The conditional nature of publication bias: a meta-regression analysis

Abstract

Abstract Publication bias is pervasive in social and behavioral sciences because journals and scholars tend to reward and be rewarded for statistically significant findings. However, the determinants of the severity of publication bias are less well understood. We argue that publication bias depends on whether an independent variable is a key variable or statistical control in traditional regression modeling. The bias should be severe only for the key variable that relates to a central question and hypothesis in a study. We offer an empirical strategy to detect the conditional nature of publication bias. As an illustration, we perform a meta-regression of 229 model estimates from 36 articles in the democracy-foreign direct investment literature. We find that publication bias is most severe when democracy is a key variable, but appears weak when democracy is a control. Our research demonstrates that empirical estimates for key and control variables follow different data generation processes and makes a novel contribution to the study of publication bias that affects many research areas.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.