Define various types of hypothesis.
Define various types of hypothesis.
Ans.
Types of Research Hypothesis
Research hypothesis may be of two types: descriptive and relational. Descriptive hypothesis are those propositions which tentatively state the existence, size, form or distribution of some variable, subject to verification by research. For example, “The current unemployment rate in Rajasthan exceeds 20 per cent of its work force” is a descriptive hypothesis about the size or extent of unemployment in the State. Similarly, a hypothesis about the form of unemployment may be, “The unemployment in the State is concentrated among its tribal population”. Relational hypothesis are those propositions which state the relationship between two variables. For example, “Families with higher incomes spend more recreation”, or “High on unemployment among the tribal population of Rajasthan is the result of 3 successive droughts in the State”. In relational hypotheses there always exist two, variables-the independent variable and the dependent variable. The independent variable is that which causes or leads to an effect on another variable. The variable in which the change or effect is produced is known as the dependent variable.
The Null Hypothesis
A hypothesis in a research is generally expressed in the form of a null hypothesis. The null hypothesis is a hypothesis of no difference. It is usually formulated for the express purpose of being rejected. If it is rejected, the alternative hypothesis may be accepted. The alternative hypothesis is the operational statement of the experimenter’s research hypothesis.
Suppose on the basis of certain social theory we predict that two specified groups of people differ in the amount of time they spend in reading newspapers. This prediction would be our research hypothesis. Confirmation of this prediction would lend support to the social theory from which it is derived. The test this research hypothesis we state it in operational form as the alternative hypothesis that the mean amount of time spend in reading newspapers by the members of the two populations is unequal. Null hypothesis would be that the mean amount of time spent in reading newspapers by the members of the two population is the same. If the data permit us to reject the null hypothesis, then the alternative hypothesis can be accepted, and this would support the research hypothesis.
To take another example, suppose, on the basis of past experience we predict that a new method of therapy is more effective in treating psychiatric patients. This prediction is our research hypothesis. To test this research hypothesis we state it in operational form thus. The percentage of recoveries is cases where new therapy is followed is greater than in case when the new therapy is not followed. This is alternative hypothesis the null hypothesis would be that the percentage of recoveries in both types of cases is the same. The new therapy does not increase the percentage of recoveries. A null hypothesis because it is a hypothesis of no difference or no effect.
