Fig. 2: Forest plot of individual study effect sizes and pooled estimate. | Nature Mental Health

Fig. 2: Forest plot of individual study effect sizes and pooled estimate.

From: Creative arts-based interventions for the improvement of PTSD symptoms in young people: a meta-analysis with a focus on non-Western populations

Fig. 2: Forest plot of individual study effect sizes and pooled estimate.

Effect sizes (θᵢ individual study Hedges’ g) and their 95% CIs were calculated for each study. The pooled effect size (θ) was estimated using a random-effects model with restricted maximum likelihood (REML). Heterogeneity across studies was evaluated using Cochran’s Q test (Q(32) = 1,752.01, P < 0.001) which tests the null hypothesis that all study effects are equal (θᵢ = θⱼ). The I² statistic (I² = 98.58%), H2 = 70.26 and τ² = 0.06. Further quantify the magnitude of between-study variability beyond chance. The overall pooled effect was statistically significant (z = 10.84, P < 0.001), with an estimated effect size of 0.85 (95% CI 0.70 to 1.00). All tests were two-sided. No adjustments for multiple comparisons were applied as this analysis addressed a single pooled hypothesis test.

Back to article page