效果量，很多书里都解释为效果量

是不是“效应大小”呢？

Effect Sizes & Confidence Intervals

    * Meta analysis reports findings in terms of effect sizes.  The effect size provides information about how much change is evident across all studies and for subsets of studies.
    * There are many different types of effect size, but they fall into two main types:
          o standardized mean difference (e.g., Cohen's d or Hedges g) or
          o correlation (e.g., Pearson's r)
    * It is possible to convert one effect size into another, so each really just offers a differently scaled measure of the strength of an effect or a relationship.
    * The standardised mean effect size is basically computed as the difference score divided by the standard deviation of the scores.
    * In meta-analysis, effect sizes should also be reported with:
          o the number of studies and the number of effects used to create the estimate.
          o confidence intervals to help readers determine the consistency and reliability of the mean estimated effect size.
    * For more information about calculating effect sizes and confidence intervals, see:
          o Effect size calculation - David Wilson
          o Overview of different effect sizes
          o Effect sizes calculation program
    * Tests of statistical significance can also be conducted and on the effect sizes.
    * Different effect sizes are calculated for different constructs of interest, as predetermined by the researchers based on what issues are of interest in the research literature.
    * Rules of thumb and comparisons with field-specific benchmarks can be used to interpret effect sizes.  According to an arbitrary but commonly used interpretation of effect size by Cohen (1988), a standardised mean effect size of 0 means no change, negative effect sizes mean a negative change, with .2 a small change, .5 a moderate change, and .8 a large charge. Wolf (1986), on the other hand, suggests that .25 is educationally significant and .50 is clinically significant.

什么是效果规模值(effect size)？

这是用来描述两个变量之间关系强度 (strength of the relationship between two variables) 的指标（具体参见前文“解释变量关系时必须考虑的四个问题”）。最常见的效果规模值有用于回归分析类的为皮尔逊相关系数 r 和用于均值比较类的 Cohen's d 两种，其它还有用比例比较类的Odds ratio 等（详见http://en.wikipedia.org/wiki/Effect_size）。

为何要转换成效果规模值？

效果规模值是标准化指标，绝对值在0和1之间，0表示完全无效、1表示完全有效，而两极之间的0.3或0.7通常代表弱效或强效的临界点。

[img]file:///C:/Documents%20and%20Settings/saber/Application%20Data/Tencent/Users/395242626/QQ/WinTemp/RichOle/RD6R8PH)YP_OPG[10O53L)4.jpg[/img]
这个里面的柱状长度是表示effect size吗？怎么跟上面的数字大小不符合呢？而且图下面解释说31的效果明显，16,和6效果不明显，希望有人帮我解答啊