Search openICPSR

Find and share social, behavioral, and health sciences research data.

  • Search terms can be anywhere in the study: title, description, variables, etc.
  • Because our holdings are large, we recommend using at least two query terms:
    rural economy
    home ownership
    higher education
    ghana adolescents
  • Keywords help delimit the breadth of results. Therefore, use as many as required to achieve your desired results:
    elementary education federal funding
  • Our search will find studies with derivative expressions of your query terms: A search for "nation" will find results containing "national"
  • Use quotes to search for an exact expression:
    "social mobility"
  • You can combine exact expressions with loose terms:
    "united states" inmates
  • Exclude results by using a MINUS sign:
    elections -sweden -germany
    elections -sweden -germany
  • On the results page, you will be able to sort and filter to further refine results.
  • Please note that your search queries only openICPSR data holdings.
Name File Type Size Last Modified
  Data 09/30/2019 03:47:PM
  Output 09/30/2019 03:50:PM text/x-stata-syntax 11.8 KB 09/30/2019 11:47:AM text/x-stata-syntax 6.9 KB 09/30/2019 11:50:AM
03-Mathematica-Supplement.nb text/plain 44.6 KB 09/30/2019 11:39:AM
ElwertPfeffer2019.pdf application/pdf 835.7 KB 09/14/2019 06:03:PM
README.txt text/plain 744 bytes 09/30/2019 11:49:AM

Project Citation: 

Elwert, Felix, and Pfeffer, Fabian. The Future Strikes Back. Using Future Treatments to Detect and Reduce Hidden Bias. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2019-09-30.

Project Description

Summary:  View help for Summary
Conventional advice discourages controlling for post-outcome variables in regression analysis. By contrast, we show that controlling for commonly available post-outcome (i.e. future) values of the treatment variable can help detect, reduce, and even remove omitted variable bias (unobserved confounding). The premise is that the same unobserved confounder that affects treatment also affects the future value of the treatment. Future treatments thus proxy for the unmeasured confounder, and researchers can exploit these proxy measures productively. We establish several new results: Regarding a commonly assumed data-generating process involving future treatments, we (1) introduce a simple new approach and show that it strictly reduces bias; (2) elaborate on existing approaches and show that they can increase bias; (3) assess the relative merits of alternative approaches; (4) analyze true state dependence and selection as key challenges. (5) Importantly, we also introduce a new non-parametric test that uses future treatments to detect hidden bias even when future-treatment estimation fails to reduce bias. We illustrate these results empirically with an analysis of the effect of parental income on children’s educational attainment.

Scope of Project

Subject Terms:  View help for Subject Terms PSID
Geographic Coverage:  View help for Geographic Coverage United States
Time Period(s):  View help for Time Period(s) 1968 – 1992

Related Publications

Published Versions

Export Metadata

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.