Causation and the Rules of Inference Classes 4 and 5.

Slides:



Advertisements
Similar presentations
How do we know when we know. Outline  What is Research  Measurement  Method Types  Statistical Reasoning  Issues in Human Factors.
Advertisements

Sample size estimation
Deriving Biological Inferences From Epidemiologic Studies.
Causality Causality Hill’s Criteria Cross sectional studies.
Traps and pitfalls in medical statistics Arvid Sjölander.
1 Case-Control Study Design Two groups are selected, one of people with the disease (cases), and the other of people with the same general characteristics.
COURSE: JUST 3900 INTRODUCTORY STATISTICS FOR CRIMINAL JUSTICE Instructor: Dr. John J. Kerbs, Associate Professor Joint Ph.D. in Social Work and Sociology.
FUNDAMENTAL RESEARCH ISSUES © 2012 The McGraw-Hill Companies, Inc.
Correlation Chapter 9.
Introduction to Risk Factors & Measures of Effect Meg McCarron, CDC.
Fundamentals of Forensic DNA Typing Slides prepared by John M. Butler June 2009 Appendix 3 Probability and Statistics.
Epidemiology Kept Simple
Evaluating Hypotheses Chapter 9. Descriptive vs. Inferential Statistics n Descriptive l quantitative descriptions of characteristics.
Evaluating Hypotheses Chapter 9 Homework: 1-9. Descriptive vs. Inferential Statistics n Descriptive l quantitative descriptions of characteristics ~
Introduction to Hypothesis Testing CJ 526 Statistical Analysis in Criminal Justice.
Introduction to Hypothesis Testing CJ 526 Statistical Analysis in Criminal Justice.
Using Statistics in Research Psych 231: Research Methods in Psychology.
Sample Size and Statistical Power Epidemiology 655 Winter 1999 Jennifer Beebe.
Descriptive Statistics
Research Methods AP Psych – Chapter 2 Psychology’s Scientific Method
Thomas Songer, PhD with acknowledgment to several slides provided by M Rahbar and Moataza Mahmoud Abdel Wahab Introduction to Research Methods In the Internet.
Research Methods in Crime and Justice Chapter 5 Causality.
© 2011 The McGraw-Hill Companies, Inc. Chapter 2 Psychology’s Scientific Method.
Overview Definition Hypothesis
POSC 202A: Lecture 1 Introductions Syllabus R Homework #1: Get R installed on your laptop; read chapters 1-2 in Daalgard, 1 in Zuur, See syllabus for Moore.
Chapter 8 Introduction to Hypothesis Testing
Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins Chapter 7: Gathering Evidence for Practice.
Chapter 3 The Research Design. Research Design A research design is a plan of action for executing a research project, specifying The theory to be tested.
Epidemiology The Basics Only… Adapted with permission from a class presentation developed by Dr. Charles Lynch – University of Iowa, Iowa City.
Web of Causation; Exposure and Disease Outcomes Thomas Songer, PhD Basic Epidemiology South Asian Cardiovascular Research Methodology Workshop.
The Argument for Using Statistics Weighing the Evidence Statistical Inference: An Overview Applying Statistical Inference: An Example Going Beyond Testing.
Chapter 8 Introduction to Hypothesis Testing
Measures of Association
Presentation on Type I and Type II Errors How can someone be arrested if they really are presumed innocent? Why do some individuals who really are guilty.
© 2011 The McGraw-Hill Companies, Inc. with snazzy editions by Mrs. Short Chapter 2 Psychology’s Scientific Method.
Slides to accompany Weathington, Cunningham & Pittenger (2010), Chapter 3: The Foundations of Research 1.
Research Process Parts of the research study Parts of the research study Aim: purpose of the study Aim: purpose of the study Target population: group whose.
Introduction to Biostatistics, Harvard Extension School © Scott Evans, Ph.D.1 Statistical Inference: Hypothesis Testing.
Introductory Topics PSY Scientific Method.
STUDYING BEHAVIOR © 2009 The McGraw-Hill Companies, Inc.
Chapter 2 Nature of the evidence. Chapter overview Introduction What is epidemiology? Measuring physical activity and fitness in population studies Laboratory-based.
Inferential Statistics Body of statistical computations relevant to making inferences from findings based on sample observations to some larger population.
IT 와 인간의 만남 KAIST 지식서비스공학과 Experimental Research KSE966/986 Seminar Uichin Lee Sept. 21, 2012.
Causation.
Research Methods in Psychology Chapter 2. The Research ProcessPsychological MeasurementEthical Issues in Human and Animal ResearchBecoming a Critical.
Reading Health Research Critically The first four guides for reading a clinical journal apply to any article, consider: the title the author the summary.
Causation and the Rules of Inference Classes 4 and 5.
 Descriptive Methods ◦ Observation ◦ Survey Research  Experimental Methods ◦ Independent Groups Designs ◦ Repeated Measures Designs ◦ Complex Designs.
METHODS IN BEHAVIORAL RESEARCH NINTH EDITION PAUL C. COZBY Copyright © 2007 The McGraw-Hill Companies, Inc.
©2010 John Wiley and Sons Chapter 2 Research Methods in Human-Computer Interaction Chapter 2- Experimental Research.
L6172 Law and Social Science R eview April 23, 2007.
Introduction to Biostatistics, Harvard Extension School © Scott Evans, Ph.D.1 Hypothesis Testing And the Central Limit Theorem.
Graduate School for Social Research Autumn 2015 Research Methodology and Methods of Social Inquiry socialinquiry.wordpress.com Causality.
Design of Clinical Research Studies ASAP Session by: Robert McCarter, ScD Dir. Biostatistics and Informatics, CNMC
Showing Cause, Introduction to Study Design Principles of Epidemiology.
MSE 600 Descriptive Statistics Chapter 11 & 12 in 6 th Edition (may be another chapter in 7 th edition)
Hypothesis Testing and Statistical Significance
Measures of disease frequency Simon Thornley. Measures of Effect and Disease Frequency Aims – To define and describe the uses of common epidemiological.
Methods of Presenting and Interpreting Information Class 9.
Sample size calculation
© 2010 Jones and Bartlett Publishers, LLC
Significance testing Introduction to Intervention Epidemiology
Causation Analysis in Occupational and Environmental Medicine
Showing Cause, Introduction to Study Design
© 2011 The McGraw-Hill Companies, Inc.
Introduction to Inductive Statistics
Critical Appraisal วิจารณญาณ
Interpreting Epidemiologic Results.
Relationship Relation: Association: real and spurious Statistical:
Research Techniques Made Simple: Interpreting Measures of Association in Clinical Research Michelle Roberts PhD,1,2 Sepideh Ashrafzadeh,1,2 Maryam Asgari.
Presentation transcript:

Causation and the Rules of Inference Classes 4 and 5

Causal Reasoning  Elements of causation in traditional positivist frameworks (Hume, Mill, et al.) Correlation Correlation Temporal Precedence Temporal Precedence Constant Conjunction (Hume) Constant Conjunction (Hume) Absence of spurious effects Absence of spurious effects  Valid causal stories have utilitarian value Causal mechanisms are reliable when they can support predictions and control, as well as explanations Causal mechanisms are reliable when they can support predictions and control, as well as explanations  We distinguish causal description from causal explanation We don’t need to know the precise causal mechanisms to make a “causal claim We don’t need to know the precise causal mechanisms to make a “causal claim Instead, we can observe the relationship between a variable and an observable outcome to conform to the conceptual demands of “causation” Instead, we can observe the relationship between a variable and an observable outcome to conform to the conceptual demands of “causation”

 “Essentialist” Concerns Cause present-cause absent demand Cause present-cause absent demand Threshold effects Threshold effects Indirect causation Indirect causation Distal versus proximal causes temporallyDistal versus proximal causes temporally Leveraged causationLeveraged causation Temporal delay Temporal delay Multiple causation versus spurious causation Multiple causation versus spurious causation  Experimental versus Epidemiological Causation Experiments test specific hypotheses through manipulation and control of experimental conditions Experiments test specific hypotheses through manipulation and control of experimental conditions Epidemiological studies presumes a probabilistic view of causation based on naturally occurring observations Epidemiological studies presumes a probabilistic view of causation based on naturally occurring observations “A’s blow was followed by B’s death” versus “A’s blow caused B’s death” “A’s blow was followed by B’s death” versus “A’s blow caused B’s death”

Criteria for Causal Inference  Strength (is the risk so large that we can easily rule out other factors)  Consistency (have the results have been replicated by different researchers and under different conditions)  Specificity (is the exposure associated with a very specific disease as opposed to a wide range of diseases)  Temporality (did the exposure precede the disease)  Biological gradient (are increasing exposures associated with increasing risks of disease)  Plausibility (is there a credible scientific mechanism that can explain the association)  Coherence (is the association consistent with the natural history of the disease)  Experimental evidence (does a physical intervention show results consistent with the association)  Analogy (is there a similar result to which we can draw a relationship) Source: Sir Austin Bradford Hill, The Environment and Disease: Association or Causation, 58 Proc. R. Soc. Med. 295 (1965)

Errors in Causal Inference  Two Types of Error Type I Error (α) – a false positive, or the probability of falsely rejecting the null hypothesis of no relationship Type I Error (α) – a false positive, or the probability of falsely rejecting the null hypothesis of no relationship Type II Error (β) – a false negative, or the probability of falsely accepting the null hypothesis of no relationship Type II Error (β) – a false negative, or the probability of falsely accepting the null hypothesis of no relationship The two types of error are related in study design, and one makes a tradeoff in the error bias in a study The two types of error are related in study design, and one makes a tradeoff in the error bias in a study Statistical Power = 1 – β -- probability of correctly rejecting the null hypothesis Statistical Power = 1 – β -- probability of correctly rejecting the null hypothesis

Justice System - Trial Defendant InnocentDefendant Guilty Reject Presumption of Innocence (Guilty Verdict) Type I ErrorCorrect Fail to Reject Presumption of Innocence (Not Guilty Verdict) CorrectType II Error Statistics - Hypothesis Test Null Hypoth TrueNull Hypoth False Reject Null Hypothesis Type I ErrorCorrect Fail to Reject Null Hypothesis CorrectType II Error

Interpreting Causal Claims  In Landrigan, the Court observes that many studies conflate the magnitude of the effect with statistical significance: Can still observe a weak effect that is statistically significant (didn’t happen by chance) Can still observe a weak effect that is statistically significant (didn’t happen by chance) Can observe varying causal effects at different levels of exposure, causal effect is not indexed Can observe varying causal effects at different levels of exposure, causal effect is not indexed

 Alternatives to Statistical Significance Odds Ratio – the odds of having been exposed given the presence of a disease (ratio) compared to the odds of not having been exposed given the presence of the disease (ratio) Odds Ratio – the odds of having been exposed given the presence of a disease (ratio) compared to the odds of not having been exposed given the presence of the disease (ratio) Risk Ratio – the risk of a disease in the population given exposure (ratio) compared to the risk of a disease given no exposure (ratio, or the base rate) Risk Ratio – the risk of a disease in the population given exposure (ratio) compared to the risk of a disease given no exposure (ratio, or the base rate) Attributable Risk – Attributable Risk – (Rate of disease among the unexposed – Rate of disease among the exposed) (Rate of disease among the exposed)  Effect Size versus Significance Such indicia help mediate between statistical significance and effect size, which are two different ways to think about causal inference Such indicia help mediate between statistical significance and effect size, which are two different ways to think about causal inference Can there be causation without significance? Yes Can there be causation without significance? Yes Allen v U.S. (588 F. Supp. 247 (1984)Allen v U.S. (588 F. Supp. 247 (1984) In re TMI, 922 F. Supp. 997 (1996)In re TMI, 922 F. Supp. 997 (1996)

 Thresholds Asbestos Litigation – relative risk must exceed 1.5, while others claim 2.0 relative risk and 1.5 attributable risk Asbestos Litigation – relative risk must exceed 1.5, while others claim 2.0 relative risk and 1.5 attributable risk RR=1.24 was “significant” but “…far removed from proving ‘specific’ causation” (Allison v McGhan, 184 F 3d 1300 (1999))RR=1.24 was “significant” but “…far removed from proving ‘specific’ causation” (Allison v McGhan, 184 F 3d 1300 (1999)) Probability standard seems to be at 50% causation, or a risk ratio of 2.0 (“ a two-fold increase” – Marder v GD Searle, 630 F. Supp (1986)). Probability standard seems to be at 50% causation, or a risk ratio of 2.0 (“ a two-fold increase” – Marder v GD Searle, 630 F. Supp (1986)). Landrigan – 2.0 is a “piece of evidence”, not a “password” to a finding of causation Landrigan – 2.0 is a “piece of evidence”, not a “password” to a finding of causation But exclusion of evidence at a RR=1.0 risks a Type II errorBut exclusion of evidence at a RR=1.0 risks a Type II error

 Epstein and King “We thus recommend that researchers not change the object of their inferences because causal inference is difficult. Instead, they should make their questions as precise as possible, follow the best advice science has to offer about reducing uncertainty and bias, and communicate the appropriate level of uncertainty readers should have in interpreting their results….” ( 38 ) “We thus recommend that researchers not change the object of their inferences because causal inference is difficult. Instead, they should make their questions as precise as possible, follow the best advice science has to offer about reducing uncertainty and bias, and communicate the appropriate level of uncertainty readers should have in interpreting their results….” ( 38 )

Epstein and King – Foundational Requirements  Replicability – transparency of theory, data and method  Social Good Peer Review Peer Review Research data should be in the public domain via data archiving Research data should be in the public domain via data archiving  Theory – should lead to observables  Control for Rival Hypotheses and “Third Factors”  Pay Attention to Measurement Validity and Reliability Validity and Reliability  Relevance of Samples, Size of Samples, Randomness of Samples, Avoid Selection Bias in Samples  Statistical Inferences and Estimation – use triangulation through multiple methods

Case Study  Pierre v Homes Trading Company  Lead paint exposure in childhood produced behavioral and social complications over the life course, resulting in criminal activity and depressed earnings as an adult  Evidence – epidemiological study of birth cohort exposed to lead paint in childhood and their future criminality and life outcomes