SADC Course in Statistics Comparing Means from Paired Samples (Session 13)

Slides:



Advertisements
Similar presentations
Introductory Mathematics & Statistics for Business
Advertisements

1 Session 8 Tests of Hypotheses. 2 By the end of this session, you will be able to set up, conduct and interpret results from a test of hypothesis concerning.
SADC Course in Statistics Analysis of Variance for comparing means (Session 11)
SADC Course in Statistics Common Non- Parametric Methods for Comparing Two Samples (Session 20)
SADC Course in Statistics Basic summaries for demographic studies (Session 03)
Basic Sampling Concepts
SADC Course in Statistics Estimating population characteristics with simple random sampling (Session 06)
The Poisson distribution
SADC Course in Statistics Comparing several proportions (Session 15)
Overview of Sampling Methods II
SADC Course in Statistics Further ideas concerning confidence intervals (Session 06)
SADC Course in Statistics Introduction to Non- Parametric Methods (Session 19)
SADC Course in Statistics Tests for Variances (Session 11)
Assumptions underlying regression analysis
SADC Course in Statistics Basic principles of hypothesis tests (Session 08)
SADC Course in Statistics The binomial distribution (Session 06)
SADC Course in Statistics Sampling weights: an appreciation (Sessions 19)
SADC Course in Statistics Inferences about the regression line (Session 03)
SADC Course in Statistics Importance of the normal distribution (Session 09)
SADC Course in Statistics Revision of key regression ideas (Session 10)
Correlation & the Coefficient of Determination
SADC Course in Statistics Samples and Populations (Session 02)
SADC Course in Statistics Confidence intervals using CAST (Session 07)
SADC Course in Statistics Sample size determinations (Session 11)
SADC Course in Statistics Multi-stage sampling (Sessions 13&14)
SADC Course in Statistics Assessing data critically Module B1 Session 17.
SADC Course in Statistics Graphical summaries for quantitative data Module I3: Sessions 2 and 3.
SADC Course in Statistics Choosing appropriate methods for data collection.
SADC Course in Statistics Comparing two proportions (Session 14)
SADC Course in Statistics Linking tests to confidence intervals (and other issues) (Session 10)
SADC Course in Statistics Preparing & presenting demographic information: 2 (Session 06)
SADC Course in Statistics Basic Life Table Computations - II (Session 13)
SADC Course in Statistics Review and further practice (Session 10)
SADC Course in Statistics Revision using CAST (Session 04)
SADC Course in Statistics Introduction to Statistical Inference (Session 03)
SADC Course in Statistics (Session 09)
SADC Course in Statistics General approaches to sample size determinations (Session 12)
SADC Course in Statistics To the Woods discussion (Sessions 10)
Objectives and data needs
SADC Course in Statistics Review of ideas of general regression models (Session 15)
SADC Course in Statistics Developing a sampling strategy (Session 05)
SADC Course in Statistics Case Study Work (Sessions 16-19)
SADC Course in Statistics Setting the scene (Session 01)
SADC Course in Statistics Objectives and analysis Module B2, Session 14.
SADC Course in Statistics Revision on tests for means using CAST (Session 17)
SADC Course in Statistics Revision on tests for proportions using CAST (Session 18)
Probability Distributions
9.4 t test and u test Hypothesis testing for population mean Example : Hemoglobin of 280 healthy male adults in a region: Question: Whether the population.
Simple Linear Regression and Correlation by Asst. Prof. Dr. Min Aung.
Module 16: One-sample t-tests and Confidence Intervals
CHAPTER 15: Tests of Significance: The Basics Lecture PowerPoint Slides The Basic Practice of Statistics 6 th Edition Moore / Notz / Fligner.
Two Sample Hypothesis Testing for Proportions
SADC Course in Statistics Introduction and Study Objectives (Session 01)
Hypothesis Testing Steps of a Statistical Significance Test. 1. Assumptions Type of data, form of population, method of sampling, sample size.
SADC Course in Statistics Comparing Means from Independent Samples (Session 12)
ESTIMATION AND HYPOTHESIS TESTING: TWO POPULATIONS
SADC Course in Statistics Paddy results: a discussion (Session 17)
CHAPTER 10 ESTIMATION AND HYPOTHESIS TESTING: TWO POPULATIONS Prem Mann, Introductory Statistics, 7/E Copyright © 2010 John Wiley & Sons. All right reserved.
Hypothesis testing – mean differences between populations
Two Sample Tests Nutan S. Mishra Department of Mathematics and Statistics University of South Alabama.
CHAPTER 14 Introduction to Inference BPS - 5TH ED.CHAPTER 14 1.
SADC Course in Statistics Forecasting and Review (Sessions 04&05)
BPS - 3rd Ed. Chapter 141 Tests of significance: the basics.
Introduction to Hypothesis Testing: the z test. Testing a hypothesis about SAT Scores (p210) Standard error of the mean Normal curve Finding Boundaries.
Chapter 9: Hypothesis Tests for One Population Mean 9.2 Terms, Errors, and Hypotheses.
Hypothesis Testing with TWO Samples. Section 8.1.
MATH 2311 Section 8.3.
Section 11.2: Carrying Out Significance Tests
Section 8.3 Addition Popper 34: Choice A for #1-10
MATH 2311 Section 8.3.
Presentation transcript:

SADC Course in Statistics Comparing Means from Paired Samples (Session 13)

To put your footer here go to View > Header and Footer 2 Learning Objectives By the end of this session, you will be able to recognise whether two samples are paired or independent explain why there is a gain in precision with paired samples carry out a paired t-test and interpret results from such a test present and report on conclusions from paired t-tests

To put your footer here go to View > Header and Footer 3 Paired samples - aims In comparing two samples we would aim to improve the precision of the comparison wherever possible, i.e. reduce the standard error used in the test statistic. e.g. If the aim is to compare the average weight of males with that of females amongst malnourished children, it would be better to assess pairs of children of the same age, having one male and one female in each pair.

To put your footer here go to View > Header and Footer 4 Benefits of pairing The paired approach means that gender difference would be clearer within a pair of the same age, thus removing age to age variability from the comparison. The situation could be further improved by matching the children, not only by age, but other characteristics too, e.g. their location (rural or urban), levels of poverty, etc. Matching leads to a study of the differences between each pair, say d i for pair i.

To put your footer here go to View > Header and Footer 5 An example Suppose we have data for the mean annual income (in 1000s), of doctors and dentists in the UK, from 10 different regions. The data (fictitious) are given in the table. Region DoctorsDentists

To put your footer here go to View > Header and Footer 6 Null and alternative hypotheses The hypotheses to be tested are: H 0 : = 0 versus H 1 : This is equivalent to H 0 : d = 0 versus H 1 : d 0 where d refers the mean of the population of differences between the two groups. Thus, the two-sample case reduces to a single sample situation. Hence ideas covered in Session 09 applies to the single sample formed from differences between each set of pairs.

To put your footer here go to View > Header and Footer 7 Test procedure The ten differences are found to be: Further, = 1.22, while = Hence the t-statistic for testing H 0 is: t = = = 3.15 Comparing with t-tables with 9 d.f. shows this result is significant at the 2% significance level. The exact p-value = 0.012

To put your footer here go to View > Header and Footer 8 Conclusions and other issues Conclusion: There is some evidence of a difference between the mean salaries at a 2% significance level. Note: If we had ignored the pairing by region and conducted an independent samples (or two-sample t-test), the test statistic is t= 1.95 on 18 d.f. This is clearly non- significant, thus leading to a different conclusion from above. So take care to recognise pairing where it occurs.

To put your footer here go to View > Header and Footer 9 References Armitage, P., Matthews J.N.S. and Berry G. (2002). Statistical Methods in Medical Research. 4th edn. Blackwell. Clarke, G.M. and Cooke, D. (2004). A Basic Course in Statistics. 5th edn. Edward Arnold. Johnson, R.A. and Bhattacharyya, G.K. (2001). Statistics Principles and Methods. 4th edn. Wiley.

To put your footer here go to View > Header and Footer 10 Some practical work follows…