# Register now to access 7 million high quality study materials (What's Course Hero?) Course Hero is the premier provider of high quality online educational resources. With millions of study documents, online tutors, digital flashcards and free courseware, Course Hero is helping students learn more efficiently and effectively. Whether you're interested in exploring new subjects or mastering key topics for your next exam, Course Hero has the tools you need to achieve your goals.

32 Pages

### EPI-820_Lect7_Stat_Inference

Course: EPI 820, Fall 2008
School: Michigan State University
Rating:

Word Count: 1704

#### Document Preview

Evidence-Based EPI-820 Medicine LECTURE 7: CLINICAL STATISTICAL INFERENCE Mat Reeves BVSc, PhD 1 Objectives Understand the theoretical underpinnings and the flaws associated with the current approach to clinical statistical testing (the frequentist approach). Understand the difference between testing and estimation Understand the advantages of the CI and the CI functions. Understand the logic of a Bayesian...

Register Now

#### Unformatted Document Excerpt

Coursehero >> Michigan >> Michigan State University >> EPI 820

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
Evidence-Based EPI-820 Medicine LECTURE 7: CLINICAL STATISTICAL INFERENCE Mat Reeves BVSc, PhD 1 Objectives Understand the theoretical underpinnings and the flaws associated with the current approach to clinical statistical testing (the frequentist approach). Understand the difference between testing and estimation Understand the advantages of the CI and the CI functions. Understand the logic of a Bayesian Approach 2 Personal Statistical History. Post-DVM Clue-less. Sceptical of the role of statistics Thinks research = the search for P < 0.05 PhD Era: Increasing obsession with stat methods Lots of tools! SLR, ANOVA, MLR, LR, LL & Cox Thinks statistics = real science Post-PhD: Healthy scepticism for the way stats are used Stats = methods which have inherent limitations Not a substitute for clear scientific thought or understanding the scientific method 3 Review of Significance Tests Substantive hypothesis: Cows on BST will tend to gain weight Null hypothesis (Ho): the mean body wt. of cows trt with BST is not different from the mean body wt. of control cows Ux = Uy Alternative hypothesis (Ha): the mean body wt. of cows trt with BST is different from the mean body wt. of control cows Ux Uy 4 Review of Significance Tests - Logically, if Ho is refuted Ha is confirmed - investigator seeks to 'nullify' Ho Expt: 20 cows randomized to BST (X) and control (Y). Measure wt. gain. Calculate mean wt. change per group. 5 Review of Significance Tests Assumptions: i) Sample statistic (X - Y) is one instance of an infinitely large number of sample statistics obtained from an infinite number of replications of the expt., under the same conditions (frequentist assumption) ii) Populations are normally distributed, equal variance iii) The Ho is true 6 Review of Significance Tests (t-test) t = X Y S xy Where: N (0, 1) df = (n1 1) (n2 1) Sxy= ( 11 + ). S 2 n1 n2 = standard error of the difference between two independent means. S2 = estimate of pooled population variance - t may take on any value, no value is logically inconsistent with Ho! Smaller t values are more consistent with Ho being true. - all else equal, larger ns increase value of t (higher power). 7 Review of Significance Tests Large values of t indicate: i) test assumptions are true, a rare event has occurred ii) one of the assumptions of the test is false, and by convention it is assumed that the Ho is not true. - By convention, relative frequency of t where we decide to choose (ii) above as a logical conclusion is set to 5% (alpha level or significance level) - Expt: t = 2.55, p = 0.02, reject Ho - result is significant 8 Review of Significance Tests - Type 1 error (alpha), occurs 5% of the time when Ho is true - Type II error (beta), occurs B% of the time when Ho is false - Alpha and beta are inversely related - Fixing alpha at 5%, means Sp is 95% - Beta is not set 'a priori, hence Se (power) tends to be low - Scientific caution dictates that set alpha small - Scientific ignorance dictates we ignore beta! 9 Alpha and beta are inversely related 10 Relationship between diagnostic test result and disease status DISEASE PRESENT (D+) ABSENT (D-) POSITIVE (T+) TP ab cd FN FP PVP= a a+b PVN= d c+d TEST NEGATIVE (T-) TN Se= a/a + c Se= P(T+|D+) Sp= d/b + d Sp= P(T-|D-) 11 Relationship between significance test results and truth TRUTH Ho False Ho True REJECT Ho TP (1 - B) FP Type I (a) SIGNF. TEST ACCEPT Ho PVP= TP TP + FP FN Type II (B) Se= TP/TP + FN Se= Power (1 - B) TN (1 - a) PVN= TN TN + FN Sp= TN/TN + FP 12 Power - Probability of rejecting Ho when Ho is false - Se = TP/(TP + FN) or (1 - B) - Power is a function of: i) Alpha (increase by making Ha one sided i.e., Ux > Uy) (consistent with changing the cut-off value) ii) Reliability (as measured by SE of the difference) - Power increases with decreasing SE - SE decreases with increasing sample size (= decr variance) iii) Size of treatment effect 13 The Consequences of Low Power i) difficult to interpret negative results - truly no effect - expt unable to detect true difference ii) increase proportion of type 1 errors in literature iii) fail to identify many important associations iv) low power means low precision (indicated by the confidence interval) 14 Questions? What proportion of statistically significant findings published in the literature are false positive (Type 1) errors? What well known measure is this proportion? and, what elements does this figure therefore depend on? 15 Hypothetical outcomes of 500 experiments, a= 0.05, Power= 0.50, and 20% prevalence of false Hos TRUTH Ho FALSE Ho TRUE REJECT Ho 50 20 SIGNF. TEST ACCEPT Ho PV+ = 50/70 = 71% 50 100 Se = 50% 380 400 Sp = 95% 16 N = 500 If all signf. results published, 29% are Type 1 errors The P value - probability of obtaining a value of the test statistic (X) at least as large as the one observed, given the Ho is true - P (>=X | Ho true) Common Incorrect Interpretations - It is NOT P (Ho true|Data)!!! - We can never state the probability of a hypothesis being true! (under the frequentist approach) - The probability that the results were due to chance! 17 Criticisms of Significance Tests i) Decision vs Inference (Neyman-Pearson) - pioneers of modern statistics were interested in producing results enabled that decisions to be made - problem of automatic acceptance or rejection based on an arbitrary cutoff (P= 0.04 vs P=0.06) - results should adjust your degree of belief in a hypothesis rather than forcing you to accept an artificial dichotomy - "intellectual economy" 18 Criticisms of Significance Tests ii) Asymmetry of significance tests - frequently, the experimental data can be found to be consistent with a Ho of no effect or a Ho of a 20% increase - acceptance of both Ho's given the data leads to 2 very different conclusions! - asymmetry was recognized by Fisher, hence convention is to identify theory with the Ha but to test the Ho - Is there an effect? is the wrong question! Should ask: What is the size of the effect? 19 Criticisms of Significance Tests iii) Corroborative power of significance tests - Both Fisherian and Neyman-Pearson schools make no assumption about the prior probability of Ho - Both schools presume Ho is almost always false - rejection of Ho does nothing to illuminate which of the vast number of Has are supported by the data! - Failing to reject Ho does not prove Ho is true (Popper: 'we can falsify hypotheses but not confirm them') 20 Criticisms of Significance Tests iv) Effect size and significance tests - Test statistics and p values are a function of both effect size and sample size - Cannot infer size of an effect by inspection of the P value reporting P< 0.00001 has no scientific merit! - Highly significant results may be derived from trivial effects if sample size is large. - Confidence intervals give plausible range for the unknown popl parameter (signf tests show what the parameter is not!) 21 Relationship between the Size of the Sample and the Size of the P Value Example RCT: Intervention: new a/b for pneumonia. Outcome: Recovery Rate = % of patients in clinical recovery by 5 days Facts: Known = Existing drug of choice results in 35% recovery rate at 5 days Unknown = New drug improves recovery rate by 5% (to 40%) 22 P values Generated by RCT by Sample Size Sample Size (N = 2x) 100 500 600 700 800 1000 P value (Chi-square) 0.465 0.103 0.074 0.053 0.039 0.021 23 Conclusion? Significance testing should be abandoned and replaced with interval estimation (point estimate and CI)! Why? - not couched in pseudo-scientific hypothesis testing language - do not imply any decision making implications - give plausible range to unknown popl parameter - gives clue as to sample size (width of the CI) - avoids danger of inferring a large effect when result if highly significant 24 Interval estimation - view "experimentation" as a measurement exercise - want an unbiased, precise measure of effect - Point estimate: best estimate of the true effect, given the data (aka MLE) and it indicates the magnitude of effect (but is imprecise) - Confidence intervals indicate degree of precision of estimate. Represent a set of all possible values for the parameter that are consistent with the data - width of CI depends on variability and level of confidence (%) 25 Interval estimation - 90% CI: - 90% of such intervals will include the true unknown popl. parameter (necessary frequentist interpretation...

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

Michigan State University - EPI - 820
EPI-820 Evidence-Based MedicineLECTURE 8: PROGNOSIS Mat Reeves BVSc, PhD1Objectives: 1. Review definitions. 2. Understand concept of natural history and inception cohort studies. 3. Define commonly used measures of prognosis. 4. Understand o
Michigan State University - EPI - 826
EPI 826 Fall, 1998 Final Assignment DUE: December 16, 1998 Consider the enclosed data for 137 patients from a lung cancer trial. The variables in this data set are listed as follows: Variable # 1 2 3 4 5 6 7 8 9 10 11 1. 2. Variable Name Treatment La
Michigan State University - EPI - 826
MOCK FINAL EXAM HM 826 Prostate Data Set Treatment of prostate cancer depends on the extent of spread of the cancer to the lymph nodes. The extent of this involvement is determined by a laprotomy. However, this is a surgical procedure. An alternative
Michigan State University - EPI - 826
Alka IndurkhyaEPI 8269/1/99Lecture 2 Analysis of Frequency Data- r c contingency tables; SAS code for r x c analysis Rosner: Chapter 10 (continued) 1. We will analyze frequency data arranged in the form of (stratified) r c contingency tables.
Michigan State University - EPI - 826
Alka IndurkhyaEPI 8269/15/99Lecture 5 Multiple Logistic Regression Chapter 1, Kleinbaum In simple or multiple linear regression we were relating one or more independent variables ( X1 , X2 , Xk ) to a normally distributed outcome variable Y. In
Michigan State University - PPL - 801
School Culture and Change 1An Examination of School Culture in an Environment of Change John Phillips EAD 801 Michigan State University April 25, 2005School Culture and Change 2 An Examination of School Culture in an Environment of ChangeInitia
Michigan State University - PPL - 802
INTRODUCTION51IntroductionThe quest to understand human learning has, in the past four decades, undergone dramatic change. Once a matter for philosophical argument, the workings of the mind and the brain are now subject to powerful research too
Michigan State University - PPL - 802
SUMMARY1SummaryIn December 1998, the National Research Council released How People Learn, a report that synthesizes research on human learning. The research put forward in the report has important implications for how our society educates: for t
Michigan State University - PPL - 802
How People LearnbridgingResearch and PracticeM. Suzanne Donovan, John D. Bransford, and James W. Pellegrino, editors Committee on Learning Research and Educational Practice Commission on Behavioral and Social Sciences and Education National Resea
Michigan State University - PPL - 802
RESPONSES FROM THE EDUCATION AND POLICY COMMUNITIES253Responses from the Education and Policy CommunitiesThe Committee on Learning Research and Educational Practice invited members of the teacher, administrator, policy, and research communities
Michigan State University - PPL - 803
SCIENCE TE801/TE803Instructor: Louise Kirks E-mail: kirks@msu.edu Course DescriptionStrand: Elementary Science Standard: All teachers will examine the interrelationships among standards, instruction, and assessment to improve student achievement in
Michigan State University - PPL - 806
How I BeganI have had a recent desire to visit Santa Fe, New Mexico. So, I did an internet search to find information on attractions, accommodations, and events that would make my visit fun. Using Ask Jeeves I simply searched Santa Fe, New Mexico.
Michigan State University - PPL - 807
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 2006, p. 57995805 0099-2240/06/\$08.00 0 doi:10.1128/AEM.00109-06 Copyright 2006, American Society for Microbiology. All Rights Reserved.Vol. 72, No. 9Selecting Lactic Acid Bacteria for Their Safety a
Michigan State University - PPL - 807
Opposing Viewpoints Resource Center DocumentPage 1 of 9Cells from hell: toxic algae has killed 400 sea lions off the coast of California and four people in Montreal.Nancy Baron. Saturday Night 115.6 (June 3, 2000): p30-6. Full Text : COPYRIGHT 20
Michigan State University - PPL - 807
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2005, p. 40524056 0099-2240/05/\$08.00 0 doi:10.1128/AEM.71.7.40524056.2005 Copyright 2005, American Society for Microbiology. All Rights Reserved.Vol. 71, No. 7Mediterranean Fruit Fly as a Potential V
Michigan State University - PPL - 809
LABOR AS A QUASI-FIXED FACTOR: Effect of Turnover on Employment RelationshipI. Turnover from 2 perspectives II. Efficiency Wages as solution to turnoverLIR 8092 SIDES TO VOLUNTARY TURNOVERFIRMS VIEWTurnover is a cost Replacement Costs Loss o
Michigan State University - PPL - 809
Introduction to the Labor MarketLIR 809MAJOR QUESTIONS OF LABOR ECONOMICSConcern with PRICING &amp; ALLOCATION of Labor 4 Practical Questions Who works? What determines how much people are paid and in what form? Who can move where &amp; why? Why is ther
Michigan State University - PPL - 809
LIR 809 February 3, 2005 Questions you should be able to answer after doing the reading and attending class 1. What is meant by the term opportunity cost? Looking back at the basic assumptions underlying the market perspective, which one addresses th
Michigan State University - PPL - 809
DEMAND FOR LABOROverview Short-run Demand for Labor Long-run Demand for LaborLIR 809OVERVIEW: Question of interest:How do firms decide how many people to hire and what to pay them?Demand for labor is DerivedPrimary role of firm is to produce
Michigan State University - PPL - 813
Business and Personal LawChad Bobb Business and Information Technology AcademyIntroduction Franklin College Danville High School Pike High School MOUS Certification Michigan State University Ivy Tech State College PersonalThe Course
Michigan State University - PPL - 813
Sampling Scheme:Design:StratifiedRandomSampleandWeightingIn a simple random sample, each member of the population has the same chance of being selected into the sample. In a stratified random sample, each member of the population has a
Michigan State University - PPL - 891
LIR 891: Lecture 10 Impasse Resolution Procedures I. Competing Ends: A. Permit public employees to negotiate their wages, hours and working conditions B. Protect public and government from excessive influence over public policy and the interruptions
Michigan State University - ES - 200
191Switzer/4formulated as a granular was not injurious at the same rate. It would appear that the granular is not absorbed by the grass leaves to the extent that occurs with a wettable powder applied as a foliar spray. A difference in rooting dept
Michigan State University - ES - 200
Michigan State University - PRR - 213
PRR 213 Recreation, Park, and Tourism Links/IdeasJanuary 10-17Recreation, Play, and Leisure http:/roswell-usa.com/city/recreation/leisure29.htmJanuary 22 January 24-29America at Work, School, Leisure - 1894-1915 http:/memory.loc.gov/ammem/awlhtm
Michigan State University - PRR - 213
PRR 475 Evaluation in Parks and RecreationFall 1999 LECT: T-TH 9:10-10:00, Room 155 COM LAB: T 10:20-11:10 Room NR 5, OR 3:00-3:50 Room NR 5 1 hour per week extra in micro-labs Micro-Computer Labs for Selected Dates (TBA) INSTRUCTOR: Daniel J. Styne
Michigan State University - PRR - 215
Park, Recreation &amp; Tourism Resources Department PRR 215-1Professor J.L. BristorRecreation Program ManagementSpring Semester 1999 M &amp; W 12:40-2:30 p.m. 223 Natural Resources (ONE HOUR ARRANGED)Recreation Program ManagementDEPARTMENT OF PARK, R
Michigan State University - PRR - 215
Introduction to Parks, Recreation and Leisure PPR 213 Study Questions EXAMINATION IV Spring Semester 2000Sharpe, Odegard &amp; Sharpe 1. Identify the personal qualifications which are important for effective park management. 2. Discuss the educational r
Michigan State University - PRR - 370
A Training Manual for Americans With Disabilities Act Compliance in Parks and Recreation Settingsby Carol Stensrud Ed.D., C.T.R.S., R.T.R.Venture Publishing, Inc. State College, PADisclaimerThe author and publisher caution the professional to
Michigan State University - PRR - 370