The truth, the whole truth, and nothing but the truth

The search for reliable and valid evidence from research in education concerning the “truth” of a matter is neither easy nor straightforward. What can we safely take from research findings? What kinds of studies can yield the most reliable and valid evidence? Do we go for close-grained detail or an overall generalization? How selective and biased are we in sifting through evidence and what it tells us? How much do we project our own tastes onto data and our interpretation of the messages coming from them? How can we overcome bias? The opening pages of Silver’s volume The Signal and the Noise (2012) include a disturbing and potentially disruptive comment. He argues that we live in an age when information floods out faster than we can cope with, and so, to survive, we look at it selectively and subjectively, mindless of the “distortions” of the “truth” that such behaviour risks (p. 17). Silver’s message reinforces Kahneman’s (2011) Thinking Fast and Slow, in which Kahneman indicates that, whilst “humans are not irrational” (p. 411), they often need help in making more accurate and rational judgements, decisions, and choices. We tend to see what we want to see, and we quietly, subconsciously look for information that will support our preconceptions and preferences, even if this contradicts what the evidence is actually telling us. None of this is new; consider Jane Austen’s quip in her novel Persuasion: “How quick come the reasons for approving what we like” (Austen, 1818/1993, p. 12). It is easy to mistake the “truth”. Add to this phenomenon Silver’s (2012) comment that the “truth” is the “signal” which arises out of the “noise”, whilst “noise” is that which “distracts us” from such “truth” (p. 17). “Noise” comprises contextual material and, indeed, research methods. But in the variable-dense, highly contingent, setting-specific, context-rich, conditional, and person-dependent world of education, how acceptable is it to consign contexts to being somehow separate, separable from, or somehow lesser than “signals” and “truths”. What if the reverse of Silver’s claim turns out to be the “truth”, and that, in reality, contextual “noise” is actually the “signal” that we should accept? How rational is it to seek and separate the “signal” from contextual “noise” in educational research? Surely research in education includes context as part of the “signal”? There is a compelling argument that suggests that research findings in education are, and should be, inescapably and symbiotically, constitutionally, wedded to context, contextually rooted in it and referenced to it. As the review of Biesta’s (2020) book Educational Research: An Unorthodox Introduction notes in this issue, researchers in education must “recover the constituent features of educational aspects of educational research: what makes educational research specifically educational”. Moreover, the conjectural nature of science makes it clear that what happens to come out of the “noise” of specific pieces of research – the “signal” – is neither incontrovertible nor an absolute “truth”. The provisional nature of scientific “truth” (qua the best that we currently know) is only a tentative, preferably evidence-informed “best bet” (Major & Higgins, 2019). Can and should decontextualized research data in education release educationists from the mire of contexts? Should we go for the small-scale, high granularity