We present explorable multiverse analysis reports, a new approach to statistical reporting where readers of research papers can explore alternative analysis options by interacting with the paper itself. This approach draws from two recent ideas: i) multiverse analysis, a philosophy of statistical reporting where paper authors report the outcomes of many different statistical analyses in order to show how fragile or robust their findings are; and ii) explorable explanations, narratives that can be read as normal explanations but where the reader can also become active by dynamically changing some elements of the explanation. Based on five examples and a design space analysis, we show how combining those two ideas can complement existing reporting approaches and constitute a step towards more transparent research papers. CCS CONCEPTS • Human-centered computing → Human computer interaction (HCI).

Background: Vulvar vestibulitis syndrome (VVS) is a chronic, persistent syndrome characterised by vestibular pain, tenderness, and erythema. The aetiology of VVS is unknown and few of the hypothesised risk factors have been tested in controlled studies. Methods: Using a matched case-control study design, medical, sexual, health behaviour, and diet history of 28 women with VVS were compared with 50 friend controls without VVS to identify possible causal factors. Results: Cases were more likely than controls to report every vaginal and urinary symptom at the time of interview measured, particularly vaginal soreness or pain (60.7%) and pain during intercourse (64.3%). There were no significant diVerences between cases and controls with respect to sexual behaviour. Cases were more likely than controls to report self reported history of physician diagnosed bacterial vaginosis (OR=22.2, 95%CI=2.8, 177.2, p value=0.0001), vaginal yeast infections (OR=4.9, 95%CI=1.4, 18.0, p value=0.01), and human papillomavirus (OR=7.1, 95%CI=0.6, 81.2, p value=0.08). There were no diVerences between cases and controls with respect to dietary intake of oxalate. Cases were more likely than controls to report poor health status (OR=5.7, 95%CI=1.1, 28.7, p value=0.02) and history of depression for 2 weeks or more during the past year (OR=4. 4, 95%CI=1.6, 12.3, p value=0.002). Conclusion: Self reported history of bacterial vaginosis, yeast infections, and human papillomavirus were strongly associated with VVS. An infectious origin for VVS should be pursued in larger controlled studies, using questionnaire and laboratory measures. (Sex Transm Inf 1999;75:320-326)

No abstract

Bayesian statistical analysis is steadily growing in popularity and use. Choosing priors is an integral part of Bayesian inference. While there exist extensive normative recommendations for prior setting, little is known about how priors are chosen in practice. We conducted a survey (N = 50) and interviews (N = 9) where we used interactive visualizations to elicit prior distributions from researchers experienced with Bayesian statistics and asked them for rationales for those priors. We found that participants' experience and philosophy influence how much and what information they are willing to incorporate into their priors, manifesting as different levels of informativeness and skepticism. We also identified three broad strategies participants use to set their priors: centrality matching, interval matching, and visual probability mass allocation. We discovered that participants' understanding of the notion of "weakly informative priors"---a commonly-recommended normative approach to prior setting---manifests very differently across participants. Our results have implications both for how to develop prior setting recommendations and how to design tools to elicit priors in Bayesian analysis.

Bayesian statistical analysis is steadily growing in popularity and use. Choosing priors is an integral part of Bayesian inference. While there exist extensive normative recommendations for prior setting, little is known about how priors are chosen in practice. We conducted a survey (N = 50) and interviews (N = 9) where we used interactive visualizations to elicit prior distributions from researchers experienced with Bayesian statistics and asked them for rationales for those priors. We found that participants' experience and philosophy influence how much and what information they are willing to incorporate into their priors, manifesting as different levels of informativeness and skepticism. We also identified three broad strategies participants use to set their priors: centrality matching, interval matching, and visual probability mass allocation. We discovered that participants' understanding of the notion of "weakly informative priors"-a commonly-recommended normative approach to prior setting-manifests very differently across participants. Our results have implications both for how to develop prior setting recommendations and how to design tools to elicit priors in Bayesian analysis.