HYPOTHESIS

&
LOGIC OF HYPOTHESES
TESTING
 HYPOTHESIS
• Hypotheses are speculative statements that are subjected
to verification through a research study.

• It can also be said as ‘Hypothesis is a proposition/proposal

that is stated in a testable form and that predicts a
particular relationship between two or more variables’.

• Technically speaking, the word “HYPOTHESIS” is a Greek

word that means “an assumption subject to verification”.
 Hypothesis is Necessary or Not:

• The formulation of a hypothesis provides a study with

focus.
• It tells us what specific aspects of a research problem to
investigate.

• But hypothesis cannot be set up in all cases:-

• In mere fact finding investigations, no problems may be
raised and the need for formulating hypotheses may not
arise.
• In all analytical and experimental studies, hypothesis
should be set up in order to give a proper direction to
them.
 •Characteristics of a hypothesis

• When constructing a hypothesis, important valid

verifications should be:

• A hypothesis should be simple, specific and conceptually

clear.

• A hypothesis should be capable of verification.

• A hypothesis should be related to the existing body of

knowledge / theoretical relevance.

• A hypothesis should be able to be tested / measured.

 • TYPES OF HYPOTHESIS
• Hypotheses are classified in several ways.

• With reference to their function, hypothesis is of two categories:

• Null Hypothesis &

• Alternate Hypothesis.

• In the literature available on the concept, there are some justifications

given for having two categories of hypothesis.

• One of the reason - to confirm (verify) to the qualities of detachment

(aloofness) and objectively (neutral or impartial) to be possessed by a
researcher.

• Another reason - a convenient approach to statistical analysis (scientific

method), which is formulated to test statistical significance (or statistical
importance).
• This is because we have to be able to prove something which
is indeed true.

• The null hypothesis is what we test with statistics (data).

• This is how we decide whether to accept or reject the null

hypothesis.

• When the null hypothesis is ACCEPTED, we reject the

alternative hypothesis and;

• When the null Hypothesis is REJECTED, we accept the

alternative hypothesis.
• The null hypothesis is generally symbolized as Ho and the alternative
hypothesis as Ha.

• The null hypothesis is a very useful tool in testing the significance of

difference.

• It asserts there is no significant difference between the sample

statistic (sample average) and population parameter (population
average).

• Difference if any is due to chance and has arisen out of sampling

fluctuations.

• Acceptance: The acceptance of Null Hypotheses implies that we have no

evidence to believe the alternative hypothesis - the difference is not
significant and is due to sampling fluctuations.

• Rejection: The rejection of null hypothesis implies that it is false and

• Statistical Significance of hypothesis: Understanding the two types of
hypothesis helps us when we go for tests related to its statistical significance
(statistical importance).
• With non-directional hypotheses, it is like suggesting that there will be a
difference in mean score results, two-tailed test of statistical significance is
used.

• When it is a directional hypothesis, a one-tailed test of significance is used.

• That is, If there is a prediction of which mean will be higher and which will be
lower, then a one-tailed test of statistical significance is applicable.
• ERRORS in testing a HYPOTHESIS:

• Variety of reasons:

• The study design selected is faulty;

• The sampling procedure adopted is faulty;

• The method of data collection is inaccurate;

• The analysis is wrong;

• The statistical procedures applied are inappropriate;

• The conclusions drawn are incorrect.

• The above mentioned attributes will lead to wrong

inferences. This can result in two types of errors:
• The above mentioned attributes will lead to wrong inferences.
This can result in two types of errors:

• Type I error: Rejection of a hypothesis (Null) when it is true (α

error).

• Type II error: Acceptance of a hypothesis (Null) when it is false

(β error).

• Otherwise it is correct decision, if:

• Rejection of a hypothesis (Null) when it is false.

• Acceptance of a hypothesis (Null) when it is true.

• LOGIC OF HYPOTHESIS TESTING:

• A number of steps are involved in testing a hypothesis:

1. Formulate a hypothesis.

2. Set up a suitable significance level.

3. Choose a test criterion.

4. Compute.

5. Make Decisions.
• Step 1: Formulate a hypothesis:
• The conventional approach to hypothesis testing to set up two
hypotheses instead of one in such a way that if, one hypothesis is true,
the other is false.

• Alternatively, if one hypothesis is false or rejected, then the other is true

or accepted.

• These two hypotheses are:

• Null Hypothesis, and

• Alternative Hypothesis

• The null hypothesis is generally symbolized as Ho and the alternative

hypothesis as Ha.

• The null hypothesis is a very useful tool in testing the significance of

difference.
• Step 2: Set up a suitable significance level:
• Non-directional null hypothesis = Two-tailed test of significance

• Directional null hypothesis = One-tailed test of significance

• The probability level that we choose influences how liberal or

conservative we want to be in testing our hypothesis.

• A decision that must be made at the beginning of the research

involves probability levels and potential errors.

• The conventional levels for rejecting the null hypothesis are either .05
or .01. (it means there will be 95% confidence or 99% confidence of
null hypothesis stated).
• 5% is the chance the null hypothesis may be rejected. (that is, 95% there is confidence that
there is no difference observed).
• 1% is the chance the null hypothesis may be rejected. ( that is, 99% there is confidence
that there is no difference observed).
• When a statistical hypothesis is tested there are four
possibilities:

1. The hypothesis is true but our test rejects it. (Type I error)
or (rejection error).

2. The hypothesis is false but our test accepts it. (Type II

error) or (acceptance error).

3. The hypothesis is true and our test accepts it. (Correct

decision).

4. The hypothesis is false and out test rejects it. (Correct

decision).
• Step 3: Select test criterion:
• The next step in hypothesis testing is the selection of an appropriate
statistical technique as a test criterion.

• There are many techniques from which one is to be chosen.

• For Example: when the hypothesis pertains to a large sample (30 or

more), the z-test implying normal distribution is used.

• When a sample is small (less than 30), the use of the z-test will be
inappropriate. Instead, the t-test will be more suitable.

• The test criteria which are frequently used in hypothesis testing are z, t, F
and x² test.
 TEST STATISTICS TO BE USED FOR
DIFFERENT TESTS.

•. Test Used for Test.

Statistics
T - Test For test of Hypothesis involving Small sample,
that is, < or = 30.

Z – Test For test of Hypothesis involving large sample,

that is, > 30.

X2 - Test For testing the discrepancy between observed

frequencies and expected frequencies, without any
reference to population parameter.
(Example: Nominal Scale)
F- Test For testing the sample variances.
 • Step 4 : Compute
• After having selected the statistical technique to

verify the hypothesis, the next step is the

performance of various computations, necessary for

the application of that particular test.

• These computations include: the testing statistic as

also its standard error.

 Level of significance and
critical (critical region) value

Level of Critical Value (Z)

Significance
(one tailed test) (two tailed test)

5 percent 1.64 1.96

1 percent 2.33 2.58

 • Step 5: Make decisions:
• The last step in hypothesis testing is to draw a
statistical decision, involving the acceptance or
rejection of the null hypothesis.

• This will depend on whether the computed value of

the test criterion falls in the region of acceptance or in
the region of rejection at a given level of significance.

• If computed value is less than the critical value, then

null hypothesis is accepted, otherwise, it is rejected.
 Example for
HYPOTHESIS TEST (Numerical Explanation)

• A sample of 400 male students of a college is found to have

a mean height of 171.38 cm. Can it be regarded as a sample
from a large population with mean height 171.17cm and
standard deviation 3.30 cm?

• Solution:

• Here n=400

• Sample statistic (sample average) ¯x¯ = 171.38 cms.

• Population Mean (µ) = 171.17 cms.

• SD = 3.30 cms.
• STEP:1

• Null Hypothesis: (Ho) since sample has been drawn from a large
population there is no much difference between mean heights.

• Ho: µ = 171.17

• Alternative Hypothesis: (Ha) there is difference between the

mean heights of sample and population

• Ha: µ ≠ 171.17

• (Two tail test)

• STEP 2:

• Level of significance: ά = 5%

• STEP 3:

• Standard Error of mean: S.E (¯x¯) = SD/√n = 3.30/√400 = 0.165.

• Test Statistic:

• z = ¯x¯ - µ / S.E (¯x¯)

• = 171.38 – 171.17 / 0.165

• = 0.21 / 0.165

• = 1.27.
• STEP 4:

• Critical Value of 5% = 1.96 (from the table).

• STEP 5:

• Decision: Here calculated z is less than critical value (1.27 <

1.96) the null hypothesis is accepted.

• Hence there is no difference between sample average and

population mean.

• Note: More sums to be referred in text books.

THANK YOU

PPTs by
Dr. Thirumagal J Pillai.