Elementary Statistics and Inference. Elementary Statistics and Inference. 1.) Introduction. 22S:025 or 7P:025. Lecture 1.

Similar documents
Experimental design. Basic principles

Study Design. Study design. Patrick Breheny. January 23. Patrick Breheny Introduction to Biostatistics (171:161) 1/34

Sampling Controlled experiments Summary. Study design. Patrick Breheny. January 22. Patrick Breheny Introduction to Biostatistics (BIOS 4120) 1/34

Introduction; Study design

Introduction to Statistics Design of Experiments

Study Methodology: Tricks and Traps

What is Statistics? (*) Collection of data Experiments and Observational studies. (*) Summarizing data Descriptive statistics.

Quizzes (and relevant lab exercises): 20% Midterm exams (2): 25% each Final exam: 30%

AMS 5 EXPERIMENTAL DESIGN

Thursday, April 25, 13. Intervention Studies

UCLA STAT 251 / OBEE 216 UCLA STAT 251 / OBEE 216

Thursday, April 25, 13. Intervention Studies

Dr. Allen Back. Sep. 30, 2016

Next, we ll discuss some terminology that is typically used when discussing randomized experiments.

Chapter 13: Experiments

CHAPTER 4 Designing Studies

TEXT: Freedman-Pisani-Purves, Statistics, 3rd edition, W.W. Norton & Co. Bring book to class, WHICH WILL BE HELD IN 344 EVANS, Tu Th

Collecting Data Example: Does aspirin prevent heart attacks?

Math 2a: Lecture 1. Agenda. Course organization. History of the subject. Why is the subject relevant today? First examples

Name Date Period. MULTIPLE CHOICE. Choose the one alternative that best completes the statement or answers the question.

You can t fix by analysis what you bungled by design. Fancy analysis can t fix a poorly designed study.

Chapter 6. Experimental Studies

Lecture 4. Experiments and Observational Studies

Lecture Slides. Elementary Statistics Eleventh Edition. by Mario F. Triola. and the Triola Statistics Series 1.1-1

Gathering. Useful Data. Chapter 3. Copyright 2004 Brooks/Cole, a division of Thomson Learning, Inc.

Applied Analysis of Variance and Experimental Design. Lukas Meier, Seminar für Statistik

What is Statistics? Ronghui (Lily) Xu (UCSD) An Introduction to Statistics 1 / 23

Math 1680 Class Notes. Chapters: 1, 2, 3, 4, 5, 6

Lecture 4: Chapter 3, Section 4 Designing Studies (Focus on Experiments)

Lecture 4: Chapter 3, Section 4. (Focus on Experiments) Designing Studies. Looking Back: Review. Definitions

Chapter 15: Continuation of probability rules

Lecture 2B: Chapter 3, Section 4 Designing Studies (Focus on Experiments)

STATISTICS: METHOD TO GET INSIGHT INTO VARIATION IN A POPULATIONS If every unit in the population had the same value,say

Scientific Method in Vaccine History

Jonas 1/30/10 6:21 AM Page 1 SAMPLE

Lecture 1: Scope, Origins, and Methods in Psychology. Examples of Research in Psychology. Contents

Chapter 6. Experiments in the Real World. Chapter 6 1

REVIEW FOR THE PREVIOUS LECTURE

Trial Designs. Professor Peter Cameron

Lecture 4. Analytical Interventional Studies

Skills Center Psychology Practice Exam II

[Source: Freedman, Pisani, Purves (1998) Statistics, 3rd ed., Ch 2.]

Chapter 8 Statistical Principles of Design. Fall 2010

Sta 309 (Statistics And Probability for Engineers)

Just What the Doctor Ordered

Unit 5. Thinking Statistically

Research Design. Miles Corak. Department of Economics The Graduate Center, City University of New York

Controlled experiments

CHAPTER 1 Understanding Social Behavior

UNIT 4 ALGEBRA II TEMPLATE CREATED BY REGION 1 ESA UNIT 4

Describing Data: Part I. John McGready Johns Hopkins University

Preventing Infectious Diseases. Chapter 28 Lesson 3

Section 4.3 Using Studies Wisely. Honors Statistics. Aug 23-8:26 PM. Daily Agenda. 1. Check homework C4# Group Quiz on

Ethical Issues for Biostatisticians. The Remune Story

Sampling Reminders about content and communications:

The Nature of Science

Daily Agenda. Honors Statistics. 1. Check homework C4#9. 4. Discuss 4.3 concepts. Finish 4.2 concepts. March 28, 2017

Observational Studies and Experiments. Observational Studies

Scientific Method in Biology

Controlled Experiments

Review. Chapter 5. Common Language. Ch 3: samples. Ch 4: real world sample surveys. Experiments, Good and Bad

El Camino College. Course Syllabus Spring 2016

Introduction to Social Psychology

In this chapter we discuss validity issues for quantitative research and for qualitative research.

Homework Answers. 1.3 Data Collection and Experimental Design

Stat 13, Intro. to Statistical Methods for the Life and Health Sciences.

Questions for the Table Group

El Camino College. Course Syllabus Winter 2017

CHAPTER 9: Producing Data: Experiments

BIOSTATS 540 Fall 2018 Course Introduction Page 1 of 20. Course Introduction

TESTING TREATMENTS SYNTHESIZING INFORMATION FROM RESEARCH

AVOIDING BIAS AND RANDOM ERROR IN DATA ANALYSIS

Chapter 2 Doing Sociology: Research Methods

Culminating Assessments. Option A Individual Essay. Option B Group Research Proposal Presentation

For more information about how to cite these materials visit

Internal & External Validity

Assessing and Enhancing HIV Vaccine Trial (HVT) Consent Preparedness Among Street Drug Users

Math 140 Introductory Statistics

Statistics Success Stories and Cautionary Tales

SYLLABUS. Texas A&M Commerce. Abnormal Psychology & Developmental Psychopathology PSY Fall 2015

BIOE 301. Lecture Seventeen

Elements of the Structured Abstract

Psych 1Chapter 2 Overview

CHAPTER 6. Experiments in the Real World

Randomized Clinical Trials

Activity 19 Great Ideas in the History of Psychology. Purpose

Getting the Big Picture

Scientific Research. The Scientific Method. Scientific Explanation

Lesson Plan. Class Policies. Designed Experiments. Observational Studies. Weighted Averages

Chapter 23. Inference About Means. Copyright 2010 Pearson Education, Inc.

Why do Psychologists Perform Research?

STATISTICS 201. Survey: Provide this Info. How familiar are you with these? Survey, continued IMPORTANT NOTE. Regression and ANOVA 9/29/2013

Patrick Breheny. January 28

Chapter 13. Experiments and Observational Studies. Copyright 2012, 2008, 2005 Pearson Education, Inc.

AP Psychology Ch. 01 Psych Science & Stats Study Guide

Evidence Based Practice

aps/stone U0 d14 review d2 teacher notes 9/14/17 obj: review Opener: I have- who has

General Biostatistics Concepts

PUBLIC ATTITUDES, BELIEFS AND KNOWLEDGE ABOUT SMALLPOX

Introduction. Marius Ionescu 08/30/2011

Transcription:

Elementary Statistics and Inference 22S:025 or 7P:025 Lecture 1 1 Elementary Statistics and Inference 22S:025 or 7P:025 Introduction 2 1.) Introduction A. The course will be based on the text: Statistics, Fourth Edition, Freedman, Pisani, and Purves published by Norton & Company (2007) IMU Bookstore B. The syllabus for the course is provided as a separate handout to each student 3 1

1.) Introduction (cont.) C. This course provides an overview of the logic and techniques involved in analyzing and interpreting empirical data. Methods for displaying data graphically and for describing characteristics of data will be covered in the first part of the course dealing with descriptive statistics. The second part of the course includes topics in elementary probability theory and sampling, forming a basis for techniques of statistical inference. The problem of inference, attempting to make generalizations that go beyond the data at hand, is the focus of the third part of the course. 4 2

Elementary Statistics and Inference 22S:025 or 7P:025 Chapter One 10 2.) Chapter One A. Salk Vaccine Field Trial NFIP Years ago Polio was a dreaded and often crippling disease that attacked many young people often in the summer. Jonas Salk, in the 1950s, developed a promising drug that would protect an individual from contracting polio. His drug was tested by the National Public Health Service and the National Foundation for Infantile Paralysis (NFIP) in 1954. 11 The drug was tested with a treatment group, against a control group, using a double blind experiment. Neither the research subjects (persons) nor the doctors who measured the responses knew who was in the treatment group (those who received the drug), or the control group (those who did not receive the drug). The Experiment 2 x 10 6 children, treatment group were vaccinated, control group not vaccinated experiment run by National Foundation for Infantile Paralysis (NFIP) grades 1, 2, 3. 12 3

Refused to Participate Agreed to Participate Treatment 500,000 Control 1,000,000 500,000 1,500,000 2,000,000=Total 13 Parents who were from higher SES would be more willing to have child participate than those from lower SES however, lower income families probably had children exposed to polio more often than children from higher income high risk children have more antibodies. The design was flawed because it included more children from higher income families, and they were more likely to contract polio. The treatment and control groups should be as nearly alike as possible. 14 B. Outcome of the Salk Vaccine Trials Likelihood of being assigned to treatment group was same as for control group. Doctors not told which children were in treatment or control group. 15 4

Randomized Double Blind Experiment N Rate Treatment 200,000 28 Control 200,000 71 No Consent 350,000 46 (See page 6.) 16 National Federation of Infantile Paralysis (NFIP) Experiment N Rate Grade 2 225,000 25 Grade 1 and 3 (Control) 725,000 54 Grade 2 (No Consent) 125,000 44 Grade 2 students received the vaccine. (see page 6.) 17 NFIP biased because students it was given only to children who were volunteered by their parents and in grade 2. (Reduced from 54% to 25%) Double Blind less biased because given across grades, and randomly given to those who were volunteered. Some volunteered children (Control) did not receive vaccine. (Reduced from 71% to 28%) Bottom Line Salk Vaccine was found to be very good. Polio is now extremely rare, compared to 1950. 18 5

C. Portacaval Shunt Chapter One In some cases of cirrhosis of the liver, the patient may start to hemorrhage and die. The portacaval shunt is a risky surgical procedure that redirects the blood flow to reduce damage from hemorrhage. 19 Table 2: 51 Studies on Portacaval Shunt Degree of Enthusiasm % Marked Design Marked Moderate None Enthusiasm No Controls 24 7 1 24/32 Controls, Not Randomized 10 3 2 10/15 Randomized Controls 0 1 3 0/3 Poor study design can lead to misleading results. 20 D. Possible Research Designs Experiments Randomized Controls Patient Population Controls Not Randomized Patient Population Eligible Ineligible (two side, wrong disease, etc.) Healthier Sicker Surgery Control Surgery Control Best Design 21 6

Table 3: Three-Year Survival Rates in Studies of Portacaval Shunt Randomized Not Randomized Surgery 60% 60% Controls 60% 45% Two designs lead to different conclusions when randomized, surgery not more effective; when not randomized, the surgery seems better. 22 E. Historical Controls When it is awkward, repugnant, or difficult to develop randomized control group of subjects the treatment group might be compared against the historical group with the disease / or skill level. Comparisons with the historical control group are not as defensible as using a contemporary control group for comparing effectiveness of treatment. 23 Table 4: A Study of Coronary Therapy Randomized Controlled Historically Controlled Therapy + - + - Coronary Bypass Surgery 1 7 16 5 5-FU 0 5 2 0 BCG 2 2 4 0 DES 0 3 5 0 Randomized controls 17 of 20 studies treatments not effective. Historical controlled 27 of 32 studies treatments effective. 24 7

Table 5: 3 Year Survival Rates Group Randomized Historical Surgery 87.6% 90.9% Controls 82.2% 71.1% Using Historical controls misleading, because only more healthy patients received treatment. 25 F. See Summary at End of Chapter One pages 10-11 Terms to note: method of comparison, treatment group, control group, confounding, randomized controlled experiment, placebo, and double blind. 26 8