# Attrition Analysis Summary | paper_name | supplement_included | attrition_analysis_confidence | summary | | ---------- | ------------------- | ----------------------------- | ------------------------------------------------------------ | | Lee | Yes | 0 | No attrition analysis performed. The Nature Reporting Summary explicitly states "No participants dropped out or declined participation." No attrition occurred, so no analysis was warranted. | | Stagnaro | Yes | 92 | Substantial attrition analysis. Supplement Section 2 is a dedicated attrition section reporting 10.4% dropout, differential attrition tests (p=.979), instrumental variable models, and exit-point design preserving 72.2% of dropout outcomes. Section 13 compares Wave 1 vs Wave 2 demographics. Section 14 re-runs models on Wave 2 returners, finding substantively identical results. | | Vlasceanu | Yes | 5 | Essentially no formal attrition analysis. Reports that 29.2% of participants were excluded via preregistered attention checks. Acknowledges this "risks contributing to a selection bias" but does not compare excluded vs included participants on any variables. No differential attrition by condition examined. | | Voelkel | Yes | 98 | Extremely thorough attrition analysis (supplement Section S13.2). Preregistered differential attrition F-tests found rates ranging 1.6%-23.0% across conditions. Used preregistered inverse-probability weighting (IPW) via random forests, attriter recapture strategy (~47% recaptured), alternative control condition with similar attrition patterns, and multiple robustness checks. Results consistent across all strategies. |