Scribd
Upload
229 views25 pages

Hypothesis Testing: Ms. Anna Marie T. Ensano, MME CASTEDSWM Faculty Universidad de Sta. Isabel, Naga City

This document discusses hypothesis testing and provides examples to illustrate key concepts. It defines a hypothesis as a statement that is subject to statistical testing to determine if it is true or false. The two main types of hypotheses are the null hypothesis (Ho), which expresses no difference or relationship, and the alternative hypothesis (Ha), which is the negation of the null. There are two types of errors in hypothesis testing - Type I, rejecting a true null hypothesis, and Type II, failing to reject a false null hypothesis. Several statistical tests are described, including one-sample and two-sample t-tests, ANOVA, and two-way ANOVA. Examples are provided to demonstrate how to perform hypothesis tests to analyze claims and compare means.

Uploaded by

shane cansancio
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
229 views25 pages

Hypothesis Testing: Ms. Anna Marie T. Ensano, MME CASTEDSWM Faculty Universidad de Sta. Isabel, Naga City

This document discusses hypothesis testing and provides examples to illustrate key concepts. It defines a hypothesis as a statement that is subject to statistical testing to determine if it is true or false. The two main types of hypotheses are the null hypothesis (Ho), which expresses no difference or relationship, and the alternative hypothesis (Ha), which is the negation of the null. There are two types of errors in hypothesis testing - Type I, rejecting a true null hypothesis, and Type II, failing to reject a false null hypothesis. Several statistical tests are described, including one-sample and two-sample t-tests, ANOVA, and two-way ANOVA. Examples are provided to demonstrate how to perform hypothesis tests to analyze claims and compare means.

Uploaded by

shane cansancio
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
  • Introduction to Hypothesis Testing: Introduces the subject matter of hypothesis testing and provides an overview of the content that will be covered.
  • Understanding Hypothesis: Explains the importance of testing hypotheses in statistical analysis and introduces foundational concepts.
  • Types of Hypothesis: Covers different hypotheses types including null and alternative with examples.
  • Types of Error: Describes errors associated with hypothesis testing, specifically Type I and Type II errors.
  • Level of Significance: Discusses significance level in hypothesis testing and how it relates to type I error probability.
  • Types of Tests: Details different test types, explaining one-tailed and two-tailed hypothesis tests with examples.
  • Graphical Interpretations: Illustrates acceptance and rejection regions graphically for hypothesis testing outcomes.
  • Steps in Hypothesis Testing: Lays out a systematic 7-step process utilized in hypothesis testing from formation to conclusion.
  • Small Sample Hypothesis Test: Introduces formula and conditions for hypothesis testing on small sample means.
  • Example 1: Testing A Claim: Practical example demonstratively analyzing a perfume company's claim using hypothesis testing.
  • Critical Values Table: Provides a reference table listing critical t-values needed for statistical analysis.
  • Testing Two Sample Means: Formula and explanation for conducting hypothesis testing on the difference between two sample means.
  • Example 2: Comparing Locations: Gives a detailed example comparing pedestrian data from two locations, applying hypothesis testing.
  • Practical Activity: Introduces an activity section requiring analysis on the weight change in a diet trial and pill claim.
  • ANOVA - Analysis of Variance: Describes how ANOVA is applied to test significant differences across means of multiple groups.
  • ANOVA Tables: Displays an example table frequently used in conducting a one-way ANOVA test.
  • ANOVA with Equal Sample Sizes: Illustrates ANOVA implementation using equal sample sizes across varying marketing strategies.
  • ANOVA with Unequal Sample Sizes: Explains how ANOVA is applied with differing sample sizes assessing educational performance data.
  • The Two-Way ANOVA Table: Shows how a two-way ANOVA table is constructed to evaluate multiple factors.
  • Two-Way Analysis of Variance: Explains two factor analysis on study habit effects, illustrating split-factor methodologies.
  • Exercises: Provides exercises for practicing hypothesis testing with real-life scenarios and different datasets.
  • Conclusion: Wraps up the document thanking the reader, implicitly marking the end of the material.

HYPOTHESIS

TESTING
Ms. Anna Marie T. Ensano, MME
CASTEDSWM Faculty
Universidad de Sta. Isabel, Naga City
Why test hypothesis?

 Any individual has his own statement to claim.

 A teacher believes that students perform better to class using game


simulation than the ordinary classroom setting.

 A scientist firmly believes that the ozone layer is already deteriorating and
has hazardous effects on mankind.

 A psychologist claims the substantial relationship between the student’s self-


concept and learning.

 The truth of these statements is in question.

 Statistics helps us decide whether to accept or reject one’s statement.


What is a Hypothesis?

 A statement with many faceted ideas that are still to be


resolved whether they are true or not .

 Subjected to testing statistically.

 If it is to be true, then it is not rejected, but if it is


found to be false, it is rejected.
Types of Hypothesis
 Null Hypothesis (Ho) – must always express the idea of non-significant
difference or relationship.
Ex.: The mean weight of a newly born baby is equal to 7 lbs.
The performance of the recent graduates in the board exam
does not differ from that of last year.

 Alternative Hypothesis (Ha/H1) – the negation of the null hypothesis.


Ex.: The mean weight of a newly born baby is not equal to 7 lbs.
The performance of the recent graduates in the board exam
is better than that of last year.
Types of Error

 Type I – Rejecting the null hypothesis when in fact


it is true.

 Type 2 – Not rejecting the null hypothesis when in


fact it is false.
Level of Significance

 Alpha (α) - The probability of committing a type I error.

 Beta (β) – the probability of committing a type II error.

Choice of the value of the significant level ranges from 1% (0.01) to 5%


(0.05), depending on the risk the researcher is willing to take for making a
type I error.

A 0.01 level of significance means the researcher is giving 1% error in his


decision. Moreover, it implies that he is 99% confident of his decision to be
right.
Types of Tests

 One-tailed Test is a one-sided test hypothesis. (directional)


Ha: “greater than”, “less than”, “better”, “more than”,
etc.

 Two-tailed Test is a two-sided test hypothesis. (non-directional)


Ha: “not equal”
If Ha/H1 is “not equal”
If Ha/H1 is “less than” (<)
If Ha/H1 is “greater than” (>)
7 Steps in Hypothesis Testing

1. Formulate a null hypothesis and the alternative


hypothesis.
2. Set the level of significance.
3. Identify the type of test.
4. Determine the tabular value.
5. Compute for the required statistical test.
6. Decide whether to not reject or reject the null
hypothesis.
7. Make a Conclusion.
Small Sample Hypothesis Test for
the Mean of a Normal Population
EXAMPLE 1

A perfume company claims that the best


selling perfume contains at most 25%
alcohol. Twenty bottles were selected and
found to have a mean of 29. 7% and standard
deviation of 4.8%. Test the claim of the
perfume company at the 0.05 level of
significance.
CRITICAL VALUES t (Tabular Value)
Testing the Difference between
Two Sample Means
Example 2

A businessman prospects two possible locations for a new


restaurant. A study was conducted and focused on the pedestrian traffic
at both areas. At each location, the pedestrian are observed in 1-hour
units and, for each hour, an index of desirable characteristics is
compiled. The sample units are given below:
Magsaysay Centro
n1 = 25 n2 = 30
x1 = 245 x2 = 438
s1 = 103 s2 = 79
Test the claim that the characteristic mean indices in Magsaysay
are less than the mean indices in Centro. Use 0.05 significance level.
ACTIVITY

 A pill is supposed to contain 25.4 mg. of Phenobarbital. A random


sample of 30 pills yields a mean and standard deviation of 19.7 mg.
and 0.95 mg. respectively. Are these pills acceptable at 0.05
significance level?

 Ten randomly volunteers tested a new diet. Mean weight before and
after the diet were obtained. The obtained mean weight before the
diet was 65.4 kg with standard deviation of 7.8 kg. while after the
diet, mean weight was 62.5 kg. with standard deviation of 0.3 kg. At
0,01 level of significance, test the claim that the diet is effective;
that is, the weights become lower after the diet.
ANOVA
Analysis of Variance

 An extension of t-test used to test the non-


significant difference of means.
 Used to test the non-significance of difference
among three or more groups.
 A statistical procedure for testing the
hypothesized equality of three or more
population means.
THE ONE-WAY ANOVA TABLE

SOURCE SUM OF DEGREES MEAN F-VALUE


OF SQUARES OF SQUARE
DISTANCE FREEDOM
TREATMENT Sum of Squares c–1 MST=SST/c-1 F = MST/MSE
(Between) Treatment

ERROR Sum of Squares n–c MSE=SSE/n-c


(Within) Error

TOTAL SSTO n–1


ANOVA with Equal Sample Sizes

 Three new marketing strategies have been implemented at company


xyz. The president wants to find out the effectiveness of the
marketing strategies through analysing the sales. Test whether the
means of the sales of the product as a results of the implementation
of the new marketing strategies are equal. Use the level of
significance at 0.05. The sales of the product using the three
marketing strategies are given below.

 Marketing Strategy 1: 34, 55, 82


 Marketing Strategy 2: 57, 71, 86
 Marketing Strategy 3: 27, 34, 72
ANOVA with Unequal Sample Sizes

 The data in the following table represent final


grades given by the three professors in statistics
course. Does the difference in sample mean the
same? Use a 5% level of significance.

 Professor 1: 63, 45, 73, 77, 72


 Professor 2: 67, 45, 76, 80, 70, 70
 Professor 3: 97, 97, 87, 84, 74, 74, 64
THE TWO-WAY ANOVA TABLE
SOURCE OF DEGREES OF SUM OF SQUARES MEAN SQUARE F-RATIO
VARIATION FREEDOM

FACTOR A a–1 SSA MSA=SSA/(a-1) Fa = MSA/MSE

FACTOR B b–1 SSB MSB=SSB/(b-1) Fb = MSB/MSE

INTERACTION (a-1)(b-1) SSAB MSAB=SSAB/ Fab=MSAB/MSE


AB (a-1)(b-1)

ERROR N - ab SSE MSE=SSE/


(N-ab)

TOTAL N–1 SSTO


Two-Way Analysis of Variance

 We wish to compare the effects of study habits on


memory retention of the students. Each variable has
different categories: study habit has classification of
“with music” and “without music”, while the memory
FACTOR A High Moderate Low
retention can be classified as high, moderate and low
FACTOR B 10 9 8
retention. 15 12 10
With Music 14 11 7
13 10 10
12 13 10

16 12 12
Without Music 17 10 10
12 11 11
13 13 12
14 9 10
Exercise
 Assembly employees were divided into 3 categories according to the
number of years in service. Productivity records from each category were
gathered. Test whether longer employment means higher productivity at
1% significant level. Use the following data:
Under 1 year: 90, 70, 80, 81, 74, 78
1 – 5 years: 92, 75, 84, 92, 88
More than 5 years: 112, 75, 120, 94
 Three car models are studied in the test that involves 5 of each model. For
each of the three model cars in each of the five samples, exactly 1 gallon
of gas is placed in the tank and the car is driven until the gas is used up.
Test the claim that the 3 car models have equal distance travelled. Use α =
0.05.
Model 1: 19, 22, 17, 18, 20
Model 2: 15, 17, 20, 16, 19
Model 3: 18, 24, 20, 20, 19
THANK YOU!

You might also like