Deductive and Inductive Reasoning

  • Add to list Added to list Added
  • Download PDFopens in new window
  • Cite
  • Share
  • Embed


This Skill looks at the ways in which two different types of reasoning can successfully be used to support an argument’s conclusion: deductive reasoning, which sets out to derive a conclusion from premises on the basis of logic, and inductive reasoning, which sets out to derive a conclusion from premises on the basis of patterns and likelihood.

Starting with deduction, we will look at the ways in which deducing a conclusion from premises means knowing the difference between valid and invalid forms of argument. In a valid argument, a conclusion follows logically and inevitably from its premises, such that – if the premises are true – the conclusion must also be true. In an invalid form of argument, by contrast, a flaw in the way the argument is constructed means that its conclusion does not follow logically from its premises.

Importantly, a valid argument is not necessarily true: that is, logic by itself is not the same thing as truth. We’ll look at the ways in which perfectly logical deductions based upon incorrect premises can lead to incorrect conclusions – and some of the dangers that come with looking only at the logic of arguments, as opposed to testing the truth of their premises.

When an argument is both valid and its premises are true, it’s known as a sound argument: in this case, its conclusion must be true. As we’ll see, however, coming up with a sound argument is far harder in practice than it might appear in theory – and it’s important not to be seduced by the appeal of a logical-seeming argument that doesn’t accurately describe the way things actually are.

Following on from an analysis of particular forms of valid and invalid argument, we’ll look at inductive arguments in depth, something that entails the concept of probability: that is, of evaluating how likely something is to happen or to be true. While deductive arguments can in theory demonstrate their conclusions definitively to be true, this is never the case with inductive arguments: the best they can ever do is very strongly suggest that their conclusions are plausible. This means that, while we speak about deductive arguments being valid or sound (or invalid and unsound), we can only ever talk about inductive arguments being strong or weak.

Finally, we’ll consider what is sometimes known as the ‘problem of induction’ – that is, the fact that we often need to accept the conclusions of inductive arguments as effectively true, even while we cannot prove beyond all doubt that they are – and the ways in which seeking disproof can help us overcome this. We’ll also look at the relationship between deduction and induction, and how making your assumptions about probability explicit can turn an inductive argument into a deductive one – and, in the process, potentially clarify the claims upon which an analysis relies.

Overall, this Skill is interested in helping you to think closely about what it means to justify a particular conclusion using different types of reasoning: and how to become more confident in your analysis of the limitations of both your own and others’ reasoning.

Pause for Thought

How might you justify one of your everyday beliefs? Based on the above, do you think your reasoning is deductive or inductive?

Suggested Readings

  • Although it’s a century old, John Dewey’s chapter on ‘systematic inference’ in his classic How We Think remains fresh and stimulating, and can be read for free online at
  • For a classic and comprehensive logic textbook, Copi, Cohen, and McMahon’s Introduction to Logic (Routledge, 2016) is expansive and accessible.
  • When it comes to induction in the context of research methodologies, Kumar’s Research Methodology: A Step-By-Step Guide is one of the best (SAGE, 2014).
  • For a useful online summary of deduction and induction, complete with a handy video, this guide from the University of Warwick in the United Kingdom is short and sweet
  • For one of my own free online videos exploring the power and limits of logic, see