Chapter 7 Comparison of two independent samples

Similar documents
AMS7: WEEK 7. CLASS 1. More on Hypothesis Testing Monday May 11th, 2015

HYPOTHESIS TESTING II TESTS ON MEANS. Sorana D. Bolboacă

Comparison of Two Population Means

Stat 529 (Winter 2011) Experimental Design for the Two-Sample Problem. Motivation: Designing a new silver coins experiment

Hypothesis Testing. Hypothesis: conjecture, proposition or statement based on published literature, data, or a theory that may or may not be true

Sampling Distributions: Central Limit Theorem

Non-parametric (Distribution-free) approaches p188 CN

The One-Way Independent-Samples ANOVA. (For Between-Subjects Designs)

Review for Final. Chapter 1 Type of studies: anecdotal, observational, experimental Random sampling

Statistics: CI, Tolerance Intervals, Exceedance, and Hypothesis Testing. Confidence intervals on mean. CL = x ± t * CL1- = exp

Hypothesis testing I. - In particular, we are talking about statistical hypotheses. [get everyone s finger length!] n =

Chapter 7 Class Notes Comparison of Two Independent Samples

Hypothesis testing: Steps

SEVERAL μs AND MEDIANS: MORE ISSUES. Business Statistics

CIVL /8904 T R A F F I C F L O W T H E O R Y L E C T U R E - 8

Hypothesis testing: Steps

Lecture 7: Hypothesis Testing and ANOVA

T.I.H.E. IT 233 Statistics and Probability: Sem. 1: 2013 ESTIMATION AND HYPOTHESIS TESTING OF TWO POPULATIONS

Sampling distribution of t. 2. Sampling distribution of t. 3. Example: Gas mileage investigation. II. Inferential Statistics (8) t =

Chapter 8 Class Notes Comparison of Paired Samples

Two-Sample Inferential Statistics

Review: General Approach to Hypothesis Testing. 1. Define the research question and formulate the appropriate null and alternative hypotheses.

MBA 605, Business Analytics Donald D. Conant, Ph.D. Master of Business Administration

Topic 22 Analysis of Variance

1 Statistical inference for a population mean

Nonparametric tests. Timothy Hanson. Department of Statistics, University of South Carolina. Stat 704: Data Analysis I

Statistics Primer. ORC Staff: Jayme Palka Peter Boedeker Marcus Fagan Trey Dejong

Chapter 9 Inferences from Two Samples

7.2 One-Sample Correlation ( = a) Introduction. Correlation analysis measures the strength and direction of association between

Lecture 18: Analysis of variance: ANOVA

Introduction to Statistical Data Analysis III

CBA4 is live in practice mode this week exam mode from Saturday!

Visual interpretation with normal approximation

Outline. PubH 5450 Biostatistics I Prof. Carlin. Confidence Interval for the Mean. Part I. Reviews

Single Sample Means. SOCY601 Alan Neustadtl

8.1-4 Test of Hypotheses Based on a Single Sample

Business Statistics. Lecture 10: Course Review

The t-statistic. Student s t Test

ME3620. Theory of Engineering Experimentation. Spring Chapter IV. Decision Making for a Single Sample. Chapter IV

GROUPED DATA E.G. FOR SAMPLE OF RAW DATA (E.G. 4, 12, 7, 5, MEAN G x / n STANDARD DEVIATION MEDIAN AND QUARTILES STANDARD DEVIATION

Two-sample t-tests. - Independent samples - Pooled standard devation - The equal variance assumption

The t-test: A z-score for a sample mean tells us where in the distribution the particular mean lies

Statistics: revision

Nonparametric Statistics

Introduction to Business Statistics QM 220 Chapter 12

DETAILED CONTENTS PART I INTRODUCTION AND DESCRIPTIVE STATISTICS. 1. Introduction to Statistics

Content by Week Week of October 14 27

Analysis of Variance: Part 1

ANOVA - analysis of variance - used to compare the means of several populations.

Business Statistics: Lecture 8: Introduction to Estimation & Hypothesis Testing

The Difference in Proportions Test

The Chi-Square Distributions

CHAPTER 17 CHI-SQUARE AND OTHER NONPARAMETRIC TESTS FROM: PAGANO, R. R. (2007)

An inferential procedure to use sample data to understand a population Procedures

The Chi-Square Distributions

Chapter 15: Nonparametric Statistics Section 15.1: An Overview of Nonparametric Statistics

IT 403 Statistics and Data Analysis Final Review Guide

Statistics Handbook. All statistical tables were computed by the author.

df=degrees of freedom = n - 1

TABLES AND FORMULAS FOR MOORE Basic Practice of Statistics

2 and F Distributions. Barrow, Statistics for Economics, Accounting and Business Studies, 4 th edition Pearson Education Limited 2006

Exam 2 (KEY) July 20, 2009

Rama Nada. -Ensherah Mokheemer. 1 P a g e

Non-parametric methods

SMAM 314 Practice Final Examination Winter 2003

Chapter 9. Inferences from Two Samples. Objective. Notation. Section 9.2. Definition. Notation. q = 1 p. Inferences About Two Proportions

Wilcoxon Test and Calculating Sample Sizes

HYPOTHESIS TESTING. Hypothesis Testing

Soc 3811 Basic Social Statistics Second Midterm Exam Spring Your Name [50 points]: ID #: ANSWERS

STA 101 Final Review

Last two weeks: Sample, population and sampling distributions finished with estimation & confidence intervals

CHAPTER 9, 10. Similar to a courtroom trial. In trying a person for a crime, the jury needs to decide between one of two possibilities:

Solutions exercises of Chapter 7

CHAPTER 8. Test Procedures is a rule, based on sample data, for deciding whether to reject H 0 and contains:

Comparison of two samples

SMAM 314 Exam 3 Name. F A. A null hypothesis that is rejected at α =.05 will always be rejected at α =.01.

Section 9.4. Notation. Requirements. Definition. Inferences About Two Means (Matched Pairs) Examples

Basic Statistics. 1. Gross error analyst makes a gross mistake (misread balance or entered wrong value into calculation).

Chapters 4-6: Inference with two samples Read sections 4.2.5, 5.2, 5.3, 6.2

CH.9 Tests of Hypotheses for a Single Sample

Two sample Hypothesis tests in R.

Mock Exam - 2 hours - use of basic (non-programmable) calculator is allowed - all exercises carry the same marks - exam is strictly individual

CHAPTER 7. Hypothesis Testing

Transition Passage to Descriptive Statistics 28

Introduction to the Analysis of Variance (ANOVA) Computing One-Way Independent Measures (Between Subjects) ANOVAs

Hypothesis Testing. We normally talk about two types of hypothesis: the null hypothesis and the research or alternative hypothesis.

Data Analysis and Statistical Methods Statistics 651

Statistical methods for comparing multiple groups. Lecture 7: ANOVA. ANOVA: Definition. ANOVA: Concepts

Non-Parametric Two-Sample Analysis: The Mann-Whitney U Test

STATISTICS 4, S4 (4769) A2

Last week: Sample, population and sampling distributions finished with estimation & confidence intervals

Inferences About Two Proportions

Lecture 28 Chi-Square Analysis

Chapter 24. Comparing Means


Introduction to hypothesis testing

ECO220Y Review and Introduction to Hypothesis Testing Readings: Chapter 12

Statistics 251: Statistical Methods

Analysis of variance (ANOVA) Comparing the means of more than two groups

PSY 307 Statistics for the Behavioral Sciences. Chapter 20 Tests for Ranked Data, Choosing Statistical Tests

Transcription:

Chapter 7 Comparison of two independent samples 7.1 Introduction Population 1 µ σ 1 1 N 1 Sample 1 y s 1 1 n 1 Population µ σ N Sample y s n 1, : population means 1, : population standard deviations N 1, N : number of elements in each population n 1, n : sample sizes y 1, y : sample means s 1, s : sample standard deviations 1

Type of populations : 1. Those that occur naturally - observational study. Those created by intervention - experimental study Q: What do we want to do in Chapter 7? A: Compare samples from different populations. Q: How do we compare the samples? A: Using complementary approaches i.e. Method 1: Confidence Interval approach Method : Hypothesis testing approach Q: For two quantitative variables, what can we compare? A: Their 1. means,. standard deviations (variance) and 3. shapes

7. Standard error of y 1 y Basic Idea Chapter 6 y estimates To measure how precise y estimates we calculated SE y s n Chapter 7 y 1 estimates 1 (population 1) y estimates (population ) To measure how precise y 1 and y estimates 1 and respectively, we can calculate SE y 1 s 1 n 1 and SE y s n...but... Q: How do we compare the sample means y 1 and y? A: We look at their difference i.e. we look at y 1 y (or y y 1 )...THUS... y 1 y estimates 1 (or y y 1 estimates 1 ) 3

And now we can ask: Q:How precise does y 1 y estimate 1? A:We need to look at SE y 1 y Formula for the unpooled standard error of y 1 y SE y 1 y s 1 n 1 s n s 1 n 1 s n SE y 1 SE y SE y 1 y SE y 1 SE y Graph SE y y ( ) 1 SE y1 SE y1 Note: The 90 angle represents independent 4

samples 5

Formula for the pooled standard error of y 1 y Suppose we have populations i.e. N 1 Population 1 1 1 N Population Assume: 1 and 1 versus Assume: 1 and 1 6

Assume we take a random sample from each population i.e. n 1 Sample 1 n Sample y 1 s 1 y s Q: Since we assume that 1, how can we combine s 1 from sample 1 and s from sample, to obtain a pooled estimate for the variance in the populations and at the same also take into account that the sample sizes differ i.e. n 1 n.? A: s pooled n 1 1 s 1 n 1 s n 1 1 n 1 n 1 1 s 1 n 1 s n 1 n 7

Therefore, becomes SE y 1 y s 1 n 1 s n SE pooled s pooled n 1 s pooled n s pooled n 1 1 n 1 8

Method 1 7.3 Confidence Interval for ( 1 In chapter 6: Confidence interval for y t df, SE y df n 1 degrees of freedom 1 100% confidence level In chapter 7: Confidence interval for 1 y 1 y t df, SE y 1 y df degrees of freedom 1 100% confidence level Q:How do we find df? A: There are three methods: 1. df SE 1 SE SE4 1 n 1 1 SE 4 n 1 (a good approximation). smaller of n 1 1 and n 1 i.e. df min n 1 1, n 1 (bit conservative actual CL is larger) 3. df n 1 n (bit liberal actual CL is 9

smaller) 10

Conditions for the confidence interval to be valid 1. random samples. The random samples should be independent from each other 3. The random samples should be from normal populations 11

Method 7.4 Hypothesis Testing The hypothesis testing procedure consists of 5 steps: 1. The Null and Alternative Hypothesis H 0 : H A :. Choose the significance level 3. Calculate the t S -test statistic t S 4. Calculate the p-value p-value 5. Conclusion 1

Step 1 The null and alternative hypothesis H 0 : The null hypothesis is 1 1 0 i.e. the two population means are equal H A : The alternative hypothesis is 1 1 0 i.e. the two population means are not equal Step Choose the significance level ( ) Typical choices are : 0.1, 0.05 and 0.01which is similar to a 90%, 95% and 99% confidence interval. Q: When do we choose? A: Before we start the procedure i.e. at the beginning. Q: How do we use the? A: If p-value we reject H 0 If p-value we do not reject H 0 13

Step 3 Calculate the t s -test statistic t s y 1 y 0 t SE df y 1 y 14

Step 4 Calculate the p-value Take Note: The t-table on p677 only gives the upper tail area. Step 5 Conclusion Compare the p-value with (the significance level). Conclusion follow. Take note: We only say that we rejet or do not rejet H 0. We NEVER say that we ACCEPT 15

either one of the hypothesis. 16

Take Note: (p4) 1. The Null and Alternative Hypothesis H 0 : 1 c where c is a constant H A : 1 c. Choose the significance level 3. Calculate the t S -test statistic t s y 1 y c t SE df y 1 y 4. Calculate the p-value p-value 5. Conclusion 17

7.5 Further Discussion on the t-test. Relationship between t-test (Method ) and the confidence interval (Method 1) Both methods uses y 1 y SE y 1 y t df, 0.05 if we look at a 95% CI or an alpha of 0.05 i.e. 0. 05 Method 1: We do not reject H 0 : 1 if the confidence interval includes zero. Method : We do not reject H 0 : 1 if the p-value. Which is similar to: We do not reject H 0 : 1 if the test statistic, t s t df, 18

Q: How do we know this last statement is true? A: t s and p value t df, and 19

Thus, we fail to reject H 0 on an % level of significance if and only if t s t df, y 1 y SE y 1 y t df, y 1 y t df, SE y 1 y t df, SE y 1 y y 1 y t df, SE y 1 y y 1 y t df, SE y 1 y 0 y 1 y t df, SE y 1 y Now we see that Method 1 is similar to Method. ( denoted if and only if statements) 0

Type I and type II errors TRUE situation OUR decision H 0 true H 0 false Do not reject H 0 Type II error ( ) Reject H 0 Type I error ( ) P Type I error R reject H 0 H 0 true P Type II error R do not reject H 0 H 0 false Power 1 P Type II error R do not reject H 0 H 0 false 1

7.6 One-tailed t Tests Two tailed (previous section) H 0 : 1 H A : 1 Non-directional alternative One tailed (this section) H 0 : 1 H A : 1 1 Directional alternative You will know the direction before you collect the data

The One-tailed hypothesis testing procedure 1. H 0 : 1 H A : 1?. choose your own value 3. Calculate test statistic t S y 1 y 0 t SE df y 1 y 4. Calculate the p value Take note: This is where textbook specify a -step procedure y 1 y y 1 y y 1 y 5. Conclusion 3

7.11 The Wilcoxon-Mann-Whitney Test This test is also used to compare independent samples and is a competitor for the t-test The test can be used for populations that do not have a normal distribution Distribution free Since we do not specifically use the mean or the median of the sample Non-parametric...BUT... We still look for a difference in location i.e. we still look at the degree of separation / shift between samples The Null hypothesis and the Alternative hypothesis Let Y 1 denote the observations from sample 1 and let Y denote the observations from sample H 0 : The population distributions of Y 1 and Y are the same. The test statistic to be used: U S 1. large value: two samples are well separated 4

. small value: two samples are not that well separated 5

Example 7.39 (p90) Soil respiration and plant growth at two different locations in a forest. Growth Gap 17 0 170 315 9 13 16 190 64 15 18 14 6 The procedure: 1. Arrange the observations in increasing order. a. K 1 counts the number of observation in sample that is less that each observation in sample 1 b. K 1 counts the number of observation in sample 1 that is less that each observation in sample c. As a check we use: K 1 K n 1 n n 1 : number of observations in sample 1 n : number of observations in sample 3. U S max K 1, K 4. Determine the critical value with n the larger sample size and 6

n the smaller sample size 7

The procedure applied to example 7.39: X Y 1 Y X 5 17 6 0 6 0 13 0 6.5 14 0 8 64 15 0 8 170 16 0 8 190 18 1 8 315.5 n 1 7 9 3 n 8 K 1 49. 5 K 6. 5 U S max K 1, K max 49. 5 ; 6. 5 49. 5 Take note: You should be able to find the p value for directional as well as non-directional alternatives 8

Chapter 7: Exercises 7.3 7.4 7.5 7.6 7.7 7.11 7.1 7.14 7.19 p5 p31 7.3 7.4 7.5 7.6 7.7 7.9 7.46 7.47 7.48 7.49 7.50 7.5 p43 p63 7.77 7.78 7.79 7.80 p96 Take note: These exercises is part of the textbook and can be included in any class test, semester test or exam! 9