Sample Data Population Inference A very common paradigm in statistical studies:

Slides:



Advertisements
Similar presentations
1.3 Data Collection and Experimental Design
Advertisements

Where do data come from and Why we don’t (always) trust statisticians.
Three or more categorical variables
Data Collection Jan 28,2014 Math Fall
CONCEPTS UNDERLYING STUDY DESIGN
GATHERING DATA Chapter Experiment or Observe?
Sections – 1.4: Other Effective Sampling Methods – 1.5: Bias in Sampling – 1.6: The Design of Experiments General goals – Collect data effectively – Avoid.
Chapter 4: Designing Studies
Introduction to Data Analysis.
Correlation AND EXPERIMENTAL DESIGN
Literary Digest Poll 1936 election: Franklin Delano Roosevelt vs. Alf Landon Literary Digest had called the election since 1916 Sample size: 2.4 million!
B/W 8/19 Identify the level of measurement of the data listed on the horizontal axis in the graphs: 1)
1. Identify the variable(s) of interest (the focus) and the population of the study. 2. Develop a detailed plan for collecting data. Make sure sample.
Lecture 2 Outline: Thu, Jan 15
The eternal tension in statistics.... Between what you really really want (the population) but can never get to...
Section Decision Making with Data  NOT ALL DATA IS GOOD DATA!  “Do not put faith in what statisticians say until you have carefully considered.
Lecture 2 Outline: Tue, Sep 9 Chapter 1.2: Statistical Inference and Study Design –Types of Inference –Observational Studies vs. Randomized Experiments.
Study Design Data. Types of studies Design of study determines whether: –an inference to the population can be made –causality can be inferred random.
Chapter 51 Experiments, Good and Bad. Chapter 52 Experimentation u An experiment is the process of subjecting experimental units to treatments and observing.
The eternal tension in statistics.... Between what you really really want (the population) but can never get to...
Chapter 4 How to get the Data Part1 n In the first 3 lectures of this course we spoke at length about what care we should take in conducting a study ourselves.
EXPERIMENTS AND OBSERVATIONAL STUDIES Chance Hofmann and Nick Quigley
Research Design Interactive Presentation Interactive Presentation
Chapter 4 Gathering data
C1, L2, S1 Design Method of Data Collection Surveys and Polls Experimentation Observational Studies.
4.2 Statistics Notes What are Good Ways and Bad Ways to Sample?
SAMPLING Nuances of sample size determination Brett Oppegaard, Washington State University Vancouver Language, Texts and Technology, Spring 2011.
Sampling Defined / The idea – Making inference about a larger population What is the population – Some particular value in the population estimating.
C1, L3-4, S1 Design Method of Data Collection Surveys and Polls Experimentation Observational Studies.
RESEARCH METHODS.
Psychological Methods Original Content Copyright by HOLT McDougal. Additions and changes to the original content are the responsibility of the instructor.
{ Chapter 4: Designing Studies 4.3 Using Studies Wisely.
Chapter 12 Designing Good Samples. Doubting the Holocaust? An opinion poll conducted in 1992 for the American Jewish Committee asked: Does it seem possible.
The Scientific Method in Psychology.  Descriptive Studies: naturalistic observations; case studies. Individuals observed in their environment.  Correlational.
DATA COLLECTION METHODS Sampling
Designing Social Inquiry week 4 I36005 Soohyung Ahn Case Study 1936 PRESIDENTIAL ELECTION : Roosevelt VS Landon.
Decision Making with Data Section 8.4. Evaluate data collection procedures Sample size Random assignment Validity –Did the test measure what it was supposed.
Pitfalls of Surveys. The Literary Digest Poll 1936 US Presidential Election Alf Landon (R) vs. Franklin D. Roosevelt (D)
Chapter 12 Sample Surveys *Sample *Bias *Randomizing *Sample Size.
Sampling. Sampling Can’t talk to everybody Select some members of population of interest If sample is “representative” can generalize findings.
Chapter 41 Sample Surveys in the Real World. Chapter 42 Thought Question 1 (from Seeing Through Statistics, 2nd Edition, by Jessica M. Utts, p. 14) Nicotine.
Agresti/Franklin Statistics, 1 of 56 Chapter 4 Gathering data Learn …. How to gather “good” data About Experiments and Observational Studies.
Making Inferences. Sample Size, Sampling Error, and 95% Confidence Intervals Samples: usually necessary (some exceptions) and don’t need to be huge to.
Statistics for fun and profit Chris Williams, Ph.D. Department of Statistics University of Idaho.
5.2 Day 1: Designing Experiments. Period 3 – Seating Chart Front Board AlthisarBarnesCreidlerGreenHollowayMcDonaldOliverRoberts EvansCawthorn e AndersonLavendarJeffreysMcKeelMenaSyed.
1. Identify the variable(s) of interest (the focus) and the population of the study. 2. Develop a detailed plan for collecting data. Make sure sample.
Warm-Up.
STT 421 Day 7: September 28, 2015 September 28, 2015
Psychology I Fri, 3/2  Please get out your textbook, Ch 2, Sec 2 Outline, & Ch 1 Summative Assessment Did you know?? Scientific research does not always.
Political Beliefs and Public Opinion. Political efficacy The belief that one’s political participation really matters.
Principles of exp. design Control for effects of lurking variables Randomization to keep personal biases or other preferences out of the study Replication.
Inference: Probabilities and Distributions Feb , 2012.
Intro Stats Warm – Up 1.3 Determine the Level of Measurement of each (nominal, ordinal, interval, ratio) 1. The daily high temperatures in a city. 2. The.
Chapter Five Vocabulary. Page 1 (1) A Census of the Population This would be ideal – we would actually KNOW the values of the parameters! Really hard.
Section 4.3 Using Studies Wisely By: Michelle Rondilla & Alexander Hasson Period 4.
The Practice of Statistics, 5th Edition Starnes, Tabor, Yates, Moore Bedford Freeman Worth Publishers CHAPTER 4 Designing Studies 4.3 Using Studies Wisely.
4.1 Statistics Notes Should We Experiment or Should We Merely Observe?
Ten percent of U. S. households contain 5 or more people
Section 1.3 Objectives Discuss how to design a statistical study Discuss data collection techniques Discuss how to design an experiment Discuss sampling.
Experiments Textbook 4.2. Observational Study vs. Experiment Observational Studies observes individuals and measures variables of interest, but does not.
THE EFFECT OF SAMPLING BIAS ON BIG DATA BY USING THE READERS DIGEST POLL OF THE 1936 ELECTION AS A CASE STUDY, WE EXAMINE HOW THE SAMPLE OF DATA USED AFFECTS.
LOOKING AT SOME BASICS Can you tell the difference?
Unit 4--Lesson 2. Lesson Objectives At the end of the lesson, students can: Identify common issues with sampling and surveys Design an experiment using.
Experiments vs. Observational Studies vs. Surveys and Simulations
A very common paradigm in statistical studies:
4.3: Using Studies Wisely.
The Survey Method In a survey, people are asked to respond to a series of questions about a particular subject.
CHAPTER 4 Designing Studies
COLLECTING STATISTICAL DATA
STA 291 Fall 2009 Lecture 2 Dustin Lueker.
Presentation transcript:

Sample Data Population Inference A very common paradigm in statistical studies:

Scientific Savvy Article A New York Times article (“Scientific Savvy? In U.S., Not Much”, Aug. 30, 2005) profiled a man who studies what Americans know about science. He uses surveys to assess this knowledge.

What Americans don’t know American adults in general do not understand what molecules are (other than that they are really small). Fewer than a third can identify DNA as a key to heredity. Only about 10 percent know what radiation is. And my personal favorite: One adult American in five thinks the Sun revolves around the Earth. From the article: How is it possible to make claims like this?

A Smoking Paradox Another New York Times article (“Habits: Menthol May Add a Danger for Smokers”, Aug. 30, 2005) opens with a baffling fact: “Black smokers tend to smoke fewer cigarettes than white smokers do, researchers have found. Yet they seem to be at higher risk for smoking-related diseases like lung cancer and heart disease.” Does this mean blacks are simply more susceptible? Not necessarily! Do NOT suppose that association means causation!

A clue to the smoking puzzle? According to one of the studies, led by Carolyn C. Celebucki of the University of Rhode Island, black smokers prefer menthol cigarettes two to one over regular ones, the reverse of the pattern among white smokers. So it might be that menthol is the real culprit. From the article: Often, an association (e.g., between race and smoking- related disease) can be explained by a completely different variable (e.g., menthol content).

Main ideas Often, we wish to know about some population of interest (like adults in the U.S.). But that entire population is unreachable, so we take a REPRESENTATIVE sample to learn about it. Do NOT fall into the trap of trying to infer causation from a statistical study (unless the study is a randomized experiment).

2 types of studies to obtain data relevant to your research: Randomized Experiment Observational Study Experimenter randomly assigns people to treatment or control group Experimenter merely observes things about people in the sample

Randomized Experiment Subjects are randomly assigned to treatment and control (or treatment 1 and treatment 2) Measurements are recorded. Then differences in the data between treatment and control can be said to be CAUSED by the treatment. Example: See articles on yoga study.

Observational Study Subjects are drawn at random from two separate populations. We do not get to randomly assign subjects to the different populations. Measurements are recorded.

Observational Study Continued We then have two separate samples and differences are said to be ASSOCIATED with which population was sampled. We cannot claim that membership in a particular population caused the difference. Example: See articles on positive outlook.

To conduct a statistical study properly, one must get a representative sample. Let’s look at a disastrous example in which this was not done.

Literary Digest 1936: This magazine mailed a questionnaire to over 10 million people VS. vs.

Survey Results: 2.4 million responded! 43% were for Roosevelt Literary Digest predicted a landslide victory for Alf Landon

Election Results:

Magazine Results: Literary Digest went bankrupt shortly thereafter.

FATAL FLAWS: a)Sampling bias: The 10 million people to whom questionnaires were sent were chosen from phone lists, club membership lists, and its own subscription list b)Response bias: The people who responded were not representative; many more republicans than democrats had strong feelings (a)is called using the wrong sampling frame and (b) is called volunteer response.