Effect 1709: ML5 Risen RP:P replication p = .02, not p > .05
还没有人认领这个 Issue。
评估
- 难度
- 2/5
- 预计耗时
- 1-3 小时
- 新手友好度
- 50/100
- Issue 类型
- 缺陷
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 领域
- data
调研方向
Find the record with fred_id ManyLabs5_1g (effect 1709) in the FReD data and check its pval_type_r and pval_value_r fields against Table 4 of Ebersole et al. (2020). The proposed change is to set the type to e (=) and the value to 0.02 while keeping r = −.04. Done means the record matches the cited table, and the maintainers confirm the inequality-versus-equality coding convention first, since the thread has no reply yet.
由索引模型根据 Issue 内容生成。
描述
Effect 1709 (fred_id = ManyLabs5_1g) records pval_type_r = g (>), pval_value_r = 0.05, despite a published replication p-value of .02 for this arm.
The record identifies the ML5 RP:P-protocol replication of Risen & Gilovich (2008), original Study 6: study_r = 7_RPP, n_r = 2811, replication DOI 10.1177/2515245920958687. Its recorded es_type_r = r, es_value_r = -0.04 agrees with the overview's tables.
Source evidence in Ebersole et al. (2020):
- Table 3, p. 11: ML5 RP:P-protocol arm, N = 2,811, r = −.04, 95% CI [−.08, −.01]. The original effect is r = +.19.
- Table 4, p. 14: Risen & Gilovich → RP:P protocol, k = 1, r = −.04 [−.08, −.01], p = .02 in the p-value column. The adjacent P_orig column also has .02 but represents a different metric.
- The all-protocols and revised-protocol results have p = .20 and .87, respectively; neither is the endpoint for effect 1709.
Two independent source coders (Opus and a blind Codex agent) agreed on N, estimate, CI, p-value, and direction. I also checked the Table 4 image. The printed p-value is rounded to two decimal places; sidedness is inferred from the estimate and zero-excluding 95% CI rather than explicitly labelled in the table.
Please change the replication p-value to pval_type_r = e (=), pval_value_r = 0.02, retaining r = −.04. This makes significance against zero true under a two-sided p < .05 endpoint. It remains unsuccessful in reproducing the original direction, so reported_success = failed need not change.
The overview's narrative on pp. 10–11 says none of the RP:P-protocol replications were significant, which conflicts with this negative significant numerical result if read as significance against zero. The paper defines successful replication using significance in the original direction on p. 8. That may explain the narrative, but the reason for FReD's inequality coding is unconfirmed. Footnote 5 also distinguishes individual-report estimators from the overview meta-analyses; the recommendation here is specific to the cited overview and the matching N and effect estimate.
- 主要语言
- HTML
- 星标
- 1
- 派生
- 0
- PR 合并指标
- 30 天内没有已合并 PR
环境准备
这个项目没有提供开发容器、Dockerfile 或贡献指南，环境需要你自己搭建：先看它的 README，通用步骤见我们的新手贡献指南。
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
forrtproject/fred-data 的其他 Issue
-
Eight FLoRA replication DOIs point to PCI RR recommendations or reviews, not the replication report未关闭
难度 2/5 1-3 小时 新手友好度 68/100
forrtproject/fred-data#152 · 1 条评论 ·
-
难度 2/5 1-3 小时 新手友好度 76/100
forrtproject/fred-data#146 ·
-
难度 2/5 1-3 小时 新手友好度 78/100
forrtproject/fred-data#128 · 2 条评论 ·
-
难度 4/5 3-5 天 新手友好度 47/100
forrtproject/fred-data#154 ·
-
Decide whether report venue type (journal, preprint, conference, thesis) is recorded in the data未关闭
难度 5/5 一周以上 新手友好度 20/100
forrtproject/fred-data#147 ·
查看 forrtproject/fred-data 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 72/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 62/100
维护者通常 1 天内回复
-
new spec review
难度 2/5 1-3 小时 新手友好度 62/100
w3c/browser-specs#2708 · 1 条评论 ·
维护者通常 1 天内回复
-
data good first issue
难度 2/5 半天 新手友好度 72/100
-
难度 1/5 1 小时以内 新手友好度 68/100
dwyl/english-words#244 ·