INFERENTIAL STATISTICS

Basic Concepts

Data analysis techniques for determining

how likely it is that results based on a sample or
samples are similar to the results that would have
been obtained for the entire population. In other
words, inferential statistics is a formalized body of
techniques used to make conclusions about the
population based on samples taken from the
population.
 Two Inference Techniques that a researcher
may use: Parametric Test and Non-parametric

Parametric tests:
 Require knowledge about population
parameters
 Assumptions made about population
distribution

e.g., population is normally distributed

Sample data measured on Interval/Ratio
scale
Non-parametric tests
 Requires no knowledge about population
parameters
 Distribution-free
 Some non-parametric tests are designed to
be applied for nominal, ordinal data (  2)
INFERENTIAL STATISTICS

We shall now consider the most important

part in statistical investigation: the determination
of the statistical significance of the results
computed from the sample.

At the start of the investigation, some claims can

be made about the population. Such claims can
be denied or confirmed by evidence from the
sample.
INFERENTIAL STATISTICS

In statistics, hypothesis is used to mean

statement about one or more of the parameters
of a population or populations.
It is a preconceived idea about the value of a
population parameter which can be verified
through statistical procedure or test.

Given two or more contending hypotheses,

testing the validity of one of these statements
through statistical test is known as hypothesis
testing.
INFERENTIAL STATISTICS

STATISTICAL HYPOTHESIS

The first that should be done in conducting

test of significance is the formulation of the
Hypothesis to be tested. Generally, the hypothesis
to be tested should be stated in terms of specific
parameters of the population.
 Hypothesis Testing
Hypothesis testing is a procedure consisting of
pertinent steps whose major objective is to be
able to make a decision based on the data
gathered.

• Hypothesis testing is a technique to help

determine whether a specific treatment has
an effect on the individuals in a population.
 Hypothesis Testing
• The general goal of a hypothesis test is to rule
out chance (sampling error) as a plausible
explanation for the results from a research
study.
 Hypothesis Testing cont.

The hypothesis test is used to evaluate the

results from a research study in which
1. A sample is selected from the population.
2. The treatment is administered to the
sample.
3. After treatment, the individuals in the
sample are measured.
 Hypothesis Testing cont.

The purpose of the hypothesis test is to decide

between two explanations:
1. The difference between the sample and
the population can be explained by
sampling error (there does not appear to
be a treatment effect)
2. The difference between the sample and
the population is too large to be explained
by sampling error (there does appear to
be a treatment effect).
INFERENTIAL STATISTICS

The alternative hypothesis denoted by

𝐻1 𝑜𝑟 𝐻𝑎 refers to the assertion contradicting the
null hypothesis. Thus if Ho is proven to be true,
then the Ha should be false.

These hypotheses are usually stated as shown:

a) Ho : μ = μo c) H o : μ = μo
Ha : μ ≠ μo H a : μ > μo

b) Ho : μ = μo where μ = the population mean

Ha : μ < μo μo = particular mean
INFERENTIAL STATISTICS

The concept of hypothesis in statistical

inference is classified into two:
• Null hypothesis
• Alternative hypothesis
The statement which gives the hypothesized
or preconceived value of the population
parameter under investigation is called the null
hypothesis, denote by Ho.
 This refers to the absence of any effect claimed
for a certain action. This hypothesis also asserts
the absence of difference between observed
and expected values.
INFERENTIAL STATISTICS

 Ho from the above statements asserts that the

population mean is equal to a particular mean
or expected.
 Ha in (a) states that the population mean is not
equal to the expected value. Test involving this
type of hypothesis is known as TWO-TAILED
TEST(non-directional test).
 Ha in (b) and (c) are predictive in nature. Test
involving these alternative hypothesis is known
as ONE-TAILED TEST(directional test ).
 Example 1.
Ho: There is no significant relationship between
job satisfaction and the salary grade of
employees .

Ha: There is significant relationship between job

satisfaction and the salary grade of
employees.
Example 2:
Ho: There is no significant difference between the
strength of concrete at three different types
of enhancer.
Ha: There is a significant difference in the
mathematics achievement of students who
were subjected to three different teaching
approaches.
 Errors in Hypothesis Tests

• Because the hypothesis test relies on sample

data, and because sample data are not
completely reliable, there is always the risk
that misleading data will cause the hypothesis
test to reach a wrong conclusion.

.
Two types of error are possible
Type I Errors

• A Type I error occurs when the sample data

appear to show a treatment effect when, in
fact, there is none.

• In this case the researcher will reject the null

hypothesis and falsely conclude that the
treatment has an effect.
Two types of error are possible
Type I Errors
• Type I errors are caused by unusual,
unrepresentative samples. Just by chance the
researcher selects an extreme sample with the
result that the sample falls in the critical region
even though the treatment has no effect.

• The hypothesis test is structured so that Type I

errors are very unlikely; specifically, the
probability of a Type I error is equal to the
alpha level.
Type II Errors
• A Type II error occurs when the sample does
not appear to have been affected by the
treatment when, in fact, the treatment does
have an effect.
• In this case, the researcher will fail to reject
the null hypothesis and falsely conclude that
the treatment does not have an effect.
• Type II errors are commonly the result of a
very small treatment effect. Although the
treatment does have an effect, it is not large
enough to show up in the research study.
 As a summary:
If the hypothesis is true, then the decision
not to reject it is the correct one. Similarly, if the
hypothesis is false, then the decision to reject it is
also the correct one. However, if the hypothesis is
true and is rejected, then an error is committed.
Such an error is called the Type 1 Error.
Conversely, if the hypothesis is false and is not
rejected, then Type II Error is committed.

The probability of committing type I error is called

the level of significance and is denoted by α.
 Directional Tests

When a research study predicts a specific

direction for the treatment effect (increase or
decrease), it is possible to incorporate the
directional prediction into the hypothesis test.
The result is called a directional test or a one-
tailed test.
A directional test includes the directional
prediction in the statement of the hypotheses
and in the location of the critical region.
 Directional Tests (cont.)
• For example, if the original population has a
mean of μ = 50 and the treatment is predicted
to increase the scores, then the null
hypothesis would state that after treatment:
H0: μ < 50 (there is no increase)

In this case, the entire critical region would be

located in the right-hand tail of the distribution
because large values for M would demonstrate
that there is an increase and would tend to reject
the null hypothesis.
23
 Measuring Effect Size
A hypothesis test evaluates the statistical
significance of the results from a research study.

That is, the test determines whether or not it is

likely that the obtained sample mean occurred
without any contribution from a treatment
effect.
 Measuring Effect Size (cont.)

The hypothesis test is influenced not only by the

size of the treatment effect but also by the size
of the sample.

Thus, even a very small effect can be significant

if it is observed in a very large sample.
 Measuring Effect Size (cont.)

• Because a significant effect does not

necessarily mean a large effect, it is
recommended that the hypothesis test be
accompanied by a measure of the effect size.
• We use Cohen=s d as a standardized measure
of effect size.
• Much like a z-score, Cohen=s d measures the
size of the mean difference in terms of the
standard deviation.
 Power of a Hypothesis Test

The power of a hypothesis test is defined as the

probability that the test will reject the null
hypothesis when the treatment does have an
effect.

The power of a test depends on a variety of

factors including the size of the treatment effect
and the size of the sample.
LEVEL OF SIGNIFCANCE OF TEST

Sample observations are usually taken in

random and this might give value of variation
among sample statistics. High value of variation
results in the rejection of a true null hypothesis. As
much as possible, we want the probability of
rejecting the true null hypothesis to be very small.
The value of this probability is known as the level of
significance. This is usually represented by a Greek
letter α where the commonly used values range
from 0.01 or 1% to 0.05 or 5%.
LEVEL OF SIGNIFCANCE OF TEST
If for example, a 0.05 or 5% level of
significance is chosen in designing a test of
hypothesis, then there are about 5 chances in 100
that we would reject the hypothesis when it should
not be accepted; i.e. we are about 95% confident
that we have made the right decision. In such a
case we say that the hypothesis has been rejected
at a 0.05 level of significance, which means that we
could be wrong with probability of 0.05. The smaller
the value of the level of significance the smaller is
the risk of rejecting the true null hypothesis.
LEVEL OF SIGNIFCANCE OF TEST
A level of significance of 0.05 means that in
repeated sampling from a given population of
interest, the probability of obtaining sample results
similar to the one presently observed is 95% and
the probability of obtaining different sample results
is 5%.
TABULAR VALUE
After specifying the value of the level of
significance, the tabular value can be determined.
This value is sometimes called the critical value. It
divides the area of the graph of a normal curve into
two areas- the area of rejection and the area of
acceptance.
The level of significance represents the area of
rejection.. The value 1 – α represents the area of
acceptance region. The tabular value then serves as
basis for accepting or rejecting the hypothesis based
on the evidenced gathered.
TABULAR VALUE
Example if α = 0.01, this implies that the
probability of rejecting a true null hypothesis is 1%
For two-tailed test, we divide the value of α by two
since we have two areas of rejection. Thus, 0.01/2 =
0.005 .
From Normal table Z0.005 = 2.575, hence we can say
that the sample mean would be greater that 2.575
of standard deviation(sd) or less that -2.575 of the
sd.
TABULAR VALUE

In case of one tailed , α is no longer divided by

2, hence, Z0.01 = 2.33. The rejection region should
be the area above 2.33 of sd (H1: μ > μo ). If H1 : μ <
μo , the rejection region should be below -2.33 of
sd.
Step in Hypothesis Testing

1. Make an assertion regarding the nature of

the distribution of the data.

Generally, we assume that the data under

consideration is normally distributed.
Step in Hypothesis Testing(cont.)
2. Formulate/State null and alternate hypotheses
• Null hypothesis
– A hypothesis to be tested
– Usually represented as H 0
• Alternative hypothesis
– A hypothesis considered as an alternate to the null
hypothesis
– Usually represented as H A
 Step in Hypothesis Testing(cont.)
• Hypothesis tests concerning one parameter
Population mean, m
A null hypothesis for a hypothesis test
concerning a population mean should always
specify
m 0a single
 value for that parameter
•
 (= ) sign must appear in the null hypothesis
Therefore: H 0 : m  m0
H0 : m  mH
H0 : m  mH  0
Step in Hypothesis Testing(cont.)

• Alternative hypothesis
– The choice of the alternative hypothesis depends
on and should reveal the purpose of the
hypothesis test
– Null hypothesis and alternative hypothesis are
mutually exclusive
– Three choices are possible
• H A : m  m 0 (nondirecti onal )
H A : m  m 0 (directiona l )
H A : m  m 0 (directiona l )
Step in Hypothesis Testing(cont.)
H A : m  m0
• An alternate hypothesis with a  sign is called
a two-tailed test
• The population mean, m is different from a
specified value, m 0
• When a < sign appears in the alternate
hypothesis, the test is called a left-tailed test
• When a > sign appears in the alternate
hypothesis, the test is called a right-tailed test
 Step in Hypothesis Testing(cont.)
3. Choose an appropriate statistical test for testing
the null hypothesis.
 The choice of a statistical test is dictated by
the question for which the research is
designed; and
 Step in Hypothesis Testing(cont.)

Step 3 (cont.)
 The level, distribution and dispersion of the data
also suggest the types of statistical test to be
used.
 Secondary consideration:
The extent of your knowledge in statistics;
Availability of resources in connection with
the computation and interpretation of data
 Step in Hypothesis Testing(cont.)
4. Specify the level of significance. (α = 0.01 or
0.05)
5. Determine the critical value.
6. Determine the computed value of the test
statistics from the given condition.
7. Make a decision, by comparing the computed
value to the critical value. We shall have two
possibilities:
• if the computed value is < the critical value,
we accept the Ho and reject H1.
• if the computed value is > critical value, we
reject Ho and accept H1.
 Limits of Hypothesis Testing
• Pre-selecting level of significance
– Lacks a theoretical basis
– Used for convenience
– Binary nature of null and alternative
hypothesis

• p-value or Probability value

– Accepted approach
– The exact significance level associated with
the calculated test statistic is determined
 The p-Value

• p-value is defined as:

– The exact probability of getting a test statistic
value of a given magnitude, if the null
hypothesis is true
Comparing Classical and p-Value Approaches

• Classical • p- Value
– State hypotheses
– State hypotheses
– Decide on significance – Decide on significance
level
level
– Compute the value of
– Select test
the test statistic
– Delineate regions of
– Determine p-value
rejection/nonrejection
– P   reject null
– Calculate the test
hypothesis; otherwise do
statistic
not reject
– State your conclusion in
words – State your conclusion in
words
 Guidelines for Using P-Value

Evidence against H0
p > 0.10 ………………….Weak or none

0.05 < p ≤ 0.10 … … . .Moderate

0.01 < p ≤ 0.05 … … . .Strong

p ≤ 0.01……………………...Very strong