Exploring Trends in Qualitative Data Analysis

Since 2006, QSR International has been conducting annual online surveys about qualitative data analysis.  These surveys assist us in having a macro view of those that undertake qualitative data analysis, identifying emergent industry trends, and improving our NVivo software to meet the future needs of qualitative researchers.

In this blog post, I wanted to share some of this survey data, and depending on interest share more trends in future blogs.  The charts below have been fitted with polynomial trend lines to reduce noise and assist with visual comparison.

Why do we survey and how?

Between 2006 and 2013, over 10,000 respondents from around the world have completed these surveys, answering 20 to 40 questions per survey.  The respondents to these surveys generally have job titles of analysts, consultants, lecturers, managers, professors, students and researchers.

For the past 3 years, the surveys were conducted via SurveyMonkey, with the analysis of the open-ended questions done within NVivo.  This has been made even easier now that NVivo integrates with SurveyMonkey directly via their API.

How do researchers collect their data?

Over the years we have asked respondents about how they collect their data and have found the following trends:

Most Popular Qualitative Data Collection Approaches

  • Interviews (e.g. one-on-one, in-depth) are the most popular approach with 81% of respondents stating in 2006 that they collected some of their qualitative data via interviews and by 2013 this had risen to 88%.
  • Articles and documents (e.g. literature reviews, government reports) which rose from 51% to 61%
  • Observation (e.g. field notes) which rose from 44% to 60%
  • Focus groups (e.g. group interviews) which rose from 40% to 53%
  • Self-report questionnaires (e.g. assessments and evaluations) which fell from 46% to 38%
  • Individual accounts (e.g. diaries, narratives) which rose from 22% to 35%.

More recent surveys over the past 3 years have highlighted the emergence of popular approaches leveraging online technologies, including online surveys (39% in 2013), online content (36% in 2013), and online communications (28% in 2013).

Overall, we are seeing greater diversity in data collection approaches.

What are the popular data formats?

We also noticed some interesting trends in the digital data formats that are collected by researchers:

Most Popular Qualitative Data Formats

  • Documents (e.g. Word, PDF) have been the most popular format collected; with 73% of respondents stating that they captured some of their qualitative data as documents in 2006 and by 2013 this had risen to 86%.  This increase can be partially attributed to improvements in reference management software and databases, providing more convenient access to digital articles and papers.
  • Audio (e.g. mp3, m4a) which rose from 51% to 79%, the increase in audio collection is likely due to improvements in technology and more respondents conducting interviews, focus groups and observation research.
  • Spreadsheets (e.g. Excel, CSV) which rose from 45% to 51%
  • Web (e.g. html web pages and social media) which rose
    from 45% to 47%
  • Images (e.g. gif, jpeg) which rose from 23% to 40%
  • Video (e.g. mp4, 3gp) which rose from 18% to 40%

Given the range of data formats collected by respondents, NVivo has evolved to ensure that such needs are met, with documents supported in version 7 (2006) and prior; audio, images, and video in version 8 (2008); spreadsheets and databases in version 9 (2010); and web and social media in version 10 (2012).

What about methodologies?

We also found that the methods of qualitative data analysis continue to evolve:

Most Popular Qualitative Data Analysis Methods

  • Mixed methods has been the most popular approach reported; with 50% of respondents stating that they used mixed methods for some of their research in 2006 and by 2013 this had risen to 69%.  This corresponds with many methodologists advocating that multiple strategies such as qualitative and quantitative enhance validity of the research.
  • Grounded theory which rose from 43% to 55%
  • Ethnography which rose from 24% to 39%
  • Discourse analysis which rose from 25% to 33%
  • Evaluation which rose from 21% to 25%
  • Phenomenology which rose from 17% to 25%

More recent surveys over the past 3 years have highlighted other popular methodologies such as interpretive analysis (36% in 2013), narrative analysis (36% in 2013), conversation analysis (21% in 2013), and framework analysis (15% in 2013).  NVivo does not dictate the method, this is a decision that the researcher must decide and the software supports.

More data, more methods, more variety

Over this 7 year period we have witnessed that respondents conducting qualitative data analysis are employing a broader range of data collection approaches resulting in more varied digital data formats being collected and undertaking an increasing range of analysis methodologies.

Sometimes it is hard to notice the constant change in today’s information age; since the start of this millennium we have seen digital cameras and smart phones become ubiquitous, online search providing access to an incredible diversity of information, and the rise of social media changing how we communicate.

It will be interesting to see how technology impacts how we collect data in the future, the forms that it takes, and how we go about understanding its content.

What trends have you noticed? How has your approach changed to accommodate them?

Note: This blog post also appears at The NVivo blog.

Tags: ,

One Response to “Exploring Trends in Qualitative Data Analysis”

  1. Craig Says:

    5 stars.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: