Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Items Description of Module

Subject Name Management


Paper Name Research Methodology
Module Title DEVELOPMENT OF RESEARCH HYPOTHESIS
Module ID Module 3
Pre-Requisites Understanding the concepts of research hypothesis
Objectives To study the development of research hypothesis
Keywords Hypothesis, Descriptive, Relational, Null, Alternative,
Role Name Affiliation
Prof.Ipshita Bansal Department of Management
Principal Investigator Studies, BPSMV, Khanpur
Kalan, Sonipat

Co-Principal Investigator

Prof. S.P.Singh Department of Management


Paper Coordinator Studies, GKV, Haridwar

Prof. S.P.Singh Department of Management


Content Writer (CW) Studies, GKV, Haridwar

Content Reviewer (CR)


Language Editor (LE)

QUADRANT –I

1. Module 3 : Development of research hypothesis


2. Learning Outcomes
3. Nature of hypothesis
4. Types of hypothesis
5. Sources of hypothesis
6. Features of a good hypothesis
7. Formulation of research hypothesis
8. Summary

1. Module 3: Development of Research Hypothesis

2. LEARNING OUTCOMES:
After accomplishing this module, you shall be able to
 Know the meaning of hypothesis
 Understand the types of hypothesis
 Comprehend the sources of hypothesis
 Understand the characteristics of a good hypothesis
 Become aware of the formulation of hypotheses

3. Introduction
Having identified and formulated, the research problem, the investigator moves on to formulate
the hypotheses to be tested based on the facts already known. In the absence of hypotheses the
researcher cannot proceed with the investigation of his problem. Thus, the hypotheses guide the
researcher through a complexity of facts to visualize and closely choose only those that are
relevant to the problem under investigation.
.

4. What is hypothesis?
The research hypothesis is an assumption made on the probable direction of the results that might
be attained on accomplishment of the research process. Unlike the research problem that
generally takes on a question form, the hypothesis is always in a form. As a declarative statement,
a hypothesis is of a tentative and conjectural nature.
The statements thus formulated can lend themselves to empirical enquiry. Kerlinger (1986)
defined a hypothesis as a statement involving conjecture of the association between two or more
variables written in such a way that it can be proven or disproven with the valid and reliable data-
it is appropriate to obtain these data that we perform our study. When a proposition is formulated
for empirical testing, it is called a hypothesis
A hypothesis should be formulated in simple, unambiguous and emphatic form. A broad
hypothesis might not be empirically testable. Thus, it might be advisable to make the hypothesis
single dimensional and to be testing only one relationship between only two variables at a time.
- High organizational commitment will lead to lower turnover intentions.
- A hypothesis liable to be measured and quantified so that the statistical relationship can be
determined.
A hypothesis is a conjectural statement based on the existing literature and theories of the topic
and not based on the gut feel or subjective judgment of the researcher.
The validation of the hypothesis would necessarily involve testing the statistical significance of
the hypothesized relation. For example, the above two hypotheses would need to use correlation
and regression analysis respectively to test the stated relation.

5. Types of Hypotheses
Hypotheses are statements we assign variables to cases. A case is the entity the hypothesis
converses about. The variable is the characteristic, trait, or quality that, in the hypothesis is
ascribed to the case.
The hypotheses could be of two types:

Descriptive Relational
Hypotheses Hypotheses

Figure 1 Types of Hypotheses

5.1 Descriptive hypotheses


Descriptive hypotheses are propositions that characteristically express the existence, size,
form, or distribution of some variable. They are simply statements about the magnitude, trend
or behavior of a population under study. On the basis of past records, the researcher makes
some presumptions about the variable under study. For example:
 The current unemployment rate in a region is more than 7 percent of the labor force
 The literacy rate in a state is 95 percent.
 The attrition rate in the education sector is about 35 percent
Researchers often use a research question preferably to a descriptive hypothesis.
 Descriptive hypotheses stimulate researchers to make their thinking about the likely
relationships to be found,
 Further encourages them to think about the implications of a supported or rejected
finding.
 Hypotheses are especially use for testing statistical significance.

5.2 Relational hypotheses


Relational hypotheses state the expected relationship between two variables. For example:
Foreign cars are perceived by Indian customers to be of better quality than domestic cars
There can be relational or explanatory/causal relationship. The relationship state merely that
the variables move together in some particular manner without stating causal relationship.
Explanatory (causal) hypothesis state the existence of, or a change in, one variable caused or
led to an impact on the dependent variable. The causal variable is called the independent
variable and the other the dependent variable. However, the independent variable need not be
the sole reason for the existence of, or change in the dependent variable. For example, a rise
in family income causes an increase in the income saved. A decrease in the price of fuels
leads to an increase in the demand for cars.

6. Sources of Hypothesis.
There are various sources of hypotheses. Some important sources are discussed here:

History of
Science``

Value
orientation of Analogies
the culture

Body of Findings of
Theory other studies

Figure 2 Sources of Hypotheses

6.1 History of Science


The history of science offers forceful evidence to the fact that personal and idiosyncratic
experience of the scientist contributed a great deal to the type of form of Questions he may
ask as also to the kinds of tentative answer to these questions (hypotheses) that he can
provide. Some scientists may perceive an interesting pattern in what may merely seem a
jumble of facts to an ordinary person. The history of science is abundantly provided with
occurrences of discoveries made just because the ‘right’ person made the ‘right’ observation
due to his feature, life history and laying open to the world of events. Personal life-histories
determine the kinds of perception and conception and in turn direct a person to certain
hypotheses soon.

6.2 Analogies
Often valuable hypotheses spring from analogies. Students of sociology would come across
analogies wherein a society is compared to a biological organism, the natural law, to the
social law and thermodynamics to social dynamics, etc. Such analogies, provide certain
useful insights formulated as hypotheses arouse and direct enquiries. One of the recent
orientations to hypotheses-formulation is provided by cybernetics. The communication
models now so well entrenched in the social sciences speak volumes of the importance of
analogies as sources of fruitful hypotheses. The hypothesis that similar human types of
activities may be identified lying in the same territory came from plant ecology. When the
hypothesis was borne out by observations in society the concept of segregation as it is called
in plant ecology was accepted into sociology. It has now become an important idea in
sociological theory. Analogy may be very suggestive but care must be taken not to accept
models from other disciplines without a careful scrutiny of the concepts in terms of their
applicability to the new frame of reference in which they are proposed to be used.

6.3 Findings of other studies


Hypotheses may be drawn from the findings of other studies. The researcher may hypothesize
on the basis of the findings of other studies that similar relationship between specified
variables will prevail in the present study also. This is a common way of researchers who
design their study with a view to replicating another study conducted in a different concrete
context. Many a study in social sciences is exploratory in character. The findings of such
studies may be formulated as hypotheses for more structured studies which aim at testing the
hypotheses.
6.4 Body of theory
A hypothesis may grow from a body of theory which by way of logical deduction, may lead
to the prediction that if definite situations are in place reliable results are likely to follow. A
theory designates evident knowledge; logical deductions from this form the hypotheses which
needs to/ be true if the theory is true.

6.5 Value orientation of the culture


The value orientation of the culture lies in a science which furnishes several of its
fundamental hypotheses. Certain hypotheses and not others engage the attention of scientists
in particular societies or cultures may well be given the credit to the cultural emphases. Folk-
wisdom prevalent in the culture also serves as an important source of hypotheses.

7. Features of a good hypothesis


The criteria for judging the usability of the hypotheses is none other than that help the hypotheses
perform their designated functions vis-à-vis research and the growth of knowledge. Hence a good
usable hypothesis must satisfy the following criteria:
Empirically Testable

Closest to things observable

Conceptually clear

Specific

Concerned with a theory

Figure 3 Features of a Good Hypothesis

7.1 Empirically testable


A hypothesis should have the characteristic of being empirically testable. It is possible to
draw certain logical inferences which in turn can be tested by field observation. The
hypotheses should have empirical referents. The concepts contained in the hypothesis must
have empirical correspondence. For example, ‘Bad parents beget bad children’. This
statement hardly qualifies as a usable hypothesis.

7.2 Closest to things observable


Hypotheses should be nearest to objects that can be observed. Failing this, it would be
difficult to test their harmony with empirical facts. Hypothesis must be formulated such that
deductions can be made from it and as a result a decision is arrived at to determine if it
explains the thought about facts or does not.

7.3 Conceptually clear


The hypothesis must have conceptual clarity. The concepts used in the hypothesis should be
clearly defined - formally as well as operationally. Formal definition of the concepts will
clarify a particular concept stands for, while the operational definition clarifies the empirical
evidence of the concept in the field. Obscure hypothesis without fixing limits or roughly
stated concepts is difficult to be tested since, there is absence of generally accepted principle
for understanding the observable facts the test would consist of. The concepts contained in
the hypotheses are commonly acceptable and communicable. This would ensure continuity in
researches resulting in cumulative growth of scientific knowledge.

7.4 Hypothesis must be specific


The hypotheses would do well. Often the researchers are allured to state their hypotheses in
such general terms and with so widely elaborated a scope that they simply do not yield to test.
The concepts for which suitable tangible indexes have not been developed should be avoided
by the researchers in their hypotheses. A hypothesis should contain a clear statement of
indexes to be used. Specific formulations of hypotheses assure the practicability and
significance of research. It enhances the validity of the results as more specific the statement,
smaller the probability that it will really be borne out as a result of mere accident.

7.5 Hypothesis must be concerned with a theory


The hypothesis should have theoretical orientation. Such condition is concerned the theoretic
reasons of a hypothesis. If the hypothesis is concerned to the theory, research will make
competent, support, correct or refute the theory. A science can become cumulative only
through building on the existing body of fact and theory.

8. Formulation of Research Hypothesis


Here the question arises, how should the researcher ideally formulate the hypotheses for research?

8.1 Formulation of Alternative Hypothesis


A researcher should start with trying to determine all the alternative solutions of his research
problem. The researcher will determine which of the alternative courses of action is most
efficient in terms of certain criteria. If the researcher has a problem whose solution depends
on certain predictions and the researcher knows that that there are three theories which are
related to the problem. Now if one of the three alternative theories can predict events more
accurately than the other two, it is indeed most efficient as a solution to the problem. The
alternative hypotheses which the researcher then sets out to formulate are nothing but
statements of conditions for each of the alternative means under which conditions, it can be
said to be the most efficient. Thus, the alternative hypotheses are the statements of acceptance
conditions for each of the alternative course of action.
H1: The average examination score produced by teaching method no.2 is greater than that of
the teaching method no.1 is greater than the average test score produced by teaching method
No.2. Recommend No.1 id H1 proves correct and recommend no.2 if H2 proves to be correct.
If the test scores are equal in both teaching methods, the researcher will have to add another
course of action.
In all research, alternative courses of action, acceptance conditions and hypotheses should be
made explicit. There is no scientific way of selecting one of the alternative hypotheses as
valid unless there is no index of efficiency which can be applied to each of the alternative
courses o action. The applicability of the measure of efficiency depends on certain conditions
holding. For example, of the alternative methods of teaching, the use of examination score as
a measure of efficiency may be suitable only if each subject is allowed an equal period to
complete the common test. These conditions constitute the points of agreement among the
hypotheses. These points of agreement among the hypotheses may be either known or
assumed to be valid. Should such an assumption be made, the researcher must make it
explicit.
If the researcher sets up two hypotheses, there must be at least one point of agreement among
them and one point of disagreement. These alternative hypotheses may be represented as
under:
H1: MN (M and N to be true)
H2 MN’ M to be true but denies N

8.2 Formulation of Null Hypothesis


The alternative hypotheses and the null hypotheses together constitute the framework for the
statistical testing of hypotheses. Null hypothesis in its simplest form asserts that there is no
difference between two populations in respect of some property and that the difference found
between the samples drawn from these populations is only accidental and unimportant.
A null hypothesis asserts that the results found in research do not differ significantly from the
results expected on a probability basis or stipulated in terms of certain theory. For example, a
null hypothesis may state that:
HO: The males and females do not significantly differ in respect of the frequency of visiting
cinema.
Ordinarily, a null hypothesis is more meaningful than other hypotheses because it is exact,
easier to disprove the contrary of a hypothesis than to prove it with certainty. Hence, the
concept of null hypothesis has been found very useful. Besides, the statistical techniques are
better adapted to test a null hypothesis.

9. Hypothesis testing
A hypothesis testing involves a number of steps:

Formulate a Hypothesis
Set up a suitable significance level

Choose a Test Criterion


Computer the statistic

Make the DecisionA23f


Figure 4 Steps in Hypothesis Testing

9.1 Formulate a hypothesis


For the sake of convenience, we assume that a company introduces a new drug which it
intends to test on a few patients and on the basis of the responses of patients, make a decision
whether the drug should be introduced or not.
A Null Hypothesis is about the universe variables: H0 : µ = 100.
The null hypothesis is tested in opposition to Alternative Hypothesis, H1: µ ≠ 100
The null hypothesis is tested with the available evidence. The decision is made either to
accept the hypothesis or reject it. If the null hypothesis is rejected, we alternate hypothesis is
accepted.

9.2 Set up a suitable significance level


Two types of errors can occur in decisions to accept or reject the null hypothesis:
We commit type I error to reject the null hypothesis, when in fact, it is true.
We commit type II error in accepting the null hypothesis, when really, it is untrue.
The level of significance denotes the probability of committing type I error (α ) and usually
take it as 5% ( α = .05). This means that even after testing the hypothesis when a decision is
made, we may still commit an error to reject the null hypothesis. Alternatively the value of α
is taken as .01 but it is the choice of the researcher, based on the sensitivity of the study.
9.3 Choose a test criterion
A suitable probability distribution is chosen to be used for the available information under
consideration. The different distributions that are normally used are:
(a) Normal distribution. Z characteristic is most often used in case of large sample.
(b) t-distribution used for small samples
(c) F test and Chi square test

9.4 Compute the statistic


The sample mean is found out and then the ‘Z’ is calculated. The calculated value is
compared with tabulated value from normal distribution table against the decided criterion
(one tail or two tailed)

9.5 Make decision


If the calculated value of the test statistic is more than the tabulated value, we reject the null
hypothesis and accept the alternate hypothesis. In terms of critical region, the value of
calculated attribute falls outside the acceptance region.

Summary
 One the research problem is formulated, the investigator proceeds to formulate the
hypotheses for testing.
 A hypothesis as a statement involving conjecture of the association between two or more
variables articulated in a way that we can prove or disprove it with the valid and reliable data.
There are two types of hypotheses: Descriptive hypothesis and relational hypothesis.
 Descriptive hypotheses are assertions that characteristically express the existence, size, form,
or distribution of some variable. Relational hypotheses state the expected relationship
between two variables.
 There are various sources of hypotheses including history of science, analogies, and the
findings of other studies, body of theory and value orientation of the culture. A good
hypothesis must be empirically testable, closest to things observable
 Hypotheses must be nearest to observable objects. If they does not it would be difficult to put
their harmony to trial with facts derived from experiments. ’Hypothesis should be formulated
in a way that inferences are drawn from it and that consequently a decision can be made
whether it explains or does not explain the facts considered, conceptually clear, specific must
be concerned to a theory or theoretical orientation. A good hypothesis should be empirically
testable, closest to things observable, conceptually clear, must be specific and related to a
body of theory or some theoretical orientation.
 A hypothesis testing involves a number of steps including formulation of a hypothesis, set up
a suitable significance level, choose a test criterion, compute the statistic and make decision.

You might also like