Collecting and analyzing qualitative data

Quantitative data naturally comes with a value attached and a structure for organizing values. When we look up the population of New York and the population of Tokyo, we likely have a table visualized in our minds lining up the cities with their associated population counts. Elements on the periodic table are organized by atomic weight, while reviewer scores and total box office receipts measure the “quality” of movies.

For this and other reasons, quantitative data is easier to collect and analyze since the use of numerical values allows us to determine whether one movie is “better” than another or which cities are growing the fastest. Numbers are intuitive; even as quantitative approaches become more complicated, the fundamental assumption that numerical values can provide a sense of measurement means that quantitative research can yield easily accessible data.

Qualitative data, on the other hand, can be more challenging to measure and organize. Notice that scores provided by movie critics and movie theater profits can provide a numerical sense of which movies are considered better or more popular than others. This sense carries the assumption that bad movies are rated poorly by reviewers and don’t draw a big enough audience of moviegoers. Neither measurement, however, gets to the core essence of what is a substantively good movie, if it can be defined at all.

Imagine focus groups or qualitative research methods that involve dynamic interaction among multiple research participants. In all but the most structured contexts, there is seldom a predictable order of speakers or actors. As a result, the resulting transcript or set of field notes may initially only be ordered by time (i.e., chronological order of events captured by data collection). They may not be ordered in a way that allows for various forms of qualitative data analysis. In a nutshell, any research data collected from qualitative methods will often require some reorganization or restructuring.
As mentioned earlier, this part of the research process may not be as glamorous as collecting or analyzing data. However, extracting useful insights from analyzed data is a significant challenge without providing some order to the data.

How do you manage qualitative data?

Managing qualitative data is a matter of sorting all the data you collect in a qualitative study so that organization and qualitative analysis is feasible and even easy. The goal of effective qualitative data management is to make useful data segments for actionable insights easily understandable and searchable for data analysis and presentation to your research audience.

Qualitative data management starts with a mess. You can throw all your research data into one folder representing your research project. Even if each interview transcript, each set of field notes, and each memo is its own file, does your project have the necessary structure for making sense of the information coming from the data?

This part will offer advice about managing qualitative data that may seem self-evident but will nonetheless prove essential to the more substantive stages of data organization leading up to data analysis. By providing the foundational structure for the collected data from your study, you can set up your project for the efficient identification of themes and insights that can inform your research inquiry.

Developing a coding system for organizing data

A critical part of managing qualitative data involves creating a system of ‘codes’ or ‘labels’ to assign to segments of the data. These codes can be based on themes, concepts, ideas, or phrases that emerge from the data. The coding process facilitates a higher level of organization, enabling researchers to categorize and segment their data for more in-depth analysis. Researchers can follow various strategies for developing a coding system, such as creating a priori codes based on literature or existing theories and in vivo codes that emerge directly from the data itself. Furthermore, researchers can consider various coding techniques, such as open coding, thematic coding, or discourse-based coding, depending on the methodology and research question being pursued.

Creating a data management plan

Before even beginning data collection, researchers can benefit significantly from developing a comprehensive management plan for all their data files. This plan should address how data will be collected, how data confidentiality will be ensured, how data will be organized during the research (e.g., whether by data collection method or data type), how data will be stored and backed up to prevent loss, and how data will be disposed of after the project, if necessary. This plan acts as a roadmap for the research process, ensuring that researchers remain consistent and efficient in their data management.