Y-axis is sample size, X-axis is effect size. I think that for this meta analysis they used Cohen’s d, though I don’t remember exactly. (It was actually Glass’ delta)
When there’s no publication bias we expect studies to be somewhat evenly distributed to the left and right of the pooled estimate (red line), but we clearly see more low quality studies leaning to the bottom right. And it also almost looks like there’s a hard cut at 0, showing that researchers are clearly discouraged from publishing studies showing that people who had sexual contact with an adult as a kid had positive outcomes.
Y-axis is sample size, X-axis is effect size.
I think that for this meta analysis they used Cohen’s d, though I don’t remember exactly.(It was actually Glass’ delta)When there’s no publication bias we expect studies to be somewhat evenly distributed to the left and right of the pooled estimate (red line), but we clearly see more low quality studies leaning to the bottom right. And it also almost looks like there’s a hard cut at 0, showing that researchers are clearly discouraged from publishing studies showing that people who had sexual contact with an adult as a kid had positive outcomes.