Sample Size Determination (Two or More Samples)

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

Common Large/Small Sample Tests 1/55

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

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

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

Data Analysis and Statistical Methods Statistics 651

Sampling Distributions, Z-Tests, Power

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

Direction: This test is worth 150 points. You are required to complete this test within 55 minutes.

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

Properties and Hypothesis Testing

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

Tools Hypothesis Tests

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

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

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

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

Math 140 Introductory Statistics

1036: Probability & Statistics

- 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

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

Important Formulas. Expectation: E (X) = Σ [X P(X)] = n p q σ = n p q. P(X) = n! X1! X 2! X 3! X k! p X. Chapter 6 The Normal Distribution.

Topic 9: Sampling Distributions of Estimators

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

Stat 200 -Testing Summary Page 1

Stat 319 Theory of Statistics (2) Exercises

Lecture 5: Parametric Hypothesis Testing: Comparing Means. GENOME 560, Spring 2016 Doug Fowler, GS

Topic 18: Composite Hypotheses

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

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

Chapter 13, Part A Analysis of Variance and Experimental Design

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

Final Examination Solutions 17/6/2010

1 Inferential Methods for Correlation and Regression Analysis

Topic 9: Sampling Distributions of Estimators

Sample Size Estimation in the Proportional Hazards Model for K-sample or Regression Settings Scott S. Emerson, M.D., Ph.D.

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

Exam II Review. CEE 3710 November 15, /16/2017. EXAM II Friday, November 17, in class. Open book and open notes.

TABLES AND FORMULAS FOR MOORE Basic Practice of Statistics

Comparing your lab results with the others by one-way ANOVA

Samples from Normal Populations with Known Variances

NCSS Statistical Software. Tolerance Intervals

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.

Frequentist Inference

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

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

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

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

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

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

Chapter 13: Tests of Hypothesis Section 13.1 Introduction

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

TABLES AND FORMULAS FOR MOORE Basic Practice of Statistics

Topic 9: Sampling Distributions of Estimators

Chapter 6 Sampling Distributions

Statistical Intervals for a Single Sample

Estimation of a population proportion March 23,

Successful HE applicants. Information sheet A Number of applicants. Gender Applicants Accepts Applicants Accepts. Age. Domicile

Parameter, Statistic and Random Samples

Sampling Error. Chapter 6 Student Lecture Notes 6-1. Business Statistics: A Decision-Making Approach, 6e. Chapter Goals

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4

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

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

Chapter 22: What is a Test of Significance?

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

A statistical method to determine sample size to estimate characteristic value of soil parameters

Regression, Inference, and Model Building

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

Last Lecture. Wald Test

Lesson 2. Projects and Hand-ins. Hypothesis testing Chaptre 3. { } x=172.0 = 3.67

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

6 Sample Size Calculations

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

Chapter 23: Inferences About Means

Chapter 8: Estimating with Confidence

Parameter, Statistic and Random Samples

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

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

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

(7 One- and Two-Sample Estimation Problem )

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

STATISTICAL INFERENCE

Statistics 511 Additional Materials

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures

Describing the Relation between Two Variables

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

MA238 Assignment 4 Solutions (part a)

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

SDS 321: Introduction to Probability and Statistics

MIT : Quantitative Reasoning and Statistical Methods for Planning I

Formulas and Tables for Gerstman

Chapter 2 Descriptive Statistics

Statistics 300: Elementary Statistics

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

f(x i ; ) L(x; p) = i=1 To estimate the value of that maximizes L or equivalently ln L we will set =0, for i =1, 2,...,m p x i (1 p) 1 x i i=1

Design of Engineering Experiments Chapter 2 Basic Statistical Concepts

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

t distribution [34] : used to test a mean against an hypothesized value (H 0 : µ = µ 0 ) or the difference

CHAPTER 2. Mean This is the usual arithmetic mean or average and is equal to the sum of the measurements divided by number of measurements.

Expectation and Variance of a random variable

Transcription:

Sample Sie Determiatio (Two or More Samples) STATGRAPHICS Rev. 963 Summary... Data Iput... Aalysis Summary... 5 Power Curve... 5 Calculatios... 6 Summary This procedure determies a suitable sample sie for estimatig or testig hypotheses cocerig ay of the followig:. the differece betwee the meas ad of two ormal distributios.. the ratio of the stadard deviatios ad of two ormal distributios. 3. the differece betwee the proportios ad of two biomial distributios. 4. the differece betwee the rates ad of two Poisso distributios. 5. the pairwise differeces betwee the meas of more tha two ormal distributios. It fids a sample sie that achieves either of two goals:. geerates a cofidece iterval for the differece or ratio of specified width.. yields the desired power i a test of hypotheses cocerig the differece or ratio. Sample StatFolio: samsie.sgp Sample Data: Noe. 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) -

STATGRAPHICS Rev. 963 Data Iput The first dialog displayed by this procedure is used to specify the problem of iterest to the aalyst. Compare: the problem of iterest. It is assumed that radom samples of sie j will be take from j populatios that follow the specified distributio ad used to estimate or test the value of the idicated parameters. The procedure will determie suitable values for j. Hypothesied Differece or Ratio: the aticipated value of the differece or ratio. If performig a hypothesis test, this value forms the ull hypothesis (usually ). If costructig a cofidece iterval, this value is oly used if the desired width of the iterval is specified i relative (percetage) terms. Hypothesied Withi-Group Sigmas: the aticipated value of the stadard deviatio withi each of the j populatios sampled, assumed to be the same for all populatios. Whe comparig or more meas, this value is a critical part of the calculatio ad should either be kow exactly or be a reliable estimate from previous data. Hypothesied Meas: a approximate value for the meas j. This value is ot used i the calculatios. Hypothesied Proportios: a approximate value for the biomial proportios. This value is used to determie the likely stadard error of the differece betwee the two proportios. Hypothesied Rates: a approximate value for the Poisso rates. This value is used to determie the likely stadard error of the differece betwee the two rates. 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) -

STATGRAPHICS Rev. 963 Number of Meas: the umber of samples k whe comparig more tha meas. Percet of Data i First Sample: whe comparig two samples, the percet of data i the first sample: % () Except i rare cases, the percetage should be set to 5%. For example, the above dialog box idicates a desire to compare the meas of ormal distributios, thought to be aroud = with stadard deviatios of = 3. The ull hypothesis is that the differece betwee the meas ( - ) equals. Equal sample sies for the samples are desired. The secod dialog box elicits iformatio about the goal of the aalysis: Cotrol: specifies the goal from amog the followig choices:. Absolute error: a cofidece iterval for the differece or ratio is to be costructed. That iterval should ot deviate from the poit estimate of the differece or ratio i either directio by more tha the absolute distace W idicated. Note: whe comparig more tha meas, the iterval used is based o Tukey s multiple compariso method.. Relative error: a cofidece iterval for the differece or ratio is to be costructed. That iterval should ot deviate from the poit estimate of the differece or ratio i either 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 3

STATGRAPHICS Rev. 963 directio by more tha the relative percetage P idicated. This is idetical to the absolute error case with W set equal to P times the specified differece or ratio. 3. Power: a hypothesis test is to be performed. The power of the test (-) % should equal the percetage specified whe the true value of the differece or ratio deviates from the ull hypothesis by the idicated = Differece to Detect. Power is defied as the probability of rejectig the ull hypothesis whe it is false. If a two-sided test is to be performed, the that power must be achieved both above ad below the value specified by the ull hypothesis. Note: whe comparig more tha meas, power refers to the F test for betwee group differeces i the ANOVA table ad refers to the largest differece betwee ay meas. 4. Sample Sie: the predetermied sample sie, assumed to be the same for all samples. This optio is used to plot the power curve for a sample sie that was ot calculated by this procedure. Cofidece Level: the level of cofidece (-)% used whe costructig cofidece itervals. The value is also used as the level of Type I error whe testig hypotheses. A Type I error occurs whe the ull hypothesis is falsely rejected. Alterative Hypothesis: select Not Equal for a two-sided hypothesis test, Less Tha if the alterative hypothesis is that the parameter is less tha the value specified by the ull hypothesis, or Greater Tha if the alterative hypothesis is that the parameter is greater tha the value specified by the ull hypothesis. Sigma: whe comparig or testig ormal meas, whether the stadard deviatio is assumed to be kow ( test) or if it will be estimated from the data (t test). For example, the dialog box above idicates that the followig test is to be performed: Null hypothesis H : = Alt. hypothesis H A : The probability of a Type I error (rejectig a true ull hypothesis) is set to = 5%, while the probability of a Type II error (ot rejectig a false ull hypothesis) is set to = % whe the true absolute differece betwee the meas equals 3. The followig table may be helpful i rememberig how to set the error probabilities. Do Not Reject H Reject H H is true Correct decisio Type I error risk = H is false Type II error risk = Correct decisio power = - 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 4

Power STATGRAPHICS Rev. 963 Aalysis Summary The Aalysis Summary displays the desired goal ad the sample sies that will achieve it: Sample-Sie Determiatio Parameter to be estimated: differece betwee two ormal meas Desired power: 9.% for differece =. versus differece = 3. Type of alterative: ot equal Alpha risk: 5.% Sigma: 3. (to be estimated) The required sample sie is 3 observatios from sample ad 3 observatios from sample. I the curret example, samples of = 3 observatios from each populatio are required to achieve the power requested. Power Curve The Power Curve shows the power of the specified test of hypotheses for the derived sample sies. Power Curve alpha =.5, sigma = 3., =3, =3.8.6.4. -4-4 True Differece Betwee Meas It ca be see that the power (probability of rejectig the ull) is oly aroud whe the true differece is close to ero, by it rises to - whe the differece varies i either directio by the specified Differece to Detect. 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 5

STATGRAPHICS Rev. 963 Calculatios Normal Mea Cofidece Iterval If is assumed to be kow, fid the smallest ad such that W () If will be estimated from the data, fid the smallest ad such that t, W (3) Normal Mea Hypothesis Test If is assumed to be kow, fid the smallest ad such that If is to be estimated from the data, fid the smallest ad such that t, t, (4) (5) Normal Sigma Cofidece Iterval Fid the smallest ad such that F,, W (6) ad F W (7),, Normal Sigma Hypothesis Test Fid the smallest or such that 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 6

STATGRAPHICS Rev. 963 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 7 ) l( if > (8) or ) l( if < (9) Biomial Proportio Cofidece Itervals Fid the smallest ad such that W () Biomial Proportio Hypothesis Tests Fid the smallest ad such that si si () Poisso Rate Cofidece Itervals Fid the smallest ad such that W () Poisso Rate Hypothesis Tests

Fid the smallest ad such that STATGRAPHICS Rev. 963 (3) More Tha Normal Meas Cofidece Itervals Usig Tukey s T, fid the smallest commo sample sie such that: T, k, k( ) W (4) More Tha Normal Meas Hypothesis Test Fid the smallest commo sample sie for which the power of the betwee group F-test i the aalysis of variace table equals or exceeds that specified whe the largest differece betwee ay two meas equals, based o a o-cetral F distributio with o-cetrality parameter k (5) Note: for all oe-sided tests, replace by i the equatios for the hypothesis tests. 3 by StatPoit Techologies, Ic. Sample Sie Determiatio (Two Samples) - 8