Discussion Alan Zaslavsky Harvard Medical School.

Slides:



Advertisements
Similar presentations
Evaluation Overview - Basics. Purpose of Testing Diagnostic Formative Summative.
Advertisements

Foresight, Insight. Hindsight US Statistical Observations Fritz Scheuren NORC University of Chicago.
Handling attrition and non- response in longitudinal data Harvey Goldstein University of Bristol.
Multilevel Interventions: Study Design and Analysis Issues Paul D. Cleary, Ph.D. Yale School of Public Health Cary Gross, M.D. Yale School of Medicine.
Module B-4: Processing ICT survey data TRAINING COURSE ON THE PRODUCTION OF STATISTICS ON THE INFORMATION ECONOMY Module B-4 Processing ICT Survey data.
1 Incorporating Statistical Process Control and Statistical Quality Control Techniques into a Quality Assurance Program Robyn Sirkis U.S. Census Bureau.
Economics 105: Statistics GH 24 due Wednesday. Hypothesis Tests on Several Regression Coefficients Consider the model (expanding on GH 22) Is “race” as.
Section 1.3 Experimental Design © 2012 Pearson Education, Inc. All rights reserved. 1 of 61.
Section 1.3 Experimental Design.
Data Collection Six Sigma Foundations Continuous Improvement Training Six Sigma Foundations Continuous Improvement Training Six Sigma Simplicity.
March 2013 ESSnet DWH - Workshop IV DATA LINKING ASPECTS OF COMBINING DATA INCLUDING OPTIONS FOR VARIOUS HIERARCHIES (S-DWH CONTEXT)
© John M. Abowd 2005, all rights reserved Analyzing Frames and Samples with Missing Data John M. Abowd March 2005.
Documentation and survey quality. Introduction.
Statistical Decision-Making Process: Decision Problem Statistical Plan Data Description and Summary Analysis and Conclusions Testing Validity of Results.
SOWK 6003 Social Work Research Week 5 Measurement By Dr. Paul Wong.
Chapter 11 Quality Control.
FINAL REPORT: OUTLINE & OVERVIEW OF SURVEY ERRORS
Chapter 14 Inferential Data Analysis
Go to Table of ContentTable of Content Analysis of Variance: Randomized Blocks Farrokh Alemi Ph.D. Kashif Haqqi M.D.
MGT-491 QUANTITATIVE ANALYSIS AND RESEARCH FOR MANAGEMENT OSMAN BIN SAIF Session 14.
Arun Srivastava. Types of Non-sampling Errors Specification errors, Coverage errors, Measurement or response errors, Non-response errors and Processing.
Chapter 24: Comparing Means.
Eurostat Statistical Data Editing and Imputation.
5-1 Introduction 5-2 Inference on the Means of Two Populations, Variances Known Assumptions.
Aggregate and Systemic Components of Risk in Total Survey Error Models John L. Eltinge U.S. Bureau of Labor Statistics International Total Survey Error.
Research methodology Data Collection tools and Techniques.
Notes Science Tennessee SPI Objective:
Section 1.3 Experimental Design Larson/Farber 4th ed.
Up the Odds when Gambling with Leads Modeling and Scoring Leads for Marketing in Education Presented at the DETC 2009 Workshop, Naples, FL October 20,
Experimental Design 1 Section 1.3. Section 1.3 Objectives 2 Discuss how to design a statistical study Discuss data collection techniques Discuss how to.
List frames area frames and administrative data, are they complementary or in competition? Elisabetta Carfagna University of Bologna Department of Statistics.
Burden and Loss: The Role of Panel Survey Recordkeeping in Self-report Quality and Nonresponse ITSEW 2010 Ryan Hubbard and Brad Edwards.
Curb-stoning, a Too Neglected and Very Embarrassing Survey Problem Comments Jaki S. McCarthy Senior Cognitive Research Methodologist US Department of Agriculture.
Topic (ii): New and Emerging Methods Maria Garcia (USA) Jeroen Pannekoek (Netherlands) UNECE Work Session on Statistical Data Editing Paris, France,
The relationship between error rates and parameter estimation in the probabilistic record linkage context Tiziana Tuoto, Nicoletta Cibella, Marco Fortini.
1 Improving Data Quality. COURSE DESCRIPTION Introduction to Data Quality- Course Outline.
Stop the Madness: Use Quality Targets Laurie Reedman.
The Conditional Independence Assumption in Probabilistic Record Linkage Methods Stephen Sharp National Records of Scotland Ladywell Road Edinburgh EH12.
MGT-491 QUANTITATIVE ANALYSIS AND RESEARCH FOR MANAGEMENT OSMAN BIN SAIF Session 16.
Research & Experimental Design Why do we do research History of wildlife research Descriptive v. experimental research Scientific Method Research considerations.
Chapter 9 Audit Sampling: An Application to Substantive Tests of Account Balances McGraw-Hill/IrwinCopyright © 2012 by The McGraw-Hill Companies, Inc.
Copyright 2010, The World Bank Group. All Rights Reserved. Reducing Non-Response Section B 1.
PROCESSING OF DATA The collected data in research is processed and analyzed to come to some conclusions or to verify the hypothesis made. Processing of.
Ensuring rigour in qualitative research CPWF Training Workshop, November 2010.
Apr. 22 Stat 100. Final Wednesday April 24 About 40 or so multiple choice questions Comprehensive Study the midterms Copies and answers are at the course.
1 C. ARRIBAS, D. LORCA, A. SALINERO & A. COLMENERO Measuring statistical quality at the Spanish National Statistical Institute.
© John M. Abowd 2007, all rights reserved General Methods for Missing Data John M. Abowd March 2007.
Marketing Research Approaches. Research Approaches Observational Research Ethnographic Research Survey Research Experimental Research.
Disclosure Limitation in Microdata with Multiple Imputation Jerry Reiter Institute of Statistics and Decision Sciences Duke University.
Personnel record and report is a statement describing an event, situation and happening in a clear manner. It provide both qualitative and quantitative.
1 Quality Control for Field Operations. 2 Overview Goal To ensure the quality of survey field work Purpose To detect and deter interviewer errors and.
11 How Much of Interviewer Variance is Really Nonresponse Error Variance? Brady T. West Michigan Program in Survey Methodology University of Michigan-Ann.
Use of Tablet Computers to Implement the Local Governance Performance Index (LGPI) in Tunisia Lindsay J. Benstead - Assistant Professor of Political Science,
The hidden side of successful story – implication of wide use of administrative data sources at national statistical institutes Metka Zaletel, Irena Križman.
Section 1.3 Experimental Design.
Stages of Test Development By Lily Novita
Machine Learning in Practice Lecture 10 Carolyn Penstein Rosé Language Technologies Institute/ Human-Computer Interaction Institute.
Quantitative and Qualitative Research Approaches
An Overview of Editing and Imputation Methods for the next Italian Censuses Gianpiero Bianchi, Antonia Manzari, Alessandra Reale UNECE-Eurostat Meeting.
Synthetic Approaches to Data Linkage Mark Elliot, University of Manchester Jerry Reiter Duke University Cathie Marsh Centre.
1 General Recommendations of the DIME Task Force on Accuracy WG on HBS, Luxembourg, 13 May 2011.
Section 1.3 Objectives Discuss how to design a statistical study Discuss data collection techniques Discuss how to design an experiment Discuss sampling.
Chapter 29 Conducting Market Research. Objectives  Explain the steps in designing and conducting market research  Compare primary and secondary data.
1.3 Experimental Design. What is the goal of every statistical Study?  Collect data  Use data to make a decision If the process to collect data is flawed,
Collecting Sample Data Chapter 1 Section 4 Part 2.
Prologue.
Understanding Results
Using Measurement Scales to Build Marketing Effectiveness
Chapter 25: Paired Samples and Blocks
Check Cruising Social : Science by Kim Iles : SITCA
Presentation transcript:

Discussion Alan Zaslavsky Harvard Medical School

Fabrication as a Statistical Procedure Fabrication is like imputation – Duplication is like hot deck – Duplication with random modifications is like multiple imputation – Duplication is like weight modification Fabrication is a multilevel process – Interview, interviewer, area, … project level

Fabrication as a Game Payoffs/risks to fabricator – Reduce effort while receiving payment – Risks greater for higher-level organization/person Detection/deterrence Costs/risks to data purchaser – Paying more for less information – Wrong decisions – Loss of credibility (cliff loss function) Risks may change with greater expertise on either side

Assumptions about Fabricators Fabricators are not very sophisticated – No fancy synthesis models Fabricators are not trying to work hard – Falsifying must be easier than data collection – Will not know how to “beat” moderately sophisticated detection techniques If fabricators try harder … – Good standard synthesis methods could be hard to detect – Learning on both sides

Fabrication on the Continuum of Survey Management Related to other survey errors at scale – Inadequately designed survey questions and tools Not adapted to conditions under which survey fielded – Interviewer errors Misinterpretation of questions, procedures Interpersonal interview technique Training and motivation Monitoring of “honesty”, accuracy, technique

Detection techniques Good survey management – Timely, at all levels – Recruitment, observation – Metadata and paradata Post-survey analysis – Replication of survey: interpenetrating samples – Subject-matter expertise – Statistical outliers (single and patterns) Earlier is better

Regina Faranda Extensive checking – Subject-matter and survey expertise – Checklist: QC Statistical assumptions? – Can be stated and tested

Rita Thissen Detailed specifics of monitoring and detection systems – Technology: CARI, CAPI, … (Anecdotes rarely heard)

Mike Robbins Duplicate detection is like record linkage – Likelihood ratio Duplicate detection also important in other settings – US Census (2000?): match 330M

Robbins – Duplicate detection Duplicate detection is like record linkage – Likelihood ratio Duplicate detection also important in other settings – US Census (2000?): match 330M × 330M possible record pairs Would models be different for fabricated data, processing errors, repeated real interviews?

Example: Medicare CAHPS survey Pulled ~5000 responses (out of ~400K/year) Examined 27 substantive items Complex features – Substantial amount of screening/skipped items – Multiple choice items – Blocks of closely related items

Agreement – all pairs

Best agreement: duplicates?

Conclusions Know your data and survey methodology Thanks to speakers for sharing their experience and methods