Calls have been made before to pay attention to the correct estimation of non-experimental causal models; the major culprit is endogeneity, where the effect of x on y cannot be interpreted because it includes omitted causes such as omitted selection, simultaneity, common-method variance, and measurement error. Endogeneity problem has been noted both in psychology and management and a recent review found that more than 90% of papers published in the premier strategy journal (and one of the top journals in management), Strategic Management Journal (SMJ), were not correctly estimated. The leadership quarterly issue 21 of 2010 finds that least 66% and up to 90% of design and estimation conditions that make causal claims are invalid. These are threatening statistics for researchers. Is it possible to know from my colleagues in empirical financial accounting and corporate governance if they have seen how published research has dealt with this matter and how widespread it is? Are papers in this area using solutions such as instrumental variables, sample selection models, fixed-effects panel, sample selection, instrumental variable, regression discontinuity, and difference-in-differences models.
Im told part of the problem in publishing in top tier journal is failure to make valid conclusions taking into account this kind of problems.
Do you have examples of meta analysis in Capital Structure surveys ( from Central Bank/Bank or Borrower perspectives?
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments