Post-inductive resonance

A topic in research methodology

Much research undertaken in discovery mode, much in interpretivist research, relies upon induction – the process by which we form new conceptualisations to describe patterns in data. This relies upon creative processes to suggest new concepts, categories for describing data, interpretations of data, etcetera.

The problem of induction

At one time the natural sciences were considered to be based upon induction – the were known as the inductive sciences. It was considered that collecting and examining enough data could reveal the true nature of things. It was long recognised that induction was logically flawed (examining no finite number of examples could assure the generalisation of what was observed to all other instances of the same kind), and it became widely reconsigned that this notion had to be abandoned after Popper's seminal 'Logik der Forschung' (1935) – in the English version 'The Logic of Scientific Discovery' (1959).

In science, it is now realised that generalisation is largely based upon a theoretical notion of 'natural kinds', that different types of entity in the world (samples of hydrogen, specimens of diamond, magnetic fields) have inherent properties such that when these are identified for a sample/specimen, we can apply them to the species as a whole (i.e., assuming we have correctly identified a natural kind and that the property being studied is concerned is indeed intrinsic to that kind. This is less likely to occur to natural kinds ('student', 'teacher', 'classroom', etc.).

However, even in the natural sciences, it is understood that data always under-determines theory. That is, even if all the data collected seems to match perfectly to a theory (which would be an ideal case seldom if ever realised) this does not exclude there being other theories which could also equally account for all the data.

"The paradigm observer is not the man who sees and reports what all normal observers see and report, but the man who sees in familiar objects what no one else has seen before."

Norwood Russell Hanson

Testing induction against data

Induction is an act of insight – of human imagination, which finds a pattern that seems to make sense of data, but cannot assure the 'right' or 'best' pattern has been imagined. Once such insights have been received, the analyst needs to test them against the data set – what has been described as post-inductive resonance.

In my own doctoral research, I 'discovered' a pattern in students' thinking about chemical bonding and related topics (the octet conceptual framework), but needed to be able to convince myself that I was not simply then fitting data to the pattern rather than critically testing it.

"In my own research there were moments during the analysis of data that I became aware of hypotheses about relevant categories that seemed to describe aspects of the data. Such a hypothesis may be judged to be authentic if it 'resonates' with the data: that is if the hypothesis is found to match other parts of the data set, and is not significantly challenged by incommensurate data. In my research I referred to this process of matching, of checking hypothesised categories against data, as post-inductive resonance. It is my belief, based on my own experience of the data analysis, that to a large extent the process of post-inductive resonance occurs at a sub-conscious level. Over a period of time, immersion in a data set leads to the sudden realisation that one has interpretations that seem to fit ('resonate with') the data, but which one has not up to that point consciously thought through. One may be able to offer a post-hoc reconstruction of the match between data and interpretation, but one is not able to describe the inductive process."

Taber, 1997: 5

In grounded theory methodology, the process of constant comparison is employed to ensure that codes, categories and concepts developed offer good fit to the data collected.

Post-inductive resonance and language learning

Closely related language often have similar grammatical structures, but learning a relatively distantly related language can present similar challenges to

• making sense of a conceptual scheme at odds with one's existing thinking (so called radical conceptual change in learning, or a paradigm-shift when applied to the scientific community facing a revolutionary change in disciplinary thinking)
• a researcher seeking to find a pattern in rich complex data by testing hypotheses against new data ('constant comparison' in the terminology of grounded theory research).

The linguist Benjamin Lee Whorf described his experience of trying to learn the North American language Hopi. He would construct sentences in the language according to his understanding of the rules, and present them to a Hopi speaker (a form of hypothesis testing!)

"The sentences I made up and submitted to my Hopi informant were usually wrong. At first the language seemed merely to be irregular. Later I found it was quite regular, in terms of its own patterns. After long study and continual scrapping of my preconceived ideas, the true patterning emerged at last."

Whorf, 1938/2012

Sources cited:

• Hanson, N. R, (1958) Patterns of Discovery. An inquiry into the conceptual foundations of science. Cambridge: Cambridge University Press.
• Popper, K. R. (1934/1959). The Logic of Scientific Discovery. London: Hutchinson.
• Taber, K. S. (1997). Post-Inductive Resonance?: the principles of 'grounded theory' applied to chemical education research. Paper presented at the 4th European Conference on Research in Chemical Education, University of York.
• Whorf, B. L. (1938/2012). Some verbal categories of Hopi. In J. B. Caroll, S. C. Levinson, & P. Lee (Eds.), Language, Thought, and Reality (2nd ed., pp. 143-158). The MIT Press.

Acknowledgement: Thanks are due to Robert Fripp for advice in this topic.