No System is Perfect: Understanding How Registration-Based Editorial Processes Affect Reproducibility and Investment in Research Quality--Free Responses to Survey on Author Discretion

Abstract

This document includes the complete set of free-text responses to our survey on the 2017 Journal of Accounting Research Conference, Registered Reports of Empirical Research. We solicited responses from all authors who have published in six well-regarded accounting journals over the last decade. Respondents provide thoughts on how research quality is affected by authors’ discretion to choose: which measures and analyses to report; whether or not to report a sample or subsample; which observations to exclude from primary analyses; whether to gather additional data after observing results; to change their predictions and/or underlying theory; and to choose not to report or highlight hypotheses. Respondents also provide first-hand accounts of discretionary choices and failed replications, advice to editors and reviewers, and their thoughts on other forms of discretion, motivations for using discretion, and any insights they wish to share not covered by their previous answers.