Part IVA Analysis of Variance (ANOVA) Dr. Stephen H. Russell Weber State University.

Slides:



Advertisements
Similar presentations
Statistics Review – Part II Topics: – Hypothesis Testing – Paired Tests – Tests of variability 1.
Advertisements

Chapter 10 Estimation and Hypothesis Testing II: Independent and Paired Sample T-Test.
BPS - 5th Ed. Chapter 241 One-Way Analysis of Variance: Comparing Several Means.
CHAPTER 25: One-Way Analysis of Variance Comparing Several Means
CHAPTER 25: One-Way Analysis of Variance: Comparing Several Means ESSENTIAL STATISTICS Second Edition David S. Moore, William I. Notz, and Michael A. Fligner.
Hypothesis Testing Steps in Hypothesis Testing:
Confidence Interval and Hypothesis Testing for:
Comparing Two Population Means The Two-Sample T-Test and T-Interval.
1 Analysis of Variance This technique is designed to test the null hypothesis that three or more group means are equal.
PSY 307 – Statistics for the Behavioral Sciences
Chapter 11 Analysis of Variance
SADC Course in Statistics Comparing Means from Independent Samples (Session 12)
Statistics Are Fun! Analysis of Variance
Independent Samples and Paired Samples t-tests PSY440 June 24, 2008.
Chapter 3 Analysis of Variance
Statistics for Managers Using Microsoft® Excel 5th Edition
PSY 307 – Statistics for the Behavioral Sciences
Chapter 9 Hypothesis Testing.
BCOR 1020 Business Statistics Lecture 20 – April 3, 2008.
The t Tests Independent Samples.
Hypothesis Testing Using The One-Sample t-Test
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 10-1 Chapter 10 Analysis of Variance Statistics for Managers Using Microsoft.
Chap 10-1 Analysis of Variance. Chap 10-2 Overview Analysis of Variance (ANOVA) F-test Tukey- Kramer test One-Way ANOVA Two-Way ANOVA Interaction Effects.
Chapter 12 ANOVA.
AM Recitation 2/10/11.
The Chi-Square Distribution 1. The student will be able to  Perform a Goodness of Fit hypothesis test  Perform a Test of Independence hypothesis test.
HAWKES LEARNING SYSTEMS math courseware specialists Copyright © 2010 by Hawkes Learning Systems/Quant Systems, Inc. All rights reserved. Chapter 14 Analysis.
8 - 1 © 2003 Pearson Prentice Hall Chi-Square (  2 ) Test of Variance.
Comparing Two Population Means
Section 10.3 Comparing Two Variances Larson/Farber 4th ed1.
One Sample Inf-1 If sample came from a normal distribution, t has a t-distribution with n-1 degrees of freedom. 1)Symmetric about 0. 2)Looks like a standard.
PSY 307 – Statistics for the Behavioral Sciences Chapter 16 – One-Factor Analysis of Variance (ANOVA)
ANOVA (Analysis of Variance) by Aziza Munir
Comparing Two Variances
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 10-1 Chapter 10 Analysis of Variance Statistics for Managers Using Microsoft.
INTRODUCTION TO ANALYSIS OF VARIANCE (ANOVA). COURSE CONTENT WHAT IS ANOVA DIFFERENT TYPES OF ANOVA ANOVA THEORY WORKED EXAMPLE IN EXCEL –GENERATING THE.
One-Way ANOVA ANOVA = Analysis of Variance This is a technique used to analyze the results of an experiment when you have more than two groups.
Warsaw Summer School 2011, OSU Study Abroad Program Difference Between Means.
Analysis of Variance 1 Dr. Mohammed Alahmed Ph.D. in BioStatistics (011)
Lecture 9-1 Analysis of Variance
Chapter 12: Analysis of Variance. Chapter Goals Test a hypothesis about several means. Consider the analysis of variance technique (ANOVA). Restrict the.
Previous Lecture: Phylogenetics. Analysis of Variance This Lecture Judy Zhong Ph.D.
Copyright © 2013, 2009, and 2007, Pearson Education, Inc. Chapter 14 Comparing Groups: Analysis of Variance Methods Section 14.1 One-Way ANOVA: Comparing.
1 ANALYSIS OF VARIANCE (ANOVA) Heibatollah Baghi, and Mastee Badii.
1 Objective Compare of two population variances using two samples from each population. Hypothesis Tests and Confidence Intervals of two variances use.
Hypothesis Testing. Why do we need it? – simply, we are looking for something – a statistical measure - that will allow us to conclude there is truly.
Lecture 10: Correlation and Regression Model.
Comparing k Populations Means – One way Analysis of Variance (ANOVA)
ETM U 1 Analysis of Variance (ANOVA) Suppose we want to compare more than two means? For example, suppose a manufacturer of paper used for grocery.
Copyright © Cengage Learning. All rights reserved. 12 Analysis of Variance.
Copyright (C) 2002 Houghton Mifflin Company. All rights reserved. 1 Understandable Statistics S eventh Edition By Brase and Brase Prepared by: Lynn Smith.
Applied Quantitative Analysis and Practices LECTURE#25 By Dr. Osman Sadiq Paracha.
CHAPTER 27: One-Way Analysis of Variance: Comparing Several Means
Business Statistics: A First Course (3rd Edition)
Chapter 4 Analysis of Variance
Hypothesis test flow chart frequency data Measurement scale number of variables 1 basic χ 2 test (19.5) Table I χ 2 test for independence (19.9) Table.
Copyright © 2016, 2013, 2010 Pearson Education, Inc. Chapter 10, Slide 1 Two-Sample Tests and One-Way ANOVA Chapter 10.
Introduction to Multiple Regression Lecture 11. The Multiple Regression Model Idea: Examine the linear relationship between 1 dependent (Y) & 2 or more.
While you wait: Enter the following in your calculator. Find the mean and sample variation of each group. Bluman, Chapter 121.
Copyright © Cengage Learning. All rights reserved. 12 Analysis of Variance.
Formula for Linear Regression y = bx + a Y variable plotted on vertical axis. X variable plotted on horizontal axis. Slope or the change in y for every.
Independent Samples ANOVA. Outline of Today’s Discussion 1.Independent Samples ANOVA: A Conceptual Introduction 2.The Equal Variance Assumption 3.Cumulative.
DSCI 346 Yamasaki Lecture 4 ANalysis Of Variance.
The 2 nd to last topic this year!!.  ANOVA Testing is similar to a “two sample t- test except” that it compares more than two samples to one another.
When the means of two groups are to be compared (where each group consists of subjects that are not related) then the excel two-sample t-test procedure.
Chapter 13 f distribution and 0ne-way anova
i) Two way ANOVA without replication
Basic Practice of Statistics - 5th Edition
10 Chapter Chi-Square Tests and the F-Distribution Chapter 10
Hypothesis Tests for Two Population Standard Deviations
Presentation transcript:

Part IVA Analysis of Variance (ANOVA) Dr. Stephen H. Russell Weber State University

4A.2 Introduction to the concept of ANOVA n I wonder if there is a difference in the average amount of beef contained in 32-ounce jars of Prago (we’ll call population A) and Ragu (we’ll call population B) spaghetti sauces. H O : µ A = µ B H A : µ A  µ B Consider the data and do a “t test” of hypotheses at the.05 level of significance. We will assume the populations are normally distributed and have equal variances.

4A.3 A note on two-sample t tests... n The degrees of freedom for a one-sample problem is n – 1, as you know. n The degrees of freedom for a two-sample problem is n 1 – 1 + n 2 – 1 or n 1 + n 2 – 2 In the spaghetti sauce problem, the two sample t test the degrees of freedom would be – 2 = 9

4A.4 Grams of Beef in …. A 32-ounce jar of Prago A 32-ounce jar of Ragu These sample results yield a P-value of.003—strong evidence against the null and in favor of the alternative that these two brands are not equal. Ragu gives us more beef!

4A.5 Let’s look at this problem again in terms of variation “among samples” (between columns) and “within samples” Grams of Beef in …. A 32-ounce jar of Prago A 32-ounce jar of Ragu

4A.6 What’s the influencing factor? The brand! Does the brand matter when it comes to the amount of beef? Yes! So we say the factor matters! Grams of Beef in …. A 32-ounce jar of Prago A 32-ounce jar of Ragu

4A.7 Dependent & Independent Variables n The Dependent Variable (the variable that is “acted upon”) in this problem is the amount of meat in the spaghetti sauce. n The Independent Variable (also called “the factor”) is the brand. We say “Brand may influence the amount of meat”. So meat is “dependent” on brand. Brand is the independent variable.

4A.8 Comparing variances: The variability among columns appears to be greater than the variability within columns. Is this observation consistent with the null or the alternative? The alternative! These brands are not equal when it comes to the amount of beef. Grams of Beef in …. A 32-ounce jar of Prago A 32-ounce jar of Ragu

4A.9 The F test We want to compute a ratio of variances: What would high values for this ratio suggest? What is the expected value of this ratio if the null is true? Ratios of two variances follow a special distribution called the F Distribution. Comparing variances like this is called Analysis of Variance (ANOVA)

4A.10 The F Test  F tests are always right tailed in ANOVA problems.

4A.11 The F test Do the spaghetti sauce problem as an ANOVA problem in MINITAB

4A.12 Spaghetti Sauce Problem The test statistic (the calculated F) is The tail of rejection is found in an F Table: Degrees of freedom for the numerator = c –1 (Levels of the factor minus one). Degrees of freedom for the denominator = n – c. (Total sample size minus levels of the factor.) For this problem: Dfn = 2 – 1 = 1 Dfd = 11 – 2 = 9 What is the tail of rejection for an alpha of.05? 5.12.

4A.13 The Spaghetti Sauce Problem  The calculated F is way out in the right tail. We reject the null and conclude these two spaghetti sauces do not have equal amounts of beef

4A.14 The Spaghetti Sauce Problem A comparison of t-test and F-test results T testF test Calculated t = -4.05Calculated F = Tail of rejectionTail of rejection (with with n-2 df = dfn = 1 & dfd = 9) = 5.12 P-value =.003 P-value =.003Decision: Handily reject null NOTE: These results are the same. AND t 2 = F

4A.15 Comparing population means n Why in the heck do the complicated F test if the t test yields the same results? n Because the F test can handle more than two population means comparisons; e.g., Ho: µ 1 = µ 2 = µ 3 = µ 4 If we compared these means with t tests we’d have to do: 1 vs 2; 1 vs 3; 1 vs 4; 2 vs 3; 2 vs 4; 3 vs 4 or 6 different t tests. Here’s the problem with doing 6 “t” tests...

4A.16 At an alpha of.05 the probability of a correct decision if the null is true on any one test is or.95. The probability of six correct decisions if the null is true is.95 raised to the sixth power or.735. This means that after doing six t-tests, the probability of a Type I error is not.05. Rather it is or.265. Hence, when comparing the equality of more than two population means, we use the F test..

4A.17 Additional comments on ANOVA n ANOVA is a misleading term. ANOVA is not a test to compare population variances! n ANOVA is a very complicated area in statistics. We have discussed only One-Way ANOVA (which means one factor). n In MINITAB always click on Stat  ANOVA  One Way (Unstacked) in this class. n ANOVA tests assume n The sampled populations are normally distributed n The sampled populations have equal variances (a critical assumption for correct results)  It’s a good idea to use equal sample sizes—which minimizes the impact of violating the equal-variances assumption.

4A.18 Example problem: A furniture manufacturer wants to compare the mean drying times for four brands of stain. Each stain was applied to 10 chairs and the drying times in minutes were recorded. The hypotheses are : H O : µ 1 = µ 2 = µ 3 = µ 4 H A : Not all population means are equal Let’s use an alpha of.01. (1) What is the tail of rejection? (2) Solve the problem with MINITAB dfn = 3 dfd = 36 F for rejection = 4.39

4A.19 Homework Assignment for ANOVA Problem Set 4

4A.20 Summary of ANOVA n Analysis of variance( ANOVA): statistical technique designed to test whether the means of more than two populations are equal n Variation has two components n variation among columns, explained by the factor: measures explained variation n variation within columns, attributed to random error; measures unexplained variation n We have covered only one-way ANOVA (also called one-factor ANOVA) n ANOVA analysis assumes normal populations with equal variances.

4A.21 Homework solutions: 1.H O :  L =  M =  H H A : Not all of the population means are equal dfn = c – 1 = 2 dfd = n – c = 12 The Tail of Rejection in a F distribution is defined as 5.10 for 2.5 percent level of significance. The F statistic is 1.92, which means the variability attributable to levels of the factor is 1.92 times greater than the random variability. P-Value is.189, which is interpreted as: “If the Null is true, there is a.189 chance of observing an F statistic as contradictory (or more contradictory) to the null as the value found here.” We fail to reject the null. We do not have sufficiently strong evidence to run with the conclusion that housing prices are not the same for three areas with different levels of air pollution.

4A.22 Homework solutions: 2. H O :  Food A =  Food B =  Food C H A : Not all of the population means are equal dfn = c – 1 = 2 dfd = n – c = 15 The Tail of Rejection in a F distribution is defined as 6.36 for.01 level of significance. The F statistic is.36, which means the variability attributable to levels of the factor is.36 of the random variability;i.e., very little “factor” variability. P-Value is a HUGE.701, which is interpreted as: “If the Null is true, there is a.701 chance of observing an F statistic as contradictory (or more contradictory) to the null as the value found here.” We fail to reject the null. We do not have sufficiently strong evidence to run with the conclusion that dogs do not like these three foods equally. (In fact, random variability is greater than explained variability!)

4A.23 Homework solutions: 3.H O :  Epsilon =  Chevron =  BP H A : Not all of the population means are equal dfn = c – 1 = 2 dfd = n – c = 15 The Tail of Rejection in a F distribution is defined as 3.68 for 5 percent level of significance. The F statistic is 20.35, which means the variability attributable to levels of the factor is more than 20 times greater than the random variability. P-Value is 0.000, which is interpreted as: “If the Null is true, there is a zero chance of observing an F statistic as contradictory (or more contradictory) to the null as the value found here.” We reject the null. We have very strong evidence that these brands do not yield the same flying time.