Hypothesis Testing

RESEARCH METHODOLOGY HYPOTHESES TESTING

MEANING  Lundberg defines hypothesis as “ a

tentative generalization, the validity of which remains to be tested”  Goode and Hatt define it as “a proposition which can be put to a test to determine its validity”

Types of hypotheses 

Hypotheses are classified in several ways Descriptive hypotheses Relational hypotheses Causal hypotheses Working hypotheses Null hypotheses Statistical hypotheses Common sense hypotheses Complex hypotheses Analytical hypotheses

Descriptive hypotheses These are propositions that describe the characteristics of a variable.  The variable may be an object, person, organisation, situation or event.  eg., the rate of unemployment among arts graduates is higher than that of commerce graduates. 

Relational hypotheses  Which describe the relationship between

two variables.  The relationship suggested may be positive or negative correlation or causal relationship  Eg.,”Families with higher incomes spend more for recreation”

Causal hypotheses 

 

State the existence of, or a change in, one variable causes or leads to an effect on another variable. The first variable is called the independent variable and the latter the dependent variable. When dealing with causal relationships between variables the researcher must consider the direction in which such relationships flow which is cause and which is effect.

Working hypotheses  While planning the study of a problem,

hypotheses are formed.  Initially it may not be very specific  Which are subject to modification as the investigation proceeds .

Null hypotheses  Null hypotheses are formulated for testing

statistical significance .  E.g.., relationship between families income and expenditure.  Null hypotheses are more exact. It is easier to reject the contrary of an hypotheses than to confirm the complete certainty.

Statistical hypotheses Here it deals with the statistical population. These are derived from a sample.  These are quantitative in nature and which are numerically measurable.  This is measured by co-efficient of correlation. 

Common sense hypotheses It represents the common sense ideas. They state the existence of empirical uniformities perceived through day to day observations.  E.g., soldiers from upper class are less adjusted in the army than lower class man. 

Contd… 

Common sense hypotheses requires three tasks: 1. The removal of value judgement. 2. clarification of terms. 3. application of validity tests.

Complex hypotheses It aims at testing the existence of logically derived relationships between empirical uniformities.  This test is used logically for the further research purpose. 

Analytical hypotheses  These are concerned with the relationship

of analytic variables.  These occurs at the highest level of abstraction.  indicate relationship between changes in one property and changes in another property.

Sources of hypotheses  Theory  gives direction to research by

stating what is known. logical deduction from theory leads to new hypotheses  Observation  derived from observation, e.g., the relationship between the price and demand  analogies  can be from the casual observations from nature and framework.

Contd…  Intuition and personal experience 

contribute to the formulation of hypotheses. Personal life and experience of persons determine their perception and conception.  Findings and studies  developed out of findings of other studies in order to replicate and test.

Contd..  State of knowledge  based on the state

of knowledge in any particular. any formal theories exist, hypotheses can be deduced.  Culture  is the culture on which the researcher nurtured.  Continuity of researcher  the rejection of some hypotheses leads to the formulation of new ones.

Role of hypotheses  Guides for proper direction  Sources of data  Data needs  Suggestions to the type of search  Appropriate technique  Contributes to the development of theory

Characteristics of hypotheses  Conceptual clarity  Specificity  Testability  Availability of techniques  Theoretical relevance  consistency  Objectivity  simplicity

Evaluating hypotheses  How can hypotheses be evaluated?  Robert bales lists the following pertinent

questions to be asked about hypotheses before adopting them for investigation 1.all the concepts used in hypotheses specific? 2.is the posited relationship between variables verifiable?

Contd…. 3.is there any prior evidence as to the truth or falseness of the posited relationship? 4.can an appropriate research design be devised? 5.are the variables are applicable to other situations? 6.are the generalization a part of theoretical system?

Rules of constructing hypotheses  Search for variable measurements  Mutually exclusive and totally inclusive

categories  Sort observation into variable category  Alternative operations  Analyse the variable through their relationship.  Link two or more formal propositions

Testing of hypotheses  Hypotheses are only considered as

provisional ideas and not as facts.  They are just meant for testing through the research process.  Researcher should have an open mind and with a spirit of disinterestedness.