Biases of success rate differences shown in binomial effect size displays
Biases of success rate differences shown in binomial effect size displays
复制标题
DOI：
10.1037/1082-989x.9.2.183
复制
发表时间：
2004-06-01
影响因子：
7
通讯作者：
Hsu, LA
中科院分区：
文献类型：
--
作者：
Hsu, LA
二项效应大小显示（BESD）的目的是通过将这种相关性重新表示为成功率差异（SRD）（例如，治疗组成功率-对照组成功率）来显示“以相关性[r]为索引的[一]效果的[现实世界]重要性”（r . Rosenthal, 1994，第242页）。然而，besd中显示的SRDs通常高估了(a)二分类X和Y变量（phi系数），(b)二分类X和连续Y变量（点双列系数[r(pb)S]）以及(c)连续X和Y变量（r(xy) S）的相关性所隐含的实际SRDs。此外，r(xy)s的高估偏差大于r(pb)s。与不同相关性相关的偏倚大小的差异表明，不同相关性报告的BESD SRDs不具有可比性。随机差异指数（N. Cliff, 1993； A. Vargha & H. D. Delaney, 2000）被推荐作为BESD的替代方法。
The intent of a binomial effect size display (BESD) is to show "the [real-world] importance of [an] effect indexed by a correlation [r]" (R. Rosenthal, 1994, p. 242) by reexpressing this correlation as a success rate difference (SRD) (e.g., treatment group success rate-control group success rate). However, SRDs displayed in BESDs generally overestimate real-world SRDs implied by correlations of (a) dichotomous X and Y variables (phi coefficients), (b) dichotomous X and continuous Y variables (point-biserial coefficients [r(pb)S]), and (c) continuous X and Y variables (r(xy)s). Furthermore, overestimation biases are larger for r(xy)s than for r(pb)s. Differences in the sizes of biases linked to different correlations suggest that BESD SRDs reported for different correlations are not comparable. The stochastic difference index (N. Cliff, 1993; A. Vargha & H. D. Delaney, 2000) is recommended as an alternative to the BESD.