Exploring options in research analysis beyond the logics of deductive and inductive approaches

Dr Kerry Earl Rinehart

Te Kura Toi Tangata School of Education /The University of Waikato

What about retroduction and abduction?

Have you heard of an approach to research analysis called retroduction? Johnny Saldaña’s  interpretation of retroduction is looking to the history of what is happening now, however, Miri Levin-Rozalis,  interprets retroduction as an iterative step-by-step process of turn taking between deductive and inductive approaches. Turn-taking in a conversation between evidence and theory is how Iddo Tavory and Stefan Timmerman see abductive analysis. 

Abductive analysis, as I think of this approach, may be familiar to readers of detective stories. In a Qualitative Inquiry (QI) journal article I wrote on abductive analysis, I refer to Fred Vargas’ novels. Commissaire Jean Baptist Adamsberg, a main character in these books, first familiarises himself with all the evidence related to a current case. He then takes a long walk and ideas ‘slippery and tangled’ sprout ultimately leading to progress in solving the crime.  In the 2001 novel, Have Mercy On Us All. . .

his long walks often left him with the feeling that not entirely uninteresting notions had started to squirm inside his head … germination is germination whatever you say, and once you’ve got your idea it doesn’t matter two hoots whether it grew on a clean piece of blotting paper or on a rubbish tip.  

To become familiar with the evidence is important and so is getting away from the desk. Other authors have also linked the crime-solving processes of detectives to forms of abductive analysis (author P. D James, for example, is mentioned by Levin-Rozalis in the article linked above).

In my article I quote Sven Brinkmann who describes abduction as a way to think about qualitative analysis. Brinkmann argues this approach “is neither data driven nor hypothesis-driven … but by astonishment, mystery, and breakdowns in one’s understanding”.  Data driven approaches, for example phenomenological or grounded theory approaches are inductive. Hypothesis-driven or theory-driven approaches are deductive, and use conceptual or theoretical frameworks. Abductive reasoning arises from what captures our attention in a given situation (e.g. surprises, puzzles, or shocks) and is typically something the researcher identifies as ‘problematic’. Brinkman explains that both induction and deduction “address the relationship between data and theory”, whereas “abduction is a form of reasoning that is concerned with the relationship between situation and inquiry“.

Not only has abductive analysis been linked to processes detectives use to solve crimes, but conditions for this type of reasoning are often also mentioned. In a book by Leif G.W. Persson, readers are told that Police Superintendent Lars Martin Johansson “found running for exercise difficult. What bothered him was not the physical activity itself but simply the fact that he couldn’t think while he ran: a pure waste of time, in other words. On the other hand, he thought very well while walking”. Daniel Kahneman, author of Thinking, Fast and Slow  actually identified 17 minutes per mile as the strolling speed that enables him “to think and work”. In Kahneman’s view “the mild physical arousal of the walk may spill over into greater mental alertness”. The fictional Johansson, however “did his best thinking while swimming.”. I also find ‘swimming’ (cough). . . or floating about in the Whaingaroa Harbour at high tide conducive to productive ideas in abductive analysis.

What conditions do we need to conduct abductive analysis?

Characterising exercise as defamiliarisation or turning away from evidence emersion—and the computer-in analysis work, I propose three conditions necessary for abductive analysis: That is, these are the conditions that may increase the likelihood of connections, patterns and insights becoming untangled from the evidence gathered and occurring to researchers:

  1. Time: The abductive process takes time for researchers to become very familiar with the research evidence, and researchers need time out to turn away from the familiar in order to make new connections. Time requires patience: patience with our evidence, with our research project and with ourselves. Siobhan O’Dwyer, Sarah Pinto and Sharon McDonough (2018) describe the types of time pressures on researchers and the production of research outputs. There are also constraints related to space in terms of the structure and length of acceptable texts evident in guidelines for authors provided by journal publishers.
  • Value Prompts: Researchers need to be open to notice, recognise and respond to prompts, tuning-in, and valuing influences on our thinking during the research process.  Prompts might be palpable as tingles, nagging out-of-reach notions or hunches. Prompts may come to the surface for a researcher during a break from the analysis desk, from a conference presentation, within the pages of a novel, while daydreaming or swimming. The limited information on qualitative analysis available in published research articles where authors have categorised evidence into ‘themes’ suggested this process is straightforward or tidy. There is often little detail available on how the labelling of presented themes occured. We might read that “analysis was conducted using NVivo”. However, as Tavory and Timmerman articulate “The glaring misconception about these programs is that they conduct the analysis for the researcher. They do not.” Analysis involves not just actions of organisation, catagorisation and retrieval but also intellectual deliberation and discernment.
  • Backward Map: Researchers undertaking analysis need to trace the logics-in-hindsight of any inference, tracking backwards from the patterns that have possibilities using evidence to indicate the path to any new knowledge. Questions are included in my QI article that researchers can ask when they have an idea and then turn back to their evidence. I also ask that researchers be comfortable with greater transparency about how inferences and insights result from analysis.

Above all, valuing time and space for thinking and deliberation is an important aspect of quality research processes and outcomes. Maggie Berg and Barbara Seeber’s book is a call to slow down and engage more deeply. They remind readers to “protect time to think critically and creatively” and “what we need are the conditions in which to think, to become engrossed”. I advocate here for researchers to take time for analysis, to acknowledge influences on our analysis, and to complete backward mapping.  Providing more transparency about how analysis was conducted is useful in any published records. These conditions are likely to benefit all research analysis and subsequently the potential impact of qualitative research.


Kerry Earl Rinehart is a senior lecturer in Te Kura Toi Tangata Faculty of Education at the University of Waikato, Hamilton, Aotearoa New Zealand. Her research interests are how people experience teaching, learning, and being judged; educational uses of digital technologies; and Postgraduate Education research. She is Co-Editor of the online open access journalWaikato Journal of Education, a publication of the Wilf Malcolm Institute of Educational Research.

One comment

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