, state/province, campus, etc.)
  • Psychographics – are you able to segment your audience based on individual attitudes towards an entity? What do they value, or what personally gratifies them? What actions do they take to manifest their values or their attitudes?
  • General purchase behaviors – does your audience prefer to purchase your product or service online? In-store? A combination of both? Who among your audience are brand loyalists and willing to buy a product or service across many multiple brands?

Additionally, crosstabs are compatible with all of these different question types:

  • Multiple choice (both single and multi-select)
  • Dropdown menu
  • Matrix questions
  • Rating or Likert scale questions
  • Net Promoter Score (NPS)
  • Conjoint analysis
  • MaxDiff analysis

And they can apply to all of these types of quantitative data – nominal, ordinal, interval & ratio scales.

They are also used across an array of professionals in different industries, most commonly the following:

  • Market researchers / product researchers / customer-satisfaction managers – though feedback and product/service satisfaction surveys, the use of metadata and demographics provide actionable results to improve products and guide the focus of marketing campaigns. Customer satisfaction surveys help managers and researchers identify areas of improvement within a department or region within your business.
  • School administrators – members within the education sector can send course -or- instructor evaluation surveys to students, use variables such as subjects, the time of class, etc., to unveil any weaknesses or setbacks to improve the educational experience for students.
  • Human resources – similar to school administration, HR directors, and managers can use employee engagement, satisfaction surveys, and even exit-interview surveys to identify conflict areas or need for improvement in specific office locations, departments, and job roles.

What are the advantages & benefits of Cross-tabulation?

At its core, cross-tabulation allows you to map out correlations between variables. For example, by looking at percentages, you can determine that ‘this younger audience is more interested in a new product than the older audience is’. On top of the fact that it can find you these correlations, it’s simple and easy to interpret the data without any advanced statistical degree.

Crosstabs will help you avoid any confusion that you would have when reviewing raw data, especially since raw data (which is not arranged in an orderly manner) cannot be fully understood to present. Even if you took a stab at raw data, you could only observe or infer the data that you’re reviewing. You might notice that a lot of customers 35 years old are interested in a new product or service. But in the crosstabs, you have definitive insights that say, “75% of customers 35 years old or younger are interested in a new product or service.” So not only is the data quantifiable, but it’s also relative or comparable.

Lastly, they help deliver clean data that can be used to improve decisions throughout your organization. By having quantifiable and comparable data readily available to you, you’ll ask the right follow-up questions. As a result, you can further delve into the data and find more interesting anomalies and insights, whether through audience filters or through custom variables – based on other survey responses or other profile data you have stored in the platform.

Conclusion

Crosstabs help translate your data into outcomes and outcomes into theories. Not only does cross-tabulation enhance the data and the findings that you have, but it allows you, the researcher, to further and better understand what additional research data you need to search for and determine. The more familiar you become with running crosstabs in your research, the more familiar you are with the analytical processes. The better you will be in understanding what data you need to look for, and help guide your business to relevant results, and maximize your company’s profit.