Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

If You Use a Screen Reader

This content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.

Risky Inference: Unobserved Treatment Effects in Conflict Studies

David H. Clark and Timothy Nordstrom
International Studies Quarterly
Vol. 47, No. 3 (Sep., 2003), pp. 417-429
Published by: Wiley on behalf of The International Studies Association
Stable URL: http://www.jstor.org/stable/3693593
Page Count: 13
  • Read Online (Free)
  • Download ($42.00)
  • Subscribe ($19.50)
  • Cite this Item
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Risky Inference: Unobserved Treatment Effects in Conflict Studies
Preview not available

Abstract

This article illustrates the importance of testing empirical models in samples appropriate to the theories the models are intended to test. While social science appears to mandate that we prefer general theories to limited ones, the generality of a theory rests in its logical application to a set of observations, not solely to its statistical survival in a large data set. Theories in international relations, especially those linking domestic turmoil and international conflict, are advancing, but are sometimes applied to samples larger than the related theories indicate. This paper examines the statistical consequences of estimation in overexpansive samples with unmodeled treatment effects; we argue that samples containing cases that cannot experience the causal phenomenon in question produce unmodeled treatment effects, and we reexamine three published articles whose samples are perhaps broader than their theories suggest they should be. The empirical analyses demonstrate that overexpansive samples can produce somewhat misleading results: the new models produce interesting findings that emerge as treatment effects are identified.

Page Thumbnails

  • Thumbnail: Page 
[417]
    [417]
  • Thumbnail: Page 
418
    418
  • Thumbnail: Page 
419
    419
  • Thumbnail: Page 
420
    420
  • Thumbnail: Page 
421
    421
  • Thumbnail: Page 
422
    422
  • Thumbnail: Page 
423
    423
  • Thumbnail: Page 
424
    424
  • Thumbnail: Page 
425
    425
  • Thumbnail: Page 
426
    426
  • Thumbnail: Page 
427
    427
  • Thumbnail: Page 
428
    428
  • Thumbnail: Page 
429
    429