0% found this document useful (0 votes)
7 views6 pages

Hypothesis Testing

This document provides an introduction to hypothesis testing, detailing the concepts of null and alternative hypotheses, their definitions, and examples. It explains the significance levels, Type I and Type II errors, and the differences between one-tailed and two-tailed tests. Additionally, it outlines a five-step hypothesis testing procedure and includes examples of statistical calculations and decision rules.

Uploaded by

cutesyyymakulit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views6 pages

Hypothesis Testing

This document provides an introduction to hypothesis testing, detailing the concepts of null and alternative hypotheses, their definitions, and examples. It explains the significance levels, Type I and Type II errors, and the differences between one-tailed and two-tailed tests. Additionally, it outlines a five-step hypothesis testing procedure and includes examples of statistical calculations and decision rules.

Uploaded by

cutesyyymakulit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

HERMELITA M. ANTIVOLA, Ph.D.

PROFESSOR (STATISTICS)

MODULE 9

Introduction to Hypothesis Testing

Lesson 1. Hypothesis is a tentative, testable assertion regarding the


occurrence of certain behaviors, phenomena, or events, a prediction of
study outcomes.

Two types of hypotheses that will be explored here.

1. Null hypothesis
2. Alternative/Research hypothesis

Null Hypothesis states that the null condition exists; that is, there is
nothing new happening. It is a statement of what the researcher believes
will be the outcome of an experiment or a study. The Null hypothesis can
be written as H0. (read as H subzero). Meaning there is no change.

Alternative Hypothesis (Research Hypothesis) is a statement that is


accepted if the sample data provide enough evidence that the Null
hypothesis is false. The Alternative hypothesis can be written as: H 1. (H
sub one). It is also called as Research hypothesis.

Some examples of research hypothesis:

 Older workers are more loyal to a company.


 Companies with more than P1 billion in assets spend a higher
percentage of their annual budget on advertising than do companies
with less that P1 billion in assets.
 The implementation of a six-sigma quality approach in manufacturing
will result in greater productivity.
 The price of scrap metal is good indicator of the industrial production
index six months later.
 Airline company stock prices are positively correlated with the volume
of OPEC oil production.
 Mathematics performance of the students improve after the
remediation class has been conducted.

In order to scientifically test research hypotheses, a more formal


hypothesis structure needs to be set up using condition statistical
hypotheses.

All statistical hypotheses consist of two parts, a null hypothesis states


that there is nothing new happening, the old theory is still true, the old
standard is correct, and the system is in control. The alternative
hypothesis, on the other hand, states that the new theory is true, there
HERMELITA M. ANTIVOLA, Ph.D.
PROFESSOR (STATISTICS)

are new standards, the system is out of control, and/or something is


happening.

Level of Significance

After setting up the null and alternative hypothesis, the next step is
to state the level of significance. (1% to 15%)

To establish whether our obtained sample difference is statistically


significant – the result of a real population difference and not just sampling
error –it is customary to set up a level of significance, which is denoted by
the Greek letter α (alpha). The alpha value is the level of probability at
which the null hypothesis can be rejected with confidence, and the
alternative hypothesis can be accepted with confidence. Accordingly, we
decide to reject the null hypothesis if the probability is very small (from the
example, less than 5 chances out of 100) that the sample difference is a
product of sampling error. Conventionally, we symbolize this small
probability by p < .05. or P < 5%

Levels of significance do not give us an absolute statement as to the


correctness of the null hypothesis. Whenever we decide to reject the null
hypothesis at a certain level of significance, we open ourselves to the
chance of making the wrong decision which could be Type I or Type II
error.

Type I and Type II Errors

Type I error is committed if we reject the null hypothesis when in fact it


should be accepted.

Type II error is committed if we accept the null hypothesis when in fact it


should be rejected.

One-Tailed and Two-Tailed test.

One-Tailed test. In a one-tailed test, the region of rejection is only in the


right (upper) tail of the curve. And the null hypothesis is presented
in one direction. H0: μ1 ≤ μ2∨¿ H0: μ1 <. μ2

Example of one-tailed test hypothesis


HERMELITA M. ANTIVOLA, Ph.D.
PROFESSOR (STATISTICS)

H0: The mean income of males is greater than the mean income of
females. It can be written as:

H0: μ1 ¿ μ2; (Null hypothesis)


H1: μ1 ≤ μ2; (Alternative hypothesis)

Two-Tailed test. The hypothesis does not state a direction.

Example of two-tailed Test hypothesis

H0: There is no difference between the mean income of males and


the mean income of females. It can be written as:
H0: μ1 = μ2 (Null hypothesis)
H1: There is significant difference between the mean income of
males and the mean income of females.
H1: μ1 ≠ μ2 (Alternative Hypothesis)

Example 2. Suppose that an educational researcher wishes to test


whether a particular remedial math program significantly improves
math skills. At 5% level of significance.

Student Before(x1) After (x2) Difference (D) Difference2 (D2)


1 58 66 -8 64
2 63 68 -5 25
3 66 72 -6 36
4 70 76 -6 36
5 63 78 -15 225
6 51 56 -5 25
7 44 69 -25 625
8 58 55 3 9
9 50 55 -5 25
Σx1 = 523 Σx2=595 ΣD2 = 1,070

One-Tailed Test

 Null Hypothesis: Math ability does not improve after remediation.


 Alternative Hypothesis: Math ability improves after the remediation.

Decision:
 If the computed value is greater than or equal to the critical value at a given
level of significance, then we reject the null hypothesis, otherwise we do not
reject the null hypothesis.
T computed ≥ t critical, Reject the null hypothesis

One-Tailed Test (Sample)


HERMELITA M. ANTIVOLA, Ph.D.
PROFESSOR (STATISTICS)

Null Hypothesis: Math ability does not improve after remediation.


In Symbol: H0: μ1 = μ2; Ha: μ1≠ μ2

Alternative Hypothesis: Math ability improves after remediation.


In Symbol: H1: μ1 ≤ μ 2

Step 1. Find the mean for both the before and after tests.

Σ x 1 523 Σ x 2 595
x1 = = = 58.11; x2 = = = 66.11
n1 9 n2 9
Step 2. Find the standard deviation of the means

SD =
√ D2
N
−(X 1− X 2)2 =
9 √
1070 – (58.11-66.11)2 =
√ 118.89−¿ ¿

SD= √ 118.89−64 = √ 54.89 =7.41

Step 3. Determine the standard error of the means.

7.41 7.41
SD = = = 2.62
√ 9−1 2.83
Step 4. Translate the sample mean difference into units of standard error
of the mean difference. (t-ratio)

x1−¿ x 58.11−66.11 −8
t= 2
¿= = = - 3.05
sD 2.62 2.62
Step 5. Determine the degrees of freedom.
df =n-1 = 9-1 = 8
Step 6. Compare the t-computed (-3.05) and the t-critical (2.306)

Step 7. Decision Rule. Reject the null hypothesis. Since the computed value (-
3.05) is greater than the critical value (2.306) at 5% level of significance
with 8 degrees of freedom.

DECISION RULE:
 If the computed or obtained value is greater than the critical value at a
given level of significance with a given degrees of freedom, reject the
null hypothesis, otherwise do not reject.

t-Test: Paired Two Sample for Means

Variable 1 Variable 2
Mean 58.11111111 66.11111111
Variance 70.86111111 79.36111111
Observations 9 9
HERMELITA M. ANTIVOLA, Ph.D.
PROFESSOR (STATISTICS)

Pearson Correlation 0.589887362


Hypothesized Mean
Difference 0
df 8
-
t Stat 3.054166862
P(T<=t) one-tail 0.007860611
t Critical one-tail 1.859548038
P(T<=t) two-tail 0.015721223
t Critical two-tail 2.306004135

Example 3. Test of Differences Between Means for Independent Groups

A criminologist was interested whether there was any disparity in sentencing


based on the race of the defendant. She scheduled at random 18 burglary
convictions and compared the prison terms given to the 10 whites and 8
blacks sampled. The sentenced lengths (in years) are shown for the white
and black offenders. Using the data, test the null hypothesis that white and
blacks convicted burglary in this jurisdiction do not differ with respect to prison
sentence length.

Years in prison
Black (x1) 4 8 7 3 5 4 5 4
White (x2) 3 5 4 7 5 5 6 4 3 2

Step 1. Mean of x1 = Mean of x2 =


Variance: s12 = S22 =
HERMELITA M. ANTIVOLA, Ph.D.
PROFESSOR (STATISTICS)

Sx 1−¿ x ¿
2
=
√ (N 1 S 12 + N 2 S 22 )( N 1 + N 2 )
(N 1+ N 2−1 )(N 1 N 2)

Decision Rule: If the computed t value is less than the t-critical, do not
reject the NULL Hypothesis, or
IF the computed value is greater than the critical t-value, reject the Null
Hypothesis.

Decision Rule: Reject the null hypothesis because the computed value
(3.05) is greater than the critical t-ratio (1.75) at 5 degrees of
freedom with 5% level of significance.

Five-Step Hypothesis Testing Procedure Used.


Step 1. State the Null and Alternative Hypothesis
Step 2. Select the level of significance
Step 3. Determine the test statistic and compute
Step 4. Formulate the decision rule based on the
test result.
Step 5. Interpret the result and make the implication.

NOTE: Start using the program (in Excel) to solve the problems in hypothesis
testing.

You might also like