Economic Sociology

1. Introduction

1.1. Characteristics of the Social Sciences

Economics is the scientific study of an aspect of human behaviour. Because its subject matter involves behaviour of human beings it is called a social science, in the same way as sociology, politics or psychology.

Science is sometimes defined as a systematized branch of knowledge; but it is more than this. A science seeks to explain the relationship between observed phenomena and to draw up laws or generalizations by means of which we can make forecasts or predictions about what will happen under certain circumstances. Science produces by means of two methods which, in practice, are inextricably inter-connected: deduction and induction:

- Deduction implies the logical working out of the implications of certain definitions, and the formulation of hypotheses or theories;

- Induction comprises experimentation, observation and the testing of theories against facts.

In the simplest term, these two methods might be said to be on the one hand thinking, and on the other looking at the real world.

The usefulness of a scientific theory or hypothesis rests on its applicability: the essence of science therefore is the appeal to fact. The more precise and quantitative are the generalizations of science the more useful they are likely to be; hence it has been said that “science is measurement”.

Social sciences have particular characteristics. Because they are concerned with human behaviour and because human beings are not homogenous but have their own individualities and eccentricities, prediction in the social sciences is more difficult than in the natural sciences. However, despite individual exceptions, generalization about human behaviour can be made because such generalizations are about aggregates rather than individuals.

Example An example may help to illustrate the point. As some of the best known laws of economics relate to supply and demand, we could, for instance, draw up a law or prediction which said that: ‘if baker A in a particular town raised the price of his bread, the quality remaining the same, while other bakers in the town kept to the old price without changing quality, then baker A would sell less bread’. Now this prediction, ’baker A would sell less bread’, would still hold true if some customers bought as much bread from baker A as before for reasons of ignorance, loyalty or habit; and it would even still hold of one customer bought more bread from baker a because he fell in love with the baker’s daughter who served in the shop and he kept buying bread as an excuse to see her. These exceptions would not nullify the aggregate effect.

Another feature which makes predictions more difficult in the social sciences is that social scientist can rarely, if ever conduct experiments of the kind carried out by the natural scientist. In the physical sciences experiments can be conducted in the laboratory where the environment (temperature, humidity) can be carefully controlled so that reactions to a change in one variable can be isolated. In the social sciences this cannot be done; moreover, it is probably true that, in the social sciences, as compared with the physical sciences, there are more ‘other things’ affecting the outcome which, ideally, we should wish to keep constant.

Example To illustrate this point, suppose that the Chancellor of the Exchequer wanted to know if a reduction of income tax by 5% would stimulate work effort. He would like to be able to try out this tax change on a small scale in carefully controlled conditions to see what happens before deciding to reduce tax in the country as a whole. But he cannot do this. He therefore reduces income tax for everyone and decides to observe the results so that he can use them for future policy. What he requires is some statistical indicator of changes in work effort following the tax reduction. An adviser might point out that in the coal mining industry there are some reliable statistics of absenteeism; if the tax reduction stimulates work effort, the absenteeism in the mines can be expected to drop. Unfortunately for the Cha