Harmony or Disruption? Examining the Coexistence of Generative AI and Qualitative Research

With the initial hype of Generative AI, many started to believe in the ability of AI to replace humans. And why not? The jobs are already getting displaced in customer service, data entry and many other sectors.  

With the expectations from AI skyrocketing and the delivery mechanisms still being in the development stage, the popular opinions are tilting towards a more balanced one.  A lot of us now believe that AI will augment human intelligence and not replace it. 

In general, there seems a cautious optimism towards adopting technology to make their jobs easier and yet researchers are still navigating and figuring out how they can start utilizing the latest technological developments into something that they can take advantage of.  

In line with Alan Turing's philosophy, AI was designed to imitate human intelligence in the way  

 we speak, think and navigate information. The recent breakthroughs in AI have enabled self-attention and deep learning with models containing expansive neural networks imitating the neurons of the human brain. These models can now perform several functions; it can classify, edit,  summarize, answer questions and do variety of other tasks including generating content as per the given guidelines.

With that being said, these models aren't yet intuitive, emotional, curious or culturally sensitive. Humans still have better abilities to understand the context and make judgements and decisions depending on the context.    

 If we look at the kind of job that AI has affected more massively, we can conclude that AI can easily replace cognitively lesser complex tasks.   

Depending upon the size of the firm and geography, qualitative researchers wear several hats. A qualitative researcher is a project manager, coordinator, moderator, client servicing representative, note taker, data analyst and report writer. Some of these roles involve task that is mundane and sometimes outsourced to support  roles. Tasks such as qualitative data analysis involves annotating the unstructured data. Often outsourced , these are the tasks that Generative AI has ability to overtake.  

 With generative capabilities, AI is also a wonderful starting point for tasks like writing discussion guides, proposals, secondary market research, screener writing and even presentation writing. Reid Hoffman recently wrote an entire book called Impromptu with the help of ChatGPT. He asserts in his podcast with Sam Altman that Gen AI is excellent at solving the blank slate problem.  

The Job of a qualitative researcher also involves understanding business problems creating research strategy, data interpretation, insights discovery and report writing. These cognitively complex tasks is where qualitative researchers add the most value. 

Both AI and human have their own strengths.  Perhaps a better example of coexistence can be derived from 'Advnaced Chess'. In a Chess match in 1998, Gary Kasparov partnered with a PC running the chess software— in a match against the Bulgarian Veselin Topalov, whom he had beaten 4-0 a month earlier. This time, with both players supported by computers, the match ended in a 3-3 draw. It appeared that the use of a PC nullified the calculative and strategic advances Kasparov usually displayed over his opponent. This match provided an important illustration of how humans might work with AI. After the match, Kasparov noted that the use of a PC allowed him to focus more on strategic planning while the machine took care of the calculations.  

 The same can be concluded for other use cases including qualitative research. There is a great scope for qualitative researchers to partner with Gen AI and boost self-productivity.  

Let's examine the coexistence of Generative AI and human researchers at each and every step of research currently:

• Proposal writing and research design: Research design involves defining the research problem, setting the research objectives, choosing the research methods, selecting the sample size and criteria, and determining the data collection and analysis procedures. This requires human intelligence and context understanding and this is the place where AI has a limited role and where qual researchers should do all the heavy lifting.  Researchers can use AI to do secondary research, synthesize past research information(If any), knowledge retrieval from internal sources etc. Researchers can also use Generative AI as a sounding board while evaluating research approaches.
• Discussion guide writing: AI can solve the blank slate problem for researchers. Researchers can feed information such as research objectives, the target audience and information areas and ask AI to generate a draft discussion guide. 
  
• Transcription and Data analysis:  Technology is useful in generating base outputs that can be corrected by humans. The task of sorting the data can be time-consuming and tedious for human researchers. AI can help by automatically coding and summarizing the data using natural language processing techniques.
  
• Data interpretation: Data interpretation involves making sense of the data and drawing conclusions/insights. It involves identifying patterns, themes, relationships, contradictions, gaps, and implications in the data. Data interpretation requires human intelligence to apply critical thinking skills. Researchers can ask AI to do rudimentary analysis such as counts overview, frequency distribution, word clouds, Visual trees and basic graphs. These may not provide any insights but will be a vital element in simulating thinking that will lead to insight. 
  
• Presentation writing:  Again presentation writing blank slate problems can be solved by AI.  Human researchers can utilize their abilities to synthesize the data, identify the key insights, and craft a compelling story.  These inputs then can be fed to AI to get help on creating possible outlines and generating the best ways of communicating data. Infact, researchers can create multiple versions of draft presentation outlines and decide what makes the most sense.

In short, there are several avenues that Generative AI can contribute to in qual research. Take it as your copilot and you will have much more time available to focus on strategy, insights and innovation. 

About Me: 

I am a qualitative researcher and tech enthusiast. Like lot of other people, I am fascinated by the recent developments in Generative AI and I truly belive it has potential to change the way we do qualitative research. If you are an enthusiast or want someone to discuss your views with, feel free to reach out to me at [email protected]

About flowres:

flowres is a qualitative research platform created with the aim of transforming Zoom Subscription utilized for qualitative research into a research powerhouse. flowres also aims to remain on the latest technology frontier of Generative AI and bring the best and latest to qualitative researchers. If you like to know more about how we can help, please feel free to book a Demo here.