Social Science Research: Deduction versus Induction

Many social science research texts (and professors) talk about hypothesis testing as a deductive process. "It's like moving from a general principle to specific instances," they tell us, and then, to complete the picture they explain, "as opposed to induction, which moves from specific instances to make a general principle."

Take for instance, the online Research Methods Knowledge Base. The author explains that in a top-down, deductive research process, "we might begin with thinking up a theory about our topic of interest. We then narrow that down into more specific hypotheses that we can test. We narrow down even further when we collect observations to address the hypotheses. This ultimately leads us to be able to test the hypotheses with specific data -- a confirmation (or not) of our original theories." (ref.)

Yet, when one looks up the definition of deductive reasoning, one gets something like this:

"Deductive reasoning is reasoning which uses deductive arguments to move from given statements (premises) to conclusions, which must be true if the premises are true. An example of deductive reasoning, given by Aristotle, is
* All men are mortal. (major premise)
* Socrates is a man. (minor premise)
* Socrates is mortal. (conclusion)" [ref.]

So, is hypothesis testing a deductive process as explained by the author of the Research Methods Knowledge Base? No. In fact, the process described is mostly an inductive one. The only thing deductive about the process is "thinking up a theory about our topic of interest." The rest of the process has its foundation in induction--that is, trying to induce from the data that the general principle is correct.

It is no wonder I have been confused about this for so long--bad teaching.

The scientific method, whether it be pure or social sciences, relies primarily on induction. Don't let someone confuse you by telling you that quantitative research is deductive and qualitative research is inductive. They both utilize inductive reasoning because they rely on data to make conclusions.

So where does deductive reasoning occur? In disciplines that come up with principles without the need to refer to data, for instance, mathematics or theology. It is, as mentioned above, when we just "think up a theory" or when we use abstract reasoning to come up with general principles.

An older Wikipedia entry on Deductive Reasoning found in Answers.com explains it well:

"Alternative to deductive reasoning is inductive reasoning. Many incorrectly teach that deductive reasoning goes from general information to specific information and that inductive reasoning travels in the opposite direction. This is not accurate [emphasis added]. Deductive reasoning applies general principles to reach specific conclusions, whereas inductive reasoning examines specific information, perhaps many pieces of specific information, to derive a general principle. By thinking about phenomena such as how apples fall and how the planets move, Isaac Newton induced his theory of gravity. In the 19th century, Adams and LeVerrier applied Newton's theory (general principle) to deduce the existence, mass, position, and orbit of Neptune (specific conclusions) from perturbations in the observed orbit of Uranus (specific data)."

In The Method of Sociology, Znaniecki explains it this way:

"Sociology can be nothing but a strictly inductive science. This does not mean that it should not use deduction: no science can live without the help of deductive reasoning (...) induction is the dominant and determining method of sociology: deduction must remain entirely subservient to it as an auxiliary method (...) The fundamental distinction between [deductive and inductive methods] is that from the point of view of the deductive method the final test of a new truth is its logical agreement with a truth already established, whereas from the inductive point of view the final test of a new truth is its validity in theoretic application to empirical facts. It is obvious that a science dealing with empirical reality could not be deductive, for it would either be incapable of proving anything or else be inapplicable to its object-matter [emphasis added]" (1934, pp.218-219).

I rest my case.