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

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

Announcements. Unit 5: Inference for Categorical Data Lecture 1: Inference for a single proportion

Chapter 8: Estimating with Confidence

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

MATH/STAT 352: Lecture 15

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

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

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

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

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

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

Recall the study where we estimated the difference between mean systolic blood pressure levels of users of oral contraceptives and non-users, x - y.

Frequentist Inference

Chapter 11: Asking and Answering Questions About the Difference of Two Proportions

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

AP Statistics Review Ch. 8

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

Confidence Intervals for the Population Proportion p

DS 100: Principles and Techniques of Data Science Date: April 13, Discussion #10

Estimation of a population proportion March 23,

STAC51: Categorical data Analysis

Tests of Hypotheses Based on a Single Sample (Devore Chapter Eight)

Data Analysis and Statistical Methods Statistics 651

Lecture 6 Simple alternatives and the Neyman-Pearson lemma

October 25, 2018 BIM 105 Probability and Statistics for Biomedical Engineers 1

1 Inferential Methods for Correlation and Regression Analysis

Chapter 23: Inferences About Means

Common Large/Small Sample Tests 1/55

Stat 200 -Testing Summary Page 1

Chapter 22: What is a Test of Significance?

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

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 summary Conservative and approximate confidence intervals for a binomial p Examples. MATH1005 Statistics. Lecture 24. M.

Homework 5 Solutions

Topic 9: Sampling Distributions of Estimators

Understanding Dissimilarity Among Samples

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n.

π: ESTIMATES, CONFIDENCE INTERVALS, AND TESTS Business Statistics

Math 140 Introductory Statistics

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

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

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

Big Picture. 5. Data, Estimates, and Models: quantifying the accuracy of estimates.

(7 One- and Two-Sample Estimation Problem )

Instructor: Judith Canner Spring 2010 CONFIDENCE INTERVALS How do we make inferences about the population parameters?

Chapter 6 Sampling Distributions

Final Examination Solutions 17/6/2010

Parameter, Statistic and Random Samples

Statistics 511 Additional Materials

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures

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

STATISTICAL INFERENCE

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

TABLES AND FORMULAS FOR MOORE Basic Practice of Statistics

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

Mathacle. PSet Stats, Concepts In Statistics Level Number Name: Date: Confidence Interval Guesswork with Confidence

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara

S160 #12. Sampling Distribution of the Proportion, Part 2. JC Wang. February 25, 2016

Introduction to Econometrics (3 rd Updated Edition) Solutions to Odd- Numbered End- of- Chapter Exercises: Chapter 3

Statistics 20: Final Exam Solutions Summer Session 2007

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

STAT 155 Introductory Statistics Chapter 6: Introduction to Inference. Lecture 18: Estimation with Confidence

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

Class 27. Daniel B. Rowe, Ph.D. Department of Mathematics, Statistics, and Computer Science. Marquette University MATH 1700

Sample Size Determination (Two or More Samples)

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. Comments:

Confidence Interval for one population mean or one population proportion, continued. 1. Sample size estimation based on the large sample C.I.

MA238 Assignment 4 Solutions (part a)

Topic 9: Sampling Distributions of Estimators

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

Understanding Samples

Mathacle. PSet Stats, Concepts In Statistics Level Number Name: Date:

Power and Type II Error

Continuous Data that can take on any real number (time/length) based on sample data. Categorical data can only be named or categorised

MidtermII Review. Sta Fall Office Hours Wednesday 12:30-2:30pm Watch linear regression videos before lab on Thursday

S160 #12. Review of Large Sample Result for Sample Proportion

STAT431 Review. X = n. n )

Read through these prior to coming to the test and follow them when you take your test.

Last Lecture. Wald Test

Topic 9: Sampling Distributions of Estimators

1 Constructing and Interpreting a Confidence Interval

Final Review. Fall 2013 Prof. Yao Xie, H. Milton Stewart School of Industrial Systems & Engineering Georgia Tech

1 Models for Matched Pairs

Stat 225 Lecture Notes Week 7, Chapter 8 and 11

Math 152. Rumbos Fall Solutions to Review Problems for Exam #2. Number of Heads Frequency

Because it tests for differences between multiple pairs of means in one test, it is called an omnibus test.

BIOSTATISTICS. Lecture 5 Interval Estimations for Mean and Proportion. dr. Petr Nazarov

Simple Random Sampling!

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance

MIT : Quantitative Reasoning and Statistical Methods for Planning I

- 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

ST 305: Exam 3 ( ) = P(A)P(B A) ( ) = P(A) + P(B) ( ) = 1 P( A) ( ) = P(A) P(B) σ X 2 = σ a+bx. σ ˆp. σ X +Y. σ X Y. σ X. σ Y. σ n.

Stat 319 Theory of Statistics (2) Exercises

Properties and Hypothesis Testing

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

CONFIDENCE INTERVALS STUDY GUIDE

Expectation and Variance of a random variable

STAT 203 Chapter 18 Sampling Distribution Models

Y i n. i=1. = 1 [number of successes] number of successes = n

Exam II Covers. STA 291 Lecture 19. Exam II Next Tuesday 5-7pm Memorial Hall (Same place as exam I) Makeup Exam 7:15pm 9:15pm Location CB 234

Transcription:

A quick activity - Cetral Limit Theorem ad Proportios Lecture 21: Testig Proportios Statistics 10 Coli Rudel Flip a coi 30 times this is goig to get loud! Record the umber of heads you obtaied ad calculate the proportio of heads i the 30 flips. Whe you are doe come up ad mark the sample proportio o the dot plot. Usig my mystical statistical powers I predict that the distributio of sample proportios should be early ormally distributed with mea equal to approximately 0.5 ad a stadard deviatio of approximately 0.09. April 10, 2012 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 2 / 17 Sigle populatio proportio Statistics ad the Geeral Populatio Results from the GSS Sigle populatio proportio Two scietists wat to kow if a certai drug is effective agaist high blood pressure. The first scietist wats to give the drug to 1000 people with high blood pressure ad see how may of them experiece lower blood pressure levels. The secod scietist wats to give the drug to 500 people with high blood pressure, ad ot give the drug to aother 500 people with high blood pressure, ad see how may i both groups experiece lower blood pressure levels. Which is the better way to test this drug? The Geeral Social Survey asks the same questio, below is the distributio of resposes from the 2010 survey: All 1000 get the drug 99 500 get the drug 500 do t 571 Total (a) All 1000 get the drug (b) 500 get the drug, 500 do t Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 3 / 17 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 4 / 17

Parameter ad Poit Estimates Sigle populatio proportio Iferece o a Proportio Sigle populatio proportio We would like to estimate the proportio of all Americas who have a good ituitio about experimetal desig, i.e. would aswer 500 get the drug 500 do t? What are the parameter of iterest ad the poit estimate? Parameter of iterest: Proportio of all Americas who have a good ituitio about experimetal desig. p (a populatio proportio) Poit estimate: Proportio of sampled Americas who have a good ituitio about experimetal desig. ˆp (a sample proportio) Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 5 / 17 What percet of all Americas have a good ituitio about experimetal desig, i.e. would aswer 500 get the drug 500 do t? We ca aswer this research questio usig a cofidece iterval, which we kow are of the form poit estimate ± ME Ad we also kow from CI of meas that ME = critical value stadard error of the poit estimate. SEˆp =? Stadard error of a sample proportio p (1 p) SEˆp = Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 6 / 17 Idetifyig whe a sample proportio is early ormal Sample proportios are also early ormally distributed Back to experimetal desig... Cofidece itervals for a proportio The, accordig to the CLT: mea = p, SE = p (1 p) But of course this is true oly uder certai coditios... Idepedece Radomizatio 10% Coditio Nearly Normal Number of successes ( p) 10 Number of failures ( (1 p)) 10 If p is ukow (most cases), we use ˆp i the calculatio of the stadard error. The GSS foud that 571 out of (85%) of Americas aswered the questio o experimetal desig correctly. Estimate (usig a 95% cofidece iterval) the proportio of all Americas who have a good ituitio about experimetal desig. Give: =, ˆp = 0.85. First check assumptios & coditios. 1. Idepedece: The sample is radom, ad < 10% of all Americas, therefore we ca assume that oe respodet s respose is idepedet of aother. 2. Normality: 571 people aswered correctly (successes) ad 99 aswered icorrectly (failures), both are greater tha 10. Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 7 / 17 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 8 / 17

Cofidece itervals for a proportio We are give that =, ˆp = 0.85, we also just leared that the stadard p(1 p) error of the sample proportio is SE =. Which of the below is the correct calculatio of the 95% cofidece iterval? (a) 0.85 ± 1.96 (b) 0.85 ± 1.65 0.85 0.15 0.85 0.15 (c) 0.85 ± 1.96 0.85 0.15 (d) 571 ± 1.96 571 99 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 9 / 17 Choosig a sample size Choosig a sample size whe estimatig a proportio If the researchers were goig to coduct aother study o the same survey questio how may people should they sample i order to cut the margi of error of a 95% cofidece iterval dow to 1%? CI = ˆp ± ME ME = z SE = z p(1 p) z ˆp(1 ˆp) 0.01 0.85 0.15 1.96 Use estimate for ˆp from previous study 0.01 2 1.96 2 0.85 0.15 1.962 0.85 0.15 0.01 2 4898.04 should be at least 4,899 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 10 / 17 What if there is t a previous study? Choosig a sample size whe estimatig a proportio Choosig a sample size, cot. Choosig a sample size whe estimatig a proportio What should the researchers do if they are plaig to ask a ew survey questio where they have o idea what the populatio proportio might be... use p = 0.5 Why? if you do t kow ay better, 50-50 is a good guess p = 0.5 gives the most coservative estimate p(1 p) is largest whe p = 0.5 which results i the largest possible p(1-p) 0.00 0.10 0.20 0.0 0.2 0.4 0.6 0.8 1.0 Statistics 10 (Coli Rudel) Lecturep 21 April 10, 2012 11 / 17 Previously we just saw that if the researchers what a margi of error less tha 1% they will eed to sample at least 4,899 people whe they expect the populatio proportio to be ear 85%. How does this chage whe we expect the populatio proportio to be ear 50%? 0.01 0.5 (1 0.5) 1.96 0.01 2 1.96 2 0.5 0.5 1.962 0.5 0.5 0.01 2 9604 which is almost double the sample size! Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 12 / 17

Choosig a sample size whe estimatig a proportio Choosig a sample size whe estimatig a proportio Back to the Cois Example - Legalizig Marijuaa Now that we kow how to calculate the cofidece iterval for a sample proportio, CI = ˆp ± z p (1 p) Calculate a 90% cofidece iterval based o the proportio of heads you observed i the 30 flips. Does this iterval iclude 50%? What does your result tell you about the fairess of your coi? Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 13 / 17 The 2010 Geeral Social Survey also asked the questio; Do you thik the use of marijuaa should be made legal, or ot? 48% of the 1,259 respodets said it should be made legal. a) Is the umber 48% a sample statistic or a populatio parameter? Explai. b) Costruct a 95% cofidece iterval for the proportio of Americas who thik marijuaa should be made legal. c) Iterpret this cofidece iterval i the cotext of this questio. d) A critic poits out that this 95% cofidece iterval is oly accurate if the statistic follows a ormal distributio, or if the ormal model is a good approximatio. Is this true for these data? Explai. e) A ews piece o this study s fidigs states; Majority of Americas thik marijuaa should be legalized. Based o your cofidece iterval, is this ews piece s statemet justified? Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 14 / 17 Hypotheses Hypothesis testig for a proportio Which of the followig are the correct set of hypotheses for testig if more tha 80% of Americas have a good ituitio about experimetal desig? Hypothesis testig for a proportio Hypothesis testig for a proportio mea = 0.80, SE = 0.80 0.20 = 0.0154 (a) H 0 : µ = 0.80 H A : µ > 0.80 (b) H 0 : p = 0.85 H A : p > 0.85 (c) H 0 : p = 0.80 H A : p > 0.80 (d) H 0 : ˆp = 0.80 H A : ˆp > 0.80 Note: The SE is differet, because ow we are coductig a hypothesis test assumig H 0 is true, ad H 0 says p = 0.80. 0.8 0.85 sample proportios Z = 0.85 0.80 0.0154 p value = P(Z > 3.25) = 3.25 = 1 0.9994 = 0.0006 Sice p-value is less tha 0.05 we reject H 0. The data provide covicig evidece that more tha 80% of Americas have a good ituitio o experimetal desig. Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 15 / 17 Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 16 / 17

Example - Legalizig Marijuaa Hypothesis testig for a proportio We ca also employ a hypothesis test to examie whether a majority of americas support legalizig marijuaa. H 0 : p = 0.50 H A : p > 0.50 mea = 0.5, SE = 0.50 0.50 1259 = 0.014 Z = 0.48 0.5 0.014 = 1.43 p value = P(Z > 1.43) = 1 0.0764 = 0.9236 Therefore, we fail to reject the ull hypothesis sice p value > 0.05. There is ot evidece that a majority of americas support legalizig marijuaa. Statistics 10 (Coli Rudel) Lecture 21 April 10, 2012 17 / 17