Cover Image
QR Code

Detecting Model Dependence in Statistical Inference: A Response.

The article presents information on the need of researchers to be able to more readily identify model dependence to improve their own work and reanalyze data from existing articles and reevaluate statistical results and conclusions. Standard uncertainty measures such as standard errors and confidenc...

Full description

Published in:International Studies Quarterly Vol. 51; no. 1; pp. 231 - 242
Main Authors: King, Gary, Zeng, Langche
Format: Article
Published: Oxford University Press / USA Mar2007
Subjects:
Research methodology
Quantitative research
Scientific errors
Counterfactuals (Logic)
Scientific method
Online Access:View this record in EBSCOhost
Summary:The article presents information on the need of researchers to be able to more readily identify model dependence to improve their own work and reanalyze data from existing articles and reevaluate statistical results and conclusions. Standard uncertainty measures such as standard errors and confidence intervals can often be massively underestimated when counterfactuals are posed too far from available data and lead to high degrees of model dependence.