Mounting evidence indicates issues with low adherence to existing consensus-based guidelines for conducting systematic reviews (SRs), meaning that SRs can be subject to selective or misreporting practices. This problem arises in part from scarce guidance for reproducible reporting practices. This is compounded by the fact that existing guidelines are mainly applicable to interventional research designs, with systematic reviewers of non-interventional studies resorting to customised tools that deviate from best practice. Here, we present the development of the first comprehensive tool for conducting and reporting Non-Interventional, Reproducible, and Open Systematic Reviews (NIRO-SR). NIRO-SR is a 68-item checklist composed of two parts that provide itemised guidance on the preparation of a protocol for pre-registration (Part A) and reporting the review (Part B) in a reproducible and transparent manner. This paper, the tool, and an open repository (https://osf.io/f3brw/) provide a comprehensive resource for anyone who aims to conduct a high quality SR of non-interventional studies.

Rape cases have a disproportionately high attrition rate and low conviction rate compared to other criminal offenses. Evaluations of a rape complainant's credibility often determine whether a case progresses through the criminal justice system. Even though emotional demeanor is not related to witness honesty or accuracy, distressed rape complainants are perceived to be more credible than complainants who present with controlled affect. To understand the extent and robustness of the influence of emotional demeanor on credibility judgments of female adult rape complainants, we conducted a systematic review, metaanalysis and p-curve analysis of the experimental simulated decision-making literature on the influence of complainant emotional demeanor on complainant credibility. The meta-analysis included 20 studies with participants who were criminal justice professionals (e.g., police officers and judges), community members, and mock jurors (N = 3128). Results suggest that distressed demeanor significantly increased perceptions of complainant credibility, with a small to moderate effect size estimate. Importantly, the results of p-curve analysis suggest that reporting bias is not a likely explanation for the effect of emotional demeanor on rape complainant credibility. Sample type (whether perceivers were criminal justice professionals or prospective jurors) and stimulus modality (whether perceivers read about or watched the complainant recount the alleged rape) were not found to moderate the effect size estimate.These results suggest that effective methods of reducing reliance on emotional demeanor to make credibility judgments about rape complainants should be investigated to make credibility assessments fairer and more accurate.Public significance statement: This meta-analysis indicates that rape complainants with distressed emotional demeanor are perceived as more credible than their emotionally controlled counterparts. As emotional demeanor is not a reliable indicator of honesty, and

It is difficult to explain why verbal street harassment, where typically a male harasser yells sexually harassing statements at a female victim, has survived as a behaviour. We propose that verbal street harassment may signal a harasser’s dominance and aim to test this in our registered report.