Chapter 9. Key Ideas Hypothesis Test (Two Populations)

Similar documents
Statistical Inference Procedures

Comments on Discussion Sheet 18 and Worksheet 18 ( ) An Introduction to Hypothesis Testing

x z Increasing the size of the sample increases the power (reduces the probability of a Type II error) when the significance level remains fixed.

M227 Chapter 9 Section 1 Testing Two Parameters: Means, Variances, Proportions

TESTS OF SIGNIFICANCE

VIII. Interval Estimation A. A Few Important Definitions (Including Some Reminders)

20. CONFIDENCE INTERVALS FOR THE MEAN, UNKNOWN VARIANCE

Tools Hypothesis Tests

COMPARISONS INVOLVING TWO SAMPLE MEANS. Two-tail tests have these types of hypotheses: H A : 1 2

STA 4032 Final Exam Formula Sheet

Statistics and Chemical Measurements: Quantifying Uncertainty. Normal or Gaussian Distribution The Bell Curve

100(1 α)% confidence interval: ( x z ( sample size needed to construct a 100(1 α)% confidence interval with a margin of error of w:

11/19/ Chapter 10 Overview. Chapter 10: Two-Sample Inference. + The Big Picture : Inference for Mean Difference Dependent Samples

SOLUTION: The 95% confidence interval for the population mean µ is x ± t 0.025; 49

REVIEW OF SIMPLE LINEAR REGRESSION SIMPLE LINEAR REGRESSION

UNIVERSITY OF CALICUT

Chapter 9: Hypothesis Testing

18.05 Problem Set 9, Spring 2014 Solutions

S T A T R a c h e l L. W e b b, P o r t l a n d S t a t e U n i v e r s i t y P a g e 1. = Population Variance

IntroEcono. Discrete RV. Continuous RV s

CHAPTER 6. Confidence Intervals. 6.1 (a) y = 1269; s = 145; n = 8. The standard error of the mean is = s n = = 51.3 ng/gm.

Stat 3411 Spring 2011 Assignment 6 Answers

Overview. p 2. Chapter 9. Pooled Estimate of. q = 1 p. Notation for Two Proportions. Inferences about Two Proportions. Assumptions

Statistics Parameters

Difference tests (1): parametric

Tables and Formulas for Sullivan, Fundamentals of Statistics, 2e Pearson Education, Inc.

Chapter 8.2. Interval Estimation

Common Large/Small Sample Tests 1/55

STUDENT S t-distribution AND CONFIDENCE INTERVALS OF THE MEAN ( )

Confidence Intervals: Three Views Class 23, Jeremy Orloff and Jonathan Bloom

MATHEMATICS LW Quantitative Methods II Martin Huard Friday April 26, 2013 TEST # 4 SOLUTIONS

Estimation Theory. goavendaño. Estimation Theory

Elementary Statistics

Confidence Intervals. Confidence Intervals

TI-83/84 Calculator Instructions for Math Elementary Statistics

Chapter 8: Estimating with Confidence

Statistical Inference for Two Samples. Applied Statistics and Probability for Engineers. Chapter 10 Statistical Inference for Two Samples

STA Learning Objectives. Population Proportions. Module 10 Comparing Two Proportions. Upon completing this module, you should be able to:

Chapter 22. Comparing Two Proportions. Copyright 2010 Pearson Education, Inc.

Chapter 22. Comparing Two Proportions. Copyright 2010, 2007, 2004 Pearson Education, Inc.

Formula Sheet. December 8, 2011

This chapter focuses on two experimental designs that are crucial to comparative studies: (1) independent samples and (2) matched pair samples.

Statistical Equations

ON THE SCALE PARAMETER OF EXPONENTIAL DISTRIBUTION

Mathacle PSet Stats, Confidence Intervals and Estimation Level Number Name: Date: Unbiased Estimators So we don t have favorite.

Chapter 8 Part 2. Unpaired t-test With Equal Variances With Unequal Variances

m = Statistical Inference Estimators Sampling Distribution of Mean (Parameters) Sampling Distribution s = Sampling Distribution & Confidence Interval

Working with Two Populations. Comparing Two Means

HYPOTHESIS TESTS FOR ONE POPULATION MEAN WORKSHEET MTH 1210, FALL 2018

- E < p. ˆ p q ˆ E = q ˆ = 1 - p ˆ = sample proportion of x failures in a sample size of n. where. x n sample proportion. population proportion

BIOS 4110: Introduction to Biostatistics. Breheny. Lab #9

Section 9.2. Tests About a Population Proportion 12/17/2014. Carrying Out a Significance Test H A N T. Parameters & Hypothesis

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals

This is an introductory course in Analysis of Variance and Design of Experiments.

Confidence Intervals for the Population Proportion p

Inferential Statistics. Inference Process. Inferential Statistics and Probability a Holistic Approach. Inference Process.

Math 140 Introductory Statistics

Chapter 20. Comparing Two Proportions. BPS - 5th Ed. Chapter 20 1

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01

LECTURE 13 SIMULTANEOUS EQUATIONS

Fig. 1: Streamline coordinates

Interval Estimation (Confidence Interval = C.I.): An interval estimate of some population parameter is an interval of the form (, ),

LESSON 20: HYPOTHESIS TESTING

Lecture 5. Materials Covered: Chapter 6 Suggested Exercises: 6.7, 6.9, 6.17, 6.20, 6.21, 6.41, 6.49, 6.52, 6.53, 6.62, 6.63.

CE3502 Environmental Monitoring, Measurements, and Data Analysis (EMMA) Spring 2008 Final Review

MATH/STAT 352: Lecture 15

Widely used? average out effect Discrete Prior. Examplep. More than one observation. using MVUE (sample mean) yy 1 = 3.2, y 2 =2.2, y 3 =3.6, y 4 =4.

Comparing Two Populations. Topic 15 - Two Sample Inference I. Comparing Two Means. Comparing Two Pop Means. Background Reading

MTH 212 Formulas page 1 out of 7. Sample variance: s = Sample standard deviation: s = s

10-716: Advanced Machine Learning Spring Lecture 13: March 5

MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND.

Chem Exam 1-9/14/16. Frequency. Grade Average = 72, Median = 72, s = 20

(7 One- and Two-Sample Estimation Problem )

A quick activity - Central Limit Theorem and Proportions. Lecture 21: Testing Proportions. Results from the GSS. Statistics and the General Population

Statistical Inference About Means and Proportions With Two Populations

Questions about the Assignment. Describing Data: Distributions and Relationships. Measures of Spread Standard Deviation. One Quantitative Variable

Section 18: confidence interval & hypothesis testing using sample means (sigma unknown)

Agreement of CI and HT. Lecture 13 - Tests of Proportions. Example - Waiting Times

Class 23. Daniel B. Rowe, Ph.D. Department of Mathematics, Statistics, and Computer Science. Marquette University MATH 1700

1 Inferential Methods for Correlation and Regression Analysis

2 1. The r.s., of size n2, from population 2 will be. 2 and 2. 2) The two populations are independent. This implies that all of the n1 n2

CONFIDENCE INTERVALS STUDY GUIDE

Grant MacEwan University STAT 151 Formula Sheet Final Exam Dr. Karen Buro

MBACATÓLICA. Quantitative Methods. Faculdade de Ciências Económicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS

Generalized Likelihood Functions and Random Measures

AP Statistics Review Ch. 8

Homework 5 Solutions

Background Information

Statistics Problem Set - modified July 25, _. d Q w. i n

Chapter 6 Sampling Distributions

Isolated Word Recogniser

Statistics. Chapter 10 Two-Sample Tests. Copyright 2013 Pearson Education, Inc. publishing as Prentice Hall. Chap 10-1

Stat 200 -Testing Summary Page 1

Statistical treatment of test results

Chapter 10: H at alpha of.05. Hypothesis Testing: Additional Topics

On the Multivariate Analysis of the level of Use of Modern Methods of Family Planning between Northern and Southern Nigeria

Statistics 20: Final Exam Solutions Summer Session 2007

Chapter 8: STATISTICAL INTERVALS FOR A SINGLE SAMPLE. Part 3: Summary of CI for µ Confidence Interval for a Population Proportion p

University of California, Los Angeles Department of Statistics. Hypothesis testing

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE

Transcription:

Chapter 9 Key Idea Hypothei Tet (Two Populatio) Sectio 9-: Overview I Chapter 8, dicuio cetered aroud hypothei tet for the proportio, mea, ad tadard deviatio/variace of a igle populatio. However, ofte reearcher wat to compare two differet populatio. For example, urgeo may wat to try out a ew urgical techique for a certai ailmet, but they are ot ure if it will be better. To tet thi, they could take two ample of people. I oe ample, they could ue the exitig techique, ad i the other, they could ue the ew techique. By comparig urvival proportio of the two group, they could the determie whether the ew ample i ay better. I thi cae, the firt populatio i all people who would receive the exitig techique. The ecod populatio i all people who would receive the ew techique. By uig the method dicued i thi chapter, uch iferece ca be doe. Sectio 9-: Iferece About Two Proportio To ue the method decribed i thi ectio, we firt eed to rely o a few coditio that mut be met for everythig to work properly. Coditio. The proportio are take from two imple radom ample which are idepedet. Here, idepedet mea that obervatio from the firt populatio are ot related to, or paired with, obervatio from the ecod populatio.. For each of the two ample, there are at leat 5 uccee ad 5 failure. Notatio p = Populatio Proportio (Populatio ) p = Populatio Proportio (Populatio ) = Sample Size (Populatio ) = Sample Size (Populatio ) x = Number of Succee (Populatio ) x = Number of Succee (Populatio ) x x p = = Sample Proportio (Populatio ) p = = Sample Proportio (Populatio ) q p q = p = x x p = = Pooled Sample Proportio q = p The Tet The goal i to tet the hypothee give by: H 0 : p = p H : p p (or p > p or p < p ) I other word, we tet whether the proportio for each populatio are equal v. whether they are differet i ome way. Tet Statitic: ( p Z = p ) ( p p ) ( p p ), or Z = (For the tet tatitic, we aume H 0 i true, which mea p p = 0) The critical value ad P-Value come from the tadard ormal ditributio. Deciio are made i exactly the ame way a i Chapter 8.

Cloe to a electio, ballot iue #3 ad #4 are very cotroverial. Reearcher wat to ee whether there i a differece i the proportio of people who upport iue #3 ad thoe who upport iue #4. They radomly ample 00 total people. They ak 00 of thee people whether they upport iue #3, to which 56 ay ye. They ak the other 00 people whether they upport iue #4, to which 45 ay ye. I there a differece i the proportio of upporter for each iue? Tet thi with α = 0.05. Solutio From the iformatio give, we ee that: = 00, x = 56, p = 0. 56, = 00, x = 45, p = 0. 45 x x 56 45 0 p = = = = 0.505, q = 0.495 00 00 00 H 0 : p = p H : p p ( p p ) 0.56 0.45 0. Tet Statitic: Z = = = =. 56 0.505(0.495) 0.505(0.495) 0.0707 00 00 P-Value Method Sice H ha a ig, we wat to fid area above Z =.56 ad below - Z = -.56 for the tadard ormal ditributio. From the Z-Table, thi area i (0.0594) = 0.88. Now we compare thi area to α ad ee that 0.88 > 0.05. Therefore, we do ot reject H 0. I other word, we coclude that there i ot eough evidece to claim that there i a differece i proportio of upporter for the two iue. Gloria, a hairdreer, claim that he i better tha a fellow hairdreer amed Jule. They decide to ru a hypothei tet to ee if thi i true. Out of 9 of Gloria cutomer, 87 are atified with their haircut. Out of 67 of Jule cutomer, 56 are atified. Ru the tet to ee if Gloria cutomer have a higher percetage of atifactio tha Jule cutomer (α = 0.05). Solutio From the iformatio give, we ee that: = 9, x = 87, p = 0. 946, = 67, x = 56, p = 0. 836 x x 87 56 43 p = = = = 0.899, q = 0.0 9 67 59 H 0 : p = p H : p > p ( p p ) 0.946 0.836 0. Tet Statitic: Z = = = =. 7 0.899(0.0) 0.899(0.0) 0.048 9 67

P-Value Method Sice H ha a > ig, we wat to fid area above Z =.7 for the tadard ormal ditributio. From the Z-Table, thi area i 0.9884 = 0.06. Now we compare thi area to α ad ee that 0.06 < 0.05. Therefore, we reject H 0 ad coclude that Gloria ha a higher atifactio percetage tha Jule. Cofidece Iterval for p p Sometime, oe would like to etimate the differece betwee the proportio, rather tha jut eeig whether the proportio differ igificatly. To cotruct a cofidece iterval for p p, we have the followig: Poit Etimate: p p Critical Value: Z Stadard Error: α p q p q Thi give the followig cofidece iterval: p q p q p p ± Zα Coider the Gloria/Jule hairdreer example from the previou page. We had the followig quatitie: = 9, x = 87, p = 0. 946, = 67, x = 56, p = 0. 836 Therefore, a 95% cofidece iterval would be: p q p q 0.946(0.054) p p ± Zα = 0.946 0.836 ±.96 9 = 0.±.96(0.05) = 0.± 0.09997 ( 0.0, 0.) 0.836(0.64) 67 Notice that 0 i ot icluded i thi cofidece iterval. Due to thi fact, we could ay that there i a differece betwee the two proportio (i.e. we would reject H 0 i favor of H : p p ).

Sectio 9-3: Iferece About Two Mea: Idepedet Sample I imilar fahio to tetig for differece i proportio, oe may alo wih to tet for a differece i the mea of two populatio. I the iteret of time, we will coider the mot geeral cae, where the populatio tadard deviatio are ukow, ad o aumptio are made about them. Better hypothei tet exit whe thee value are both kow, or are ukow but aumed to be equal. Coult the textbook for more iformatio o thee tet. Agai, certai requiremet mut be met for the techique dicued i thi ectio to be theoretically oud. Coditio. σ ad σ are ukow ad o aumptio i made about the equality of σ ad σ.. The two ample are idepedet. 3. Both ample are imple radom ample. 4. Either both populatio are ormally ditributed or ad are both greater tha 30. Notatio µ = Populatio Mea (Populatio ) µ = Populatio Mea (Populatio ) = Sample Size (Populatio ) = Sample Size (Populatio ) x = Sample Mea (Populatio ) x = Sample Mea (Populatio ) = Sample Stadard Deviatio (Populatio ) = Sample Stadard Deviatio (Populatio ) Degree of Freedom = df = mi(, ) (o df i till, but here i the maller of the two ample ize) The Tet The goal i to tet the hypothee give by: H 0 : µ = µ H : µ µ (or µ > µ or µ < µ ) I other word, we tet whether the mea for each populatio are equal v. whether they are differet i ome way. Tet Statitic: ( x t = x ) ( ) ( x x ) µ µ, or t = (For the tet tatitic, we aume H 0 i true, which mea µ µ = 0) The critical value ad P-Value come from the Studet t ditributio with Degree of Freedom = mi(, ). Deciio are made i exactly the ame way a i Chapter 8. Reearcher wat to ee if the average moey pet per week by tourit i Chicago i le tha the average moey pet per week by tourit i New York City. They take a ample of 60 tourit from Chicago ad 56 tourit from NYC. Of the Chicago tourit, average pedig wa $635, with a tadard deviatio of $50. Of the NYC tourit, the average pedig wa $650, with a tadard deviatio of $30. Ru a hypothei tet with α = 0.05. Solutio From the iformatio give, we ee that: = 60, x = 635, = 50, = 56, x = 650, = 30, df = mi(, ) = 56 = 55 H 0 : µ = µ H : µ < µ ( x x ) 635 650 5 Tet Statitic: t = = = =. 974 50 30 7.599 60 56

P-Value Method Sice H ha a < ig, we wat to fid area below t = -.974 for the t ditributio. Now, otice that the t-table doe ot allow oe to directly fid thi area. However, for df = 55 we ee that -.004 ha a area below of 0.05 ad -.673 ha a area below of 0.05. Sice -.004 < t < -.673, the p-value will fall betwee 0.05 ad 0.05. (ee picture) Now we compare thi area to α ad ee that 0.05 < p < 0.05 = α. Sice p < 0.05, we reject H 0. The evidece ugget that Chicago tourit pay le per week tha thoe i New York City. Oe quetio o everyoe mid i whether there i a differece i the average umber of pet owed by Columbu familie ad the average umber of pet owed by Clevelad familie. Reearcher et out to awer thi importat quetio. They ampled 35 Columbu familie ad 48 Clevelad familie. Of the Columbu familie, the average umber of pet wa.4, with a tadard deviatio of.4. Of the Clevelad familie, the average umber of pet wa.9, with a tadard deviatio of 0.9. Ru a hypothei tet (α = 0.05) to ee if there i a differece i the average umber of pet owed by familie i the two citie. Solutio From the iformatio give, we ee that: = 35, x =. 4, =.4, = 48, x =. 9, = 0.9, df = mi(, ) = 35 = 34 H 0 : µ = µ H : µ µ ( x x ).4.9 0.5 Tet Statitic: t = = = =. 85.4 0.9 0.700 35 48 P-Value Method Sice H ha a ig, the p-value will be the area above t =.85 ad below - t = -.85 for the t ditributio. Agai, otice that the t-table doe ot allow oe to directly fid thi area. However, for df = 34 we ee that.69 ha a two-tailed area of 0.0 ad.03 ha a two-tailed area of 0.05. Sice.69 < t <.03, the p-value will fall betwee 0.05 ad 0.0. (ee picture) Now we compare thi area to α ad ee that p > 0.05 = α. Sice p > 0.05, we do ot reject H 0. There i ot ufficiet evidece to coclude that the average umber of pet i differet i the two citie.

Cofidece Iterval for p p A i the previou ectio, oe might like to etimate the differece betwee the mea, rather tha jut tetig for the differece. To cotruct a cofidece iterval for µ µ, we have the followig: Poit Etimate: x x Critical Value: t (df = mi(, ) ) Stadard Error: α Thi give the followig cofidece iterval: x x ± tα Coider the avg. umber of pet example from the previou page. We had the followig quatitie: = 35, x =. 4, =.4, = 48, x =. 9, = 0.9, df = mi(, ) = 35 = 34 Therefore, a 95% cofidece iterval would be:.4 x ± t =.4.9 ±.03 35 x α 0.9 48 = 0.5 ±.03(0.700) = 0.5 ± 0.5485 ( 0.049,.049) Notice that 0 i icluded i thi cofidece iterval. Due to thi fact, we could ay that there i ot a igificat differece betwee the two mea (i.e. we do ot reject H 0 i favor of H : µ µ, a i the example o the previou page).