Sample Size Copyright  2008 by The McGraw-Hill Companies. This material is intended for educational purposes by licensed users of LearningStats. It.

Slides:



Advertisements
Similar presentations
Statistics for Business and Economics
Advertisements

McGraw-Hill Ryerson Copyright © 2011 McGraw-Hill Ryerson Limited. Adapted by Peter Au, George Brown College.
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 7-1 Chapter 7 Confidence Interval Estimation Statistics for Managers.
Point and Confidence Interval Estimation of a Population Proportion, p
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Confidence Interval Estimation Basic Business Statistics 10 th Edition.
Chapter Topics Confidence Interval Estimation for the Mean (s Known)
8-1 Copyright ©2011 Pearson Education, Inc. publishing as Prentice Hall Chapter 8 Confidence Interval Estimation Statistics for Managers using Microsoft.
Copyright ©2011 Pearson Education 8-1 Chapter 8 Confidence Interval Estimation Statistics for Managers using Microsoft Excel 6 th Global Edition.
C82MCP Diploma Statistics School of Psychology University of Nottingham 1 Overview Central Limit Theorem The Normal Distribution The Standardised Normal.
Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.Chap 8-1 Statistics for Managers Using Microsoft® Excel 5th Edition.
Sample size. Ch 132 Sample Size Formula Standard sample size formula for estimating a percentage:
Determining the Size of
Business Statistics, A First Course (4e) © 2006 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Confidence Interval Estimation Business Statistics, A First Course.
Statistics for Managers Using Microsoft® Excel 7th Edition
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 7-1 Chapter 7 Confidence Interval Estimation Statistics for Managers.
1 BIOSTAT 6 - Estimation There are two types of inference: estimation and hypothesis testing; estimation is introduced first. The objective of estimation.
Basic Business Statistics, 11e © 2009 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Confidence Interval Estimation Basic Business Statistics 11 th Edition.
Confidence Intervals (Chapter 8) Confidence Intervals for numerical data: –Standard deviation known –Standard deviation unknown Confidence Intervals for.
Confidence Interval Estimation
Chap 8-1 Copyright ©2013 Pearson Education, Inc. publishing as Prentice Hall Chapter 8 Confidence Interval Estimation Business Statistics: A First Course.
© 2003 Prentice-Hall, Inc.Chap 6-1 Business Statistics: A First Course (3 rd Edition) Chapter 6 Sampling Distributions and Confidence Interval Estimation.
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Confidence Interval Estimation Basic Business Statistics 11 th Edition.
Confidence Intervals about a Population Proportion
Statistical Fundamentals: Using Microsoft Excel for Univariate and Bivariate Analysis Alfred P. Rovai Estimation PowerPoint Prepared by Alfred P. Rovai.
Biostatistics Class 6 Hypothesis Testing: One-Sample Inference 2/29/2000.
McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. Sampling Distributions and Estimation (Part 2) Chapter88 Sample Size Determination.
1 Chapter 12: Inference for Proportions 12.1Inference for a Population Proportion 12.2Comparing Two Proportions.
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 8-1 Confidence Interval Estimation.
Statistical Fundamentals: Using Microsoft Excel for Univariate and Bivariate Analysis Alfred P. Rovai Estimation PowerPoint Prepared by Alfred P. Rovai.
Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 7-1 4th Lesson Estimating Population Values part 2.
: An alternative representation of level of significance. - normal distribution applies. - α level of significance (e.g. 5% in two tails) determines the.
Chap 8-1 Chapter 8 Confidence Interval Estimation Statistics for Managers Using Microsoft Excel 7 th Edition, Global Edition Copyright ©2014 Pearson Education.
Statistics for Managers Using Microsoft Excel, 5e © 2008 Pearson Prentice-Hall, Inc.Chap 8-1 Statistics for Managers Using Microsoft® Excel 5th Edition.
Various Topics of Interest to the Inquiring Orthopedist Richard Gerkin, MD, MS BGSMC GME Research.
8-1 Confidence Intervals Chapter Contents Confidence Interval for a Mean (μ) with Known σ Confidence Interval for a Mean (μ) with Unknown σ Confidence.
Lesoon Statistics for Management Confidence Interval Estimation.
Basic Business Statistics, 11e © 2009 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Confidence Interval Estimation Business Statistics: A First Course 5 th Edition.
Chapter 8 Confidence Interval Estimation Statistics For Managers 5 th Edition.
Class Six Turn In: Chapter 15: 30, 32, 38, 44, 48, 50 Chapter 17: 28, 38, 44 For Class Seven: Chapter 18: 32, 34, 36 Chapter 19: 26, 34, 44 Quiz 3 Read.
Chapter 8 Confidence Interval Estimation. 8.1 Confidence Interval Estimation of the Mean This section deals with the case of known σ. There are two kinds.
Sampling and Sampling Distribution
Process Control Charts
CHAPTER 8 Estimating with Confidence
CHAPTER 8 Estimating with Confidence
Chapter 7 Confidence Interval Estimation
Sampling Distributions
Understanding Sampling Distributions: Statistics as Random Variables
Chapter 6 Inferences Based on a Single Sample: Estimation with Confidence Intervals Slides for Optional Sections Section 7.5 Finite Population Correction.
ESTIMATION.
Confidence Interval Estimation
One-Sample Hypothesis Tests
Two-Sample Hypothesis Testing
Confidence Intervals Copyright  2008 by The McGraw-Hill Companies. This material is intended for educational purposes by licensed users of LearningStats.
Sampling Distributions and Estimation
Sample Size Determination
CHAPTER 10 Estimating with Confidence
Medical Estimation Copyright  2008 by The McGraw-Hill Companies. This material is intended for educational purposes by licensed users of LearningStats.
Goodness-of-Fit Tests
Estimating
CONCEPTS OF ESTIMATION
Sampling Distribution of the Sample Mean
CHAPTER 8 Estimating with Confidence
Confidence Interval Estimation
Estimating a Population Proportion
Chapter 8: Estimating with Confidence
CHAPTER 8 Estimating with Confidence
CHAPTER 8 Estimating with Confidence
CHAPTER 8 Estimating with Confidence
CHAPTER 8 Estimating with Confidence
CHAPTER 8 Estimating with Confidence
Presentation transcript:

Sample Size Copyright  2008 by The McGraw-Hill Companies. This material is intended for educational purposes by licensed users of LearningStats. It may not be copied or resold for profit.

Sample Size For a mean: For a proportion: where n = the required sample size e = the desired error z = z value for the desired confidence level. Comment Since s is often unknown, a small-scale pilot study may be required to estimate s using s. Comment Since p is unknown, a conservative assumption is to set p = 0.5. This will guarantee a large enough sample.

Sample Size for a Proportion Problem How large a sample should be taken to estimate the proportion of dissatisfied HMO patients with an error of  0.03 and 98% confidence? Solution Since p is unknown, set p = 0.5. For 98% confidence we use z = 2.326. For a desired error of  0.03, the required sample size is 1503. Note If an estimate p is available from a preliminary sample, use it (especially if you are pretty sure that p is not 0.50).

Effect of Confidence Level Problem How large a sample should be taken to estimate the proportion of dissatisfied HMO patients with an error of  0.03? The sample size increases with the desired confidence level. Comment Samples sizes for a proportion are often quite large. For example, national political polls that require error rates of 2-3% often survey over 1000 people.

Sample Size for a Mean Problem How large a sample should be taken to estimate the mean waiting time of patients at a clinic with an error of  1.5 minutes and 95% confidence? Solution The formula requires the standard deviation, so we take a small preliminary sample and obtain s = 8.37 minutes. For 95% confidence we use z = 1.960. For a desired error of 1.5, the required sample size is 120. Note To estimate s, a small preliminary sample is probably an inevitable pre-task.

Effect of Confidence Level Problem How large a sample should be taken to estimate the mean waiting time of patients at a clinic with an error of  1.5 minutes? The sample size increases with the desired confidence level.

Tradeoffs in Sample Size Problem The client says it would be too expensive and time-consuming to survey 1503 patients. The statistician is asked for an alternative. Solution The statistician suggests lowering the confidence level to 90% (using z = 1.645) and raising the desired error to 5%. The resulting sample size is 271. The client agrees that this will suffice, and appreciates the reduced cost of data gathering.

But what if you can’t guess R? The Empirical Rule Problem How large a sample should be taken to estimate the mean waiting time of patients at a clinic with an error of  1.5 minutes and 95% confidence? We do not know s. Solution Instead of taking a small preliminary sample, if we know the approximate range (R) we can set s = R/6, since for a normal distribution the range is about m  3s. If we assume that patients wait between 0 and 60 minutes, we could set s = 60/6 = 10 (a “good guess” about s). For 95% confidence and a desired error of 1.5, the required sample size is 171. But what if you can’t guess R?

Problems with the Empirical Rule Problem Recent research suggests that the rule for estimating s = R/6 is often too conservative, i.e., it may result in sample sizes that are too small. A better rule might be to set s = R/4. Alternatively, just take a small preliminary sample and use s as an estimate of s. Suggested Reading Richard H. Browne, “Using the Sample Range as a Basis for Calculating Sample Size in Power Calculations,” The American Statistician, Vol. 55, No. 4, November, 2001, pp. 293-298.

Finite Populations For a mean: For a proportion: Finite population adjustment: Comment If the population is finite and sampling is without replacement, the sample size should be adjusted. However, this adjustment will have little effect if the sample size is less than 5% of the population (i.e., you can ignore the adjustment if n/N < 0.05).

Other Sample Size Issues The scientific aura of sample size formulas may divert attention from more fundamental issues such as: Goals of the proposed study Statistical significance vs. importance Practical and ethical criteria However, these formulas can help structure the dialogue between statistician and client. Suggested Reading Russell V. Lenth, “Some Practical Guidelines for Effective Sample Size Determination,” The American Statistician, Vol. 55, No. 3, August, 2001, pp. 187-193.

Commercial Software Multi-purpose software packages (e.g., SAS, SPSS, Minitab, Statistica) will perform sample size and power calculations. You might also try specialized vendors such as these: Biostat http://www.poweranalysis.com NCSS http://www.ncss.com nQuery http://www.statsolusa.com StatXact http://www.cytel.com Source Recent copies of Amstat News. This list is not intended to be suggestive, not comprehensive.