By Carolyn Galvin, Primary Intelligence – CCS® Strategic Partner
There’s an ongoing debate over which type of data is better: quantitative (“quant”) or qualitative (“qual”). For researchers who have used and benefited from both, there are distinct advantages and disadvantages from each. There are also instances in which each method is best suited to a specific application.
Quantitative data is all about numbers. When people talk about “big data,” they’re talking about quantitative data—exact, scientific, precise. Black or white, quantitative data is straightforward, although the interpretation of quantitative data can be manipulated—think statistics.
In his popular 1954 book, “How to Lie with Statistics,” Darrell Huff highlights the many creative ways statistics have been used to distort reality, such as truncating the bottom of a line or bar chart in a graph so that differences seem larger than they really are, or representing one-dimensional quantities as two- or three-dimensional objects to compare their sizes. In the latter example, readers often forget that images don’t scale in the same manner as quantities do.
Quantitative data is often used in science and medicine. It’s also common in market research studies when trying to collect ratings feedback, including the relative degree to which someone agrees or disagrees with specific statements, as well as ratings for product or vendor performance. Quantitative data is unambiguous in telling the user about a point in time, the results of a study or an opinion.
But while quantitative data explains the “what,” it doesn’t always explain the “why” or the “how.” Why are the numbers high, low or average? How can specific patterns in the results be explained? How can we interpret the overall trends? For this, we often look to interpretation. And for interpretation, we frequently call upon qualitative data.
Qualitative data helps explain the “why” and “how” behind the numbers. It gives meaning and context to the raw data. It provides color. Examples of qualitative data are free form responses to questions asked in telephone and in-person interviews.
It’s believed that between 80 and 85 percent of all business data is unstructured, or qualitative, data. This includes emails, reports and conversations workers and managers may have with colleagues, suppliers and partners throughout any given business day.
While qualitative data can be rich in insights and can uncover new trends and ideas, it’s free-flowing form is also a key disadvantage—qualitative data is often voluminous in nature, making extraction and usage of key insights time consuming and difficult. In market research studies, for example, collecting open-ended feedback can help identify some very interesting trends—trends that quantitative data alone may not reveal. However, wading through tens or hundreds of thousands of words is not trivial. It’s a major time and resource commitment.
Thankfully, text analytics is helping to make the analysis, interpretation and reporting of qualitative data much less labor and time intensive. Text analytics that includes sentiment analysis is even better.
Text analytics is a set of statistical, linguistic, and machine learning techniques that allow textual data—such as documents, emails and speech—to be used for business intelligence, market research investigation, and outcomes analysis. As text analytics applications have become more widespread since the 1990s, this technique has found growing usage in competitive intelligence, business intelligence, sentiment analysis, listening platforms and social media monitoring, among others.
Recommendations for Quantitative and Qualitative Usage
How do you know when and whether you should collect and use quantitative data versus qualitative data? Below are general guidelines for both types of data.
Collect quantitative data when you have:
Straightforward binary questions (yes or no, male or female)
Ratings questions (scales of 0-5, 0-7, or 0-10)
Check boxes (lists of possible choices)
Sensitive questions (such as pricing questions)
Collect qualitative data when you:
Want to understand the “why” and “how” behind the numbers
- Want to probe on certain areas of the research
- Have a new product to test
While some may see quantitative and qualitative information as black and white—an either/or proposition—many successful researchers use both types of data in combination as complementary tools in their toolbox. Quantitative data will provide hard numbers against which to benchmark your product, company and/or competitors over time, while qualitative data will provide invaluable insights behind those numbers.
A study by Gallup highlights the beneficial outcomes of collecting a combination of qualitative and quantitative feedback, especially for B2B buyers, where 60 percent of customers are indifferent, 11 percent are actively disengaged and only 29 percent are actively engaged. As Gallup points out, and as Primary Intelligence consistently confirms in our Win Loss and Customer Experience interviews, capturing only quantitative information misses significant flags in customers’ experiences with their vendors. When vendors commit to capturing verbal feedback, they typically identify consistent “themes” in customer and buyer feedback, themes that may be problematic for the long-term relationship if not addressed before they grow into larger issues and the customer is in danger of defecting to a competitor.
While it’s cheaper, more convenient and more expedient to send out or post to social media a short, online survey, the use of both qualitative and quantitative information gathering techniques, especially in complex B2B markets, is essential in understanding important customers, markets,and competitors holistically. Collecting this rich data also allows trending over time to see how customer and market needs are changing, as well as how competitor strategies are evolving.