Then, we create a new filter variable outlier, with the value TRUE if the CI of the effect size is outside of the CI of the pooled effect. We first calculate the 95% CI for each study effect size, using the standard error of the effect size ( sqrt(vi)). To filter out outliers, we will use a boolean (TRUE/FALSE) filter variable. We can use these values to filter out outliers now. The pooled effect confidence interval stretches from \(g = 0.16\) to \(g = 0.32\). Let’s see what the upper and lower bound of the pooled effect confidence interval are: m_re $ci.lb # 0.1581353 m_re $ci.ub # 0.3203839 Here, i’ll use my m_re meta-analysis output from Chapter 5.2.2 again.
for which the lower bound of the 95% confidence interval of the study is higher than the higher bound of the pooled effect confidence interval (i.e., extremely large effects). for which the upper bound of the 95% confidence interval of the study is lower than the lower bound of the pooled effect confidence interval (i.e., extremely small effects). Using this function, we can search for all studies: To detect such outliers in our dataset, the filter function in the dplyr package we introduced in Chapter 3.3.3 comes in handy again. 11.1.1 Three-level meta-analytic modelsħ.3.1 Searching for extreme effect sizes (outliers)Ī common method to detect outliers directly is to define a study as an outlier if the study’s confidence interval does not overlap with the confidence interval of the pooled effect. 11.1 Meta-Analysis is multi-level (optional). 10.1.5 Duval & Tweedie’s trim-and-fill procedure. 10.1.4 Testing for funnel plot asymmetry using Egger’s test. 7.3.1 Searching for extreme effect sizes (outliers). 7.3 Detecting outliers & influential cases. 7.2 Assessing the heterogeneity of your pooled effect size. 5.2.1 Estimators for tau 2 in the random-effects-model. 4.1 Calculating standardized mean differences. 2.1 Getting RStudio to run on your computer.
0 kommentar(er)
