Lecture 4 t-Tests. History (from Wikipedia) Introduced in 1908 by William Sealy Gosset, a chemist working for the Guinness brewery in Dublin, Ireland.

Slides:



Advertisements
Similar presentations
Introduction to the t Statistic
Advertisements

Inferential Statistics
Chapter 7: Statistical Applications in Traffic Engineering
Inference for distributions: - Comparing two means IPS chapter 7.2 © 2006 W.H. Freeman and Company.
12.5 Differences between Means (s’s known)
SADC Course in Statistics Comparing Means from Independent Samples (Session 12)
Final Jeopardy $100 $200 $300 $400 $500 $100 $200 $300 $400 $500 $100 $200 $300 $400 $500 $100 $200 $300 $400 $500 $100 $200 $300 $400 $500 LosingConfidenceLosingConfidenceTesting.
1/55 EF 507 QUANTITATIVE METHODS FOR ECONOMICS AND FINANCE FALL 2008 Chapter 10 Hypothesis Testing.
Lecture 6 Outline: Tue, Sept 23 Review chapter 2.2 –Confidence Intervals Chapter 2.3 –Case Study –Two sample t-test –Confidence Intervals Testing.
BCOR 1020 Business Statistics
Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 8-1 Business Statistics: A Decision-Making Approach 6 th Edition Chapter.
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 9-1 Chapter 9 Fundamentals of Hypothesis Testing: One-Sample Tests Basic Business Statistics.
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 10-1 Business Statistics: A Decision-Making Approach 7 th Edition Chapter.
Two Population Means Hypothesis Testing and Confidence Intervals With Unknown Standard Deviations.
S519: Evaluation of Information Systems
Two-sample problems for population means BPS chapter 19 © 2006 W.H. Freeman and Company.
Hypothesis Testing Using The One-Sample t-Test
Hypothesis Testing.
Statistical Analysis. Purpose of Statistical Analysis Determines whether the results found in an experiment are meaningful. Answers the question: –Does.
Probability Distributions and Test of Hypothesis Ka-Lok Ng Dept. of Bioinformatics Asia University.
Confidence Intervals and Hypothesis Testing - II
Descriptive statistics Inferential statistics
Hypothesis testing – mean differences between populations
Statistical Analysis Statistical Analysis
Copyright © 2006 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Slide
1 Introduction to Hypothesis Testing. 2 What is a Hypothesis? A hypothesis is a claim A hypothesis is a claim (assumption) about a population parameter:
Ttests Programming in R. The first part of these notes will address ttesting basics. The second part of these notes will address z test (or proportion.
Psy B07 Chapter 4Slide 1 SAMPLING DISTRIBUTIONS AND HYPOTHESIS TESTING.
Chapter 20 Testing hypotheses about proportions
1 ConceptsDescriptionHypothesis TheoryLawsModel organizesurprise validate formalize The Scientific Method.
Chapter 9 Tests of Hypothesis Single Sample Tests The Beginnings – concepts and techniques Chapter 9A.
Lecture 16 Section 8.1 Objectives: Testing Statistical Hypotheses − Stating hypotheses statements − Type I and II errors − Conducting a hypothesis test.
Essential Question:  How do scientists use statistical analyses to draw meaningful conclusions from experimental results?
EMIS 7300 SYSTEMS ANALYSIS METHODS FALL 2005 Dr. John Lipp Copyright © Dr. John Lipp.
10.1: Confidence Intervals Falls under the topic of “Inference.” Inference means we are attempting to answer the question, “How good is our answer?” Mathematically:
5.1 Chapter 5 Inference in the Simple Regression Model In this chapter we study how to construct confidence intervals and how to conduct hypothesis tests.
Confidence Intervals Lecture 3. Confidence Intervals for the Population Mean (or percentage) For studies with large samples, “approximately 95% of the.
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 8-1 Chapter 8 Fundamentals of Hypothesis Testing: One-Sample Tests Statistics.
Lecture 9 Chap 9-1 Chapter 2b Fundamentals of Hypothesis Testing: One-Sample Tests.
11/16/2015Slide 1 We will use a two-sample test of proportions to test whether or not there are group differences in the proportions of cases that have.
Two-Sample Hypothesis Testing. Suppose you want to know if two populations have the same mean or, equivalently, if the difference between the population.
1 9 Tests of Hypotheses for a Single Sample. © John Wiley & Sons, Inc. Applied Statistics and Probability for Engineers, by Montgomery and Runger. 9-1.
Statistical Inference for the Mean Objectives: (Chapter 9, DeCoursey) -To understand the terms: Null Hypothesis, Rejection Region, and Type I and II errors.
3-1 MGMG 522 : Session #3 Hypothesis Testing (Ch. 5)
Chapter 8 Parameter Estimates and Hypothesis Testing.
Fall 2002Biostat Statistical Inference - Confidence Intervals General (1 -  ) Confidence Intervals: a random interval that will include a fixed.
Copyright © 2006 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Slide
Tests of significance: The basics BPS chapter 14 © 2006 W.H. Freeman and Company.
Introduction to inference Tests of significance IPS chapter 6.2 © 2006 W.H. Freeman and Company.
© Copyright McGraw-Hill 2004
Statistical Inference Statistical inference is concerned with the use of sample data to make inferences about unknown population parameters. For example,
Medical Statistics Medical Statistics Tao Yuchun Tao Yuchun 7
+ Unit 6: Comparing Two Populations or Groups Section 10.2 Comparing Two Means.
Ttests INCM 9102 Quantitative Methods. Ttests The term “Ttest” comes from the application of the t-distribution to evaluate a hypothesis. Note: a “t-statistic”
Chapter 12 Tests of Hypotheses Means 12.1 Tests of Hypotheses 12.2 Significance of Tests 12.3 Tests concerning Means 12.4 Tests concerning Means(unknown.
BIOL 582 Lecture Set 2 Inferential Statistics, Hypotheses, and Resampling.
Hypothesis Tests u Structure of hypothesis tests 1. choose the appropriate test »based on: data characteristics, study objectives »parametric or nonparametric.
Chapter 7: Hypothesis Testing. Learning Objectives Describe the process of hypothesis testing Correctly state hypotheses Distinguish between one-tailed.
Statistical Inference for the Mean Objectives: (Chapter 8&9, DeCoursey) -To understand the terms variance and standard error of a sample mean, Null Hypothesis,
Hypothesis Tests for 1-Proportion Presentation 9.
T-TEST. Outline  Introduction  T Distribution  Example cases  Test of Means-Single population  Test of difference of Means-Independent Samples 
Copyright © 2013, 2009, and 2007, Pearson Education, Inc. 1 FINAL EXAMINATION STUDY MATERIAL III A ADDITIONAL READING MATERIAL – INTRO STATS 3 RD EDITION.
Statistical hypothesis Statistical hypothesis is a method for testing a claim or hypothesis about a parameter in a papulation The statement H 0 is called.
Hypothesis Testing. Steps for Hypothesis Testing Fig Draw Marketing Research Conclusion Formulate H 0 and H 1 Select Appropriate Test Choose Level.
Chapter 9 Introduction to the t Statistic
When the means of two groups are to be compared (where each group consists of subjects that are not related) then the excel two-sample t-test procedure.
Data Analysis Module: Bivariate Testing
Bivariate Testing (ttests and proportion tests)
Bivariate Testing (ttests and proportion tests)
Bivariate Testing (ttests and proportion tests)
Presentation transcript:

Lecture 4 t-Tests

History (from Wikipedia) Introduced in 1908 by William Sealy Gosset, a chemist working for the Guinness brewery in Dublin, Ireland ("Student" was his pen name because his employer regarded their use of statistics as a trade secret). Gosset had been hired due to Claude Guinness's innovative policy of recruiting the best graduates from Oxford and Cambridge to apply biochemistry and statistics to Guinness' industrial processes. [2] Gosset devised the t-test as a way to cheaply monitor the quality of stout.William Sealy GossetGuinnessbreweryDublin, Irelandpen nameOxfordCambridgebiochemistrystatistics [2]stout Gosset’s t-test was ignored until placed on firmer mathematical footing by Ronald Fisher.

Usage Used to answer these questions: Are two data sets “equivalent”? Is the experiment repeatable? In other words --- allows us to compare TWO MEANS.

My First t-Test Is the modulus of the iris different in the axial direction from the circumferential direction? Yes

Requirements The t-test, equal variance requires that the errors associated with your data are normally distributed, and the two samples are independent (i.e., “unpaired”). Before-and-after testing (e.g., experimental drug) violates the independence requirement. Need a t-test for “paired” samples. Pairing can also happen through the use of additional variable (e.g., gender) The equal variance that is part of the test name (t-test, equal variance) implies that the variance in the population(s) behind the samples is equal.

Example Set 1: SimilarSet 2: Dissimilar 1a1b2a2b mean: std. dev.:

Hypothesis: The t-test described here is a statistical test of these competing hypotheses: – H1 (Test Hypothesis): The population means behind the two samples are different. – H0 (Null Hypothesis): The population means behind the two samples are the same. A t-test is any statistical hypothesis test in which the test statistic follows a Student's t distribution if the null hypothesis is true.

In Excel… Excel 2007: Data tab / Analysis Group / Data Analysis button Choose t-Test: Two-Sample Assuming Equal Variance

Inputting…

Output… The important value is the two-tailed P-value.

What can we say? P two-tail = , i.e., the probability of the null hypothesis is 31.5%. Since this is greater than α = 0.05, we fail to reject the null hypothesis (they are the same) and can say: we cannot conclude that the population means are different (with 95% confidence) we cannot conclude that the samples are not “equivalent” we cannot conclude that the experiment is not repeatable That is as close as we can come to saying the two samples are “equivalent” based on a t-test.

N.B. (Note Well!) “we fail to reject the null hypothesis” …is not the same as… “we accept the null hypothesis that they are the same”! Informally, if P > 0.05, we say things like “the means are likely the same”, but this is not precisely correct.

Second (dissimilar) data set… P two-tail = 7.79x Since this is less than α = 0.05, we reject the null hypothesis and can say: the population means behind the two samples are different (with greater than 95% confidence) the samples are not “equivalent” the experiment is not repeatable

Picture Mean does not fall within the tails of the other distribution

Picture Mean falls within the tails of the other distribution

Informally… We have two different sets of data…e.g., blood pressure for 20 people on a drug and 20 people on sugar pills. If P < 0.05 we say the differences are significant, and If P > 0.05 we say they are insignificant. Alas, we live in a world where no one cares if two means are the same or just slightly different. In other words, informally, “fail to reject” becomes “accept”.

Warning If the P value > 0.05, we fail to reject the Null hypothesis, so we can say “we could not conclude that the population means are different”, “we could not conclude that the new data is not consistent with the published value”, or “the experimental mean is not significantly different from the published value” WE SHOULD NOT SAY: “we conclude that the experimental value agrees with the published value”, but many people do say this.

Notes Some students perform a t-test and include all three statements (bullet items) in their report. Don’t do that – pick the one statement that fits with your report. The “one-tail” P values are used for “less- than” or “greater-than” t-tests.

One-sided t-Test Calculated by the same process in Excel. For a one-sided t-test, the hypotheses being tested are: H 0 (Null Hypothesis) – the population means μ 1 and μ 2 are equal H A (Alternate (or test) Hypothesis) – the population mean μ 1 is less than population mean μ 2 Because of the “less than” comparison, you have to be careful which data set you call group 1 and which you call group 2.

One-sided t-Test In general a test is called two-sided or two- tailed if the null hypothesis is rejected for values of the test statistic falling into either tail of its sampling distribution, and it is called one- sided or one-tailed if the null hypothesis is rejected only for values of the test statistic falling into one specified tail of its sampling distributionnull hypothesissampling distributionnull hypothesissampling distribution

Question Should the one-tail P-value or the two-tail P- value be smaller?

Question Should the one-tail P-value or the two-tail P- value be smaller? One tail because you are only comparing values in one-direction (i.e., one-tail). You’re simply comparing whether two means are equal or one is less than the other --- never consider if one value is greater than the other in the one-tail test.

t-Test, unequal variance A t-test can be performed to see if the population means behind two data sets (samples) are similar enough to conclude that they could have come from the same population. One of those data sets might have a mean value with zero standard deviation – that is, it might be a constant.

t-Test, unequal variance This t-test is used to answer this question: Is my experimental result “equivalent to” (or “consistent with”) a published value? …informally Requires: the errors associated with your data are normally distributed, and the two samples are independent (unpaired).

Hypotheses H1 (Test Hypothesis): The population means behind the two samples are different – implying that the new data is inconsistent with the published value being tested. H0 (Null Hypothesis): The population means behind the two samples are the same – implying that the new data is consistent with the published value being tested.

In Excel…

Excel results… we could not conclude that the new data is not consistent with the published value of 43% Most engineers would say that the new data supports the published value of 43% efficiency.

Different published value… the new data is not consistent with the published value of 49%

Notes Saying that the new data is “consistent” with the published result of 43% efficiency does not mean that the new data is inconsistent with every other value. In fact, the new data shown in Table 1 is consistent with efficiencies ranging from 43% to 47%. t-Tests are “simpler” for saying two things are different than for saying two means are the same.

Statistically Significant ≠ Significant

References 2 biggest mistakes… – Failing to reference statements/claims that are not common knowledge. Cite anything you (or another student) look up. – Putting references in footnotes (is it 1980?). Put references inline with a complete list at the end. – Example: “… [1]”, numbered reference – Example: “… (Lamport, 1994)”, author-date – Example: “… (Lamport, p. 18)”, author-page

Reference List Grose, T. K. and J.A. Doe, “Engineering their Way to the Top,” ASEE Prism 12 (September 2002), p. 20. List all authors with initials (list only first author in the inline citation).

Reference List Grose, T. K. and J.A. Doe, “Engineering their Way to the Top,” ASEE Prism 12 (September 2002), p. 20. Article or chapter title in quotes

Reference List Grose, T. K. and J.A. Doe, “Engineering their Way to the Top,” ASEE Prism 12 (September 2002), p. 20. Book or journal title in italic

Reference List Grose, T. K. and J.A. Doe, “Engineering their Way to the Top,” ASEE Prism, 12(1), 2002, p. 20. Volume, issue (if available), year, pages

References Other examples… Grose, T. K. and J.A. Doe, “Engineering their Way to the Top,” ASEE Prism 12(1), 2002, p. 20. Hibbeler, R. C., Engineering Mechanics. Statics & Dynamics, 9 th ed., Prentice-Hall, Inc., Upper Saddle River, New Jersey, 2001, p. 82. Sir Winston Churchill, Quotes and Stories (no date), Retrieved March 30, 2003, from Turabian, K. L., J.B. Smith, and J.A. Doe, A Manual for Writers of Term Papers, Theses, and Dissertations, 5 th ed., The University of Chicago Press, Chicago, Illinois, 1987, pp I often move the (date) after the authors. The INFO pack uses a slight variation Any option is fine, be consistent and complete