Research hypothesis

Dr. Arindam Sarkar , Department of Geography
Purash Kanpur Haridas Nandi College Website: https://pkhnm.ac.in/
Email ID: arindam.srkr1@gmail.com
Research
Hypothesis
Dr. Arindam Sarkar
Department of Geography
Purash Kanpur Haridas Nandi College
arindam.srkr1@gmail.com

Definition
The main purpose of statistics is to test a hypothesis.
Hypothesis testing in statistics is a way for you to test
the results of a survey or experiment to see if you have
meaningful results. You’re basically testing whether
your results are valid by figuring out the odds that
your results have happened by chance. If your results
may have happened by chance, the experiment won’t
be repeatable and so has little use.
Source Link: https://www.statisticshowto.com/probability-and-statistics/hypothesis-
testing/#WhatisHT

Nature and Character
 When the statement can be tested by scientific method called hypothesis
 In research if researcher would like to find out the relationship among two or
more variables need state about hypothesis before going to experiment and field
survey as well as data collection
 Hypothesis is actually some prediction of expected research findings
 Hypothesis is considered as possible answer of research question
 Sometime hypothesis can address different aspects of research question
 Hypothesis should be based on existing theory, observation and experience
 Hypothesis must be tested through scientific research method (observation,
experiment and analysis)
 Survey , sampling, existing data , different tools, modern techniques etc. Can
be used for hypothesis testing.
 if you set some hypothesis in your research then you have to prove or
disprove it in your result and conclusion part.
 It is not required to set hypothesis along with research question in your
research.

Significance
 Proper explanation of expected result of research can be well explained through
research problem, research question or objectives in association with hypothesis.
 Component of clearly explained hypothesis are manipulation and measurement
of variables and identification of population.
 Hypothesis helps researcher to objectively find out new area of discovery.
 Clear and specific goal of research can be clarified in front of the researcher
through hypothesis.
 Theoretical and practical research can be linked by hypothesis.
 Hypothesis provides connection between theoretical and real world.
 Suitable as well as appropriate research type can be suggested by research
hypothesis.
 Appropriate research design and suitable tools and techniques of data analysis
can be determine by hypothesis.
 Hypothesis provides framework for make some conclusion of existing research.
 Without ant hypothesis research would not be considered as accurate research.

Source of Hypothesis
 Academic literature
 Theoretical and conceptual frame work
 Real life experience
 Previous or Published research
 Proved theory
 Environmental reality
 Tolerance
 Influential interface
 Natural law
 Contemporary scientific justification
 Expedition
 Critical review

Type of Hypothesis
 Simple Hypothesis
 Complex Hypothesis
 Associative Hypothesis
 Casual Hypothesis
 Directional Hypothesis
 Non-directional Hypothesis
 Null Hypothesis
 Alternative or Research Hypothesis

Type of Hypothesis
Simple
Hypothesis
 It is associated with relationship between two variable.
 Example: Increase of discharge in the channel increase cross
sectional area of the river
Complex
hypothesis
 It is associated with relationship between more than two
variables.
 Example: Awareness of cyclone is higher among the
inhabitants who are younger and dwelling in rural area than
those who are younger and dwelling in urban area.
Associative
hypothesis
 It deals with a relationship between variables derived from
the natural setting without manipulation .
 This hypothesis is very popular to used in correlation
research studies.
Casual
hypothesis
 It is affected with cause-effect relationship between two or
more dependent and interdependent variables.
 Here impact of dependent variables may be tested by
manipulation of independent variable.

Type of Hypothesis
Directional
hypothesis
 Specific and expected direction of relationship between two or
more variable can be emphasised by directional hypothesis.
 Nature of relationship (negative or positive) can be stated by
this hypothesis.
Non-directional
hypothesis
 it is also indicate relationship between two or more variables
but it is unable to clarify about direction and nature of
relationship between variables.
 Existence of relationship between two or more variable can
truly be identified.
Null hypothesis  It is statistical hypothesis.
 Statistical data can be tested through null hypothesis.
It has ability to identify no relationship between dependent and
independent variables.
Alternative
hypothesis
 It is also a statistical hypothesis.
 It has ability to identify existing relationship between
dependent and independent variable.

Character of accurate hypothesis
 Well understandable
 Conceptually clear
 Empirical basis (Empirical reference)
 Clear and well defined objectives
 Should be specific
 Realistic
 Connected with previous well established theory or practice
 Simple and well explained
 Consistent and continuous
 Verifiable content
 Relevant to modern context
 Purposiveness
 Available tools and techniques for testing
 Profound effect
 Economically sound
 Environmental relation

Hypothesis testing
Component of Hypothesis testing
1. Level of significance
2. Test of hypothesis
3. Type I and Type II error
4. Two tiled and one tailed research
Significance of Hypothesis testing
 Validity of assumption can be determined by hypothesis testing
 Hypothesis testing clarify about the basis of sample data
 Hypothesis about the population is likely to be true or false

The significance level, also denoted as alpha or α, is a measure of the strength of
the evidence that must be present in your sample before you will reject the null
hypothesis and conclude that the effect is statistically significant. The researcher
determines the significance level before conducting the experiment.
The significance level is the probability of rejecting the null hypothesis when it is
true. For example, a significance level of 0.05 indicates a 5% risk of concluding
that a difference exists when there is no actual difference. Lower significance
levels indicate that you require stronger evidence before you will reject the null
hypothesis.
Use significance levels during hypothesis testing to help you determine which
hypothesis the data support. Compare your p-value to your significance level. If
the p-value is less than your significance level, you can reject the null hypothesis
and conclude that the effect is statistically significant. In other words, the
evidence in your sample is strong enough to be able to reject the null hypothesis
at the population level.
Reference link: https://statisticsbyjim.com/glossary/significance-level/
Significance level
By Jim Frost

Error in hypothesis test
Situation Null hypothesis (HO), Alternative hypothesis (HA)
Situtaion-1 Ho False Rejected Acceptance of HA
Situtaion-2 Ho True Accepted Rejection of HA
Situtaion-3 Ho False Accepted Acceptance of HA
Situtaion-4 Ho True Rejected Rejection of HA
Decision for sample
True Statement Rejection HO Acceptance HO
HO True
(HA False)
False
(Type-I)
True
HO False
(HA True)
Ture False
(Type II)
During decision making time if situation I and 2 is true then 3 and 4 become false

Decision Errors
Two types of errors can result from a
hypothesis test.
Type I error
 Type I error occurs due to rejection of
null hypothesis when it is true.
 Significance level is probability of
occurring Type I error. This probability is
also known as alpha, and is often denoted
by α.
Type II error
 Type II error occurs when null
hypothesis is false and accepted.
 The probability of occurring Type II
error is known as Beta, and is often
denoted by β. The probability of not
committing a Type II error is known as
Power of the test.
Decision Rules
P-value.
 It investigate strength of evidence in
support of a null hypothesis .
 If the P-value is less than the significance
level, then you can reject the null hypothesis.
Region of acceptance
 The region of acceptance is a range of
values. If the test statistic falls within the
region of acceptance, the null hypothesis is
accepted by rejection of alternative hypothesis
 If region of acceptance is well defined then
the chance of occurring Type I error is equal to
the significance level.
Region of rejection
Set of values located outside the region of
acceptance is known as the region of rejection.
If the test statistic falls within the region of
rejection, the null hypothesis can be rejected.

Potential source of error
(in estimating a population distribution using a sample)
Sampling error Non sampling error
Because the sample is not
the whole population
Poor sampling
method
Questionnaire
or
measurement
error
Behavioural
effect
Type of error
Type Character
Measurement When there is a difference between the information desired and the
information provided by the researcher
Sampling When a sample some time does not represent the target population
Frame When a sample drawn from a population differs from the target
population
Random When selected sample is am imperfect representation of over all
population

One tailed and Two tailed research
One-tailed test
 A test of a statistical hypothesis,
where the region of rejection is on only
one side of the sampling distribution, is
called one-tailed test.
 Example: If the null hypothesis states
that the mean is less than or equal to
100.
 The alternative hypothesis would be
that the mean is greater than 100.
 The region of rejection would consist
of a range of numbers located on the
right side of sampling distribution; that
is, a set of numbers greater than 100.
Two-tailed test
 Region of rejection is located on both
sides of the sampling distribution, is
called Two-tailed hypothesis test
 Example: If null hypothesis
associated with mean that is equal to
100. The alternative hypothesis would
be that the mean is less than 100 or
greater than 100.
 The region of rejection would related
to range of numbers situated on both
sides of sampling distribution; that is,
the region of rejection would associated
with partially of numbers that were less
than 100 and partly of numbers that
were greater than 100.

Types of Hypothesis testing
Parametric test Non-Parametric test
 Distribution free test of hypothesis.
 Non-Parametric test is not associated
with any assumption about the population
parameter.
 This type of test don not depend on any
assumption about the parameter of the
population.
 Non-Parametric test assume only nominal
and ordinal data.
 Non-Parametric test needs more
observation
 Example: Sign Test, Fisher-Irwin Test,
McNemer Test, Signed Rank Test, Rank Sum
Test (U-Test , H-Test), One sample Run Test,
Spearman’s Rank Correlation, Kendall’s
Coefficient of Concordance

Research hypothesis

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Research hypothesis

Similar to Research hypothesis (20)

More from Arindam Sarkar

More from Arindam Sarkar (6)

Recently uploaded

Recently uploaded (20)

Research hypothesis