Stephanie Eckman
Stephanie Eckman
Home
AI Research
Survey Research
Talks
Publications
Contact
CV
Light
Dark
Automatic
measurement error
Data Quality in Data Science
Data scientists are increasingly turning their attention to the collection of high quality training data. Those of us with expertise in data collection can apply lessons in web survey design to them.
Aug 5, 2021 10:00 AM
Stephanie Eckman
Project
Slides
Video
stephnie
Annotation Sensitivity: Training Data Collection Methods Affect Model Performance
Small changes in how you ask annotators to label data can dramatically change your model’s behavior. We tested 5 versions of a hate speech labeling task and found significant differences in model performance.
Christoph Kern
,
Stephanie Eckman
,
Jacob Beck
,
Rob Chew
,
Bolei Ma
,
Frauke Kreuter
PDF
Project
DOI
Underreporting of Purchases in the U.S. Consumer Expenditure Survey
Motivated misreporting occurs when respondents give incorrect responses to survey questions to shorten the interview; studies have …
Stephanie Eckman
PDF
Project
DOI
Panel Conditioning in the U.S. Consumer Expenditure Survey
Do survey respondents learn to game the questionnaire over time? We tested this in the U.S. Consumer Expenditure Survey and found the opposite. Responses actually improve in later waves.
Stephanie Eckman
,
Ruben Bach
PDF
Project
DOI
Understanding the Effects of Questionnaire Changes when Experiments are Not Possible: Application to the U.S. Consumer Expenditure Survey
Many large-scale government surveys do not have the budget or time to conduct experiments to assess the effects of questionnaire …
Mar 2, 2021 12:30 PM
virtual
Stephanie Eckman
Project
stephnie
Motivated Misreporting in the Consumer Expenditure Household Survey
Motivated misreporting occurs when respondents give incorrect responses to questions to shorten a survey interview. Previous research …
Jun 12, 2020 8:30 AM
Stephanie Eckman
Project
Slides
stephnie
Blending Survey and Passive Data
As survey costs increase and response rates decrease, researchers are looking to alternative methods to collect data from study …
Feb 28, 2020 12:00 PM
United Nations, New York City, NY
Stephanie Eckman
Project
Slides
Improving Labeling Through Social Science Insights: Preliminary Results and Research Agenda
How you design a labeling interface affects the labels you get. We show that task structure, ordering, and annotator backgrounds all shape training data quality.
Jacob Beck
,
Stephanie Eckman
,
Rob Chew
,
Frauke Kreuter
PDF
Project
DOI
Motivated Misreporting in Smartphone Surveys
Filter questions are used to administer follow-up questions to eligible respondents while allowing respondents who are not eligible to …
Jessica Daikeler
,
Ruben Bach
,
Henning Silber
,
Stephanie Eckman
PDF
Project
DOI
Rotation Group Bias in Reporting of Household Purchases in the U.S. Consumer Expenditure Survey
We find rotation group bias in reports of spending in the U.S. Consumer Expenditure Survey. Contrary to our expectations, the more …
Ruben Bach
,
Stephanie Eckman
PDF
Project
DOI
«
»
Cite
×