On this basis, the hypothesis may be classified as under:
Existence of Empirical Uniformities:
There are some hypotheses which state the existence of empirical uniformities.
By and large, these represent scientific examination of common-sense propositions. These deal with a problem about which some commonsense observations already exist. Such studies can be distribution of a particular class of people in an area, behaviour pattern of specific groups, etc. According to some thinkers such types of hypothesis cannot be called a hypothesis because these deal only with statement of facts, and the whole study is nothing else but affirmation of those facts alone. But that is not true because even such statements are to be either approved or disapproved after collecting and analysing of research material.
Then another objection raised is that such type of hypothesis are not likely to take research in a long way because these deal with what is already known to everybody. But again this objection cannot be sustained. Firstly because what even every one known must be put in presentable and precise way and research can help in that. Secondly, what everyone knows may or may not be correct. So to put commonsense ideas into precisely defined concepts and subject the proposition to test is in itself an important task. The simple level of hypothesis that seeks empirical generalisation, thus, plays an important role in the growth and development of social research.
Complex Ideal Type Hypothesis:
These are the types of hypothesis which aim at testing the existence of logically derived relationships between empirical uniformities. The level of hypothesis moves beyond the expectations of simple empirical uniformities, e.
g. it is not essential that all areas must be natural areas or that all members of minority groups must be marginal men or that all cities must show perfect concentric circles. For proving these hypothesis is essentially useful.
In the words of Goode and Hatt, The function of such hypothesis is to create tools and problems for further research in otherwise very complex areas of investigation.
Relation of Analytic Variables:
There are certain hypotheses which are concerned with the relation of analytic variables. These hypotheses occur at a level of abstraction beyond that of ideal types. While differentiating the three type of hypothesis. Goode and Hatt say.
Whereas the hypothesis of empirical uniformities lead to observation of simple differences, and those dealing with ideal types lead to specific coincidences of observations, the study of analytic variables requires the formulation of relationship between changes in one property and changes in another. This type of hypothesising is more abstract than the others and also more sophisticated. It also has most flexible mode of formulation. At this level the number of variables which can be abstracted and studied is limited by theory and since the theory grows by process itself, opportunities of new research are being constantly created.