In addition to inspection of the funnel plots, correlations between effect size and sample size and methodological quality measures were calculated (Levine et al., 2009).
Tests for potential publication bias found no significant correlation between effect size and sample size (rs, p>0.05) in all cases, suggesting that file-drawer effects did not strongly bias our results.
Thus one possibility is to investigate the relation between effect size and sample size (or more specifically, precision of the effect estimate).
27 If sufficient studies are found, funnel plots will be drawn to investigate any relationship between effect size and sample size.
This weighted linear ordinary least squares regression model establishes a linear association between-study effect and its sample size weighted by a function of sample size.
This article is a follow-up study of our GWAS report to analyze potential sample stratification of the study population, the relationship between the significant effects and sample stratification, and the effect of sample stratification correction on our reported GWAS results.
Begg's rank correlation test uses the Spearman rank correlation to investigate the relationship between the standardised effect size and sample size or variance in effect size.
However, there has always been discussion about the strong negative correlation between observed effect sizes and sample sizes [ 41, 42].
This is unlikely to reflect a publication bias towards positive results amongst earlier publications since we found no association between effect magnitude, or sample size, and publication date; moreover, the summary ORs for candidate polymorphisms have not changed significantly over five years [25].
Furthermore, there is a correlation between the standardized effect sizes and sampling variances under the null hypothesis of no publication bias.
