The digital story meta-analysis is resistant to publication bias by fail-safe N and trim-and-fill tests¶

Claim · All claims
Evidence · 1 study · quant-synthesis r2 · q3
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Rosenthal's fail-safe N was 7854, meaning 7854 zero-effect studies would be needed to render the pooled effect of 0.750 non-significant, and the Mullen formula value of 16.3625 exceeds 1. → Nurullah Sahin and İsmail Coban 2020

Evidence¶

Nurullah Sahin and İsmail Coban 2020¶

Nurullah Sahin and İsmail Coban. (2020). The effect of digital story applications on students' academic achievement: A meta-analysis study. African Educational Research Journal, Special Issue 8(3), S62-S75. https://eric.ed.gov/?id=EJ1274516

q3 · i? · quant-synthesis · r2

Publication bias assessment in the Findings section using funnel plot inspection, Duval and Tweedie's trim and fill (14 artificial studies added, adjusted ES = 0.822), Rosenthal's fail-safe N (7854), and the Mullen et al. formula ([7854 / (5 * 94 + 10) = 16.3625]). The authors conclude the meta-analysis is resistant to publication bias.

"the fail-safe N number was calculated as 7854. This value indicates that in order for the common effect size calculated as 0.750 to be statistically insignificant, 7854 studies with zero effect level should be included in the meta-analysis."

Discussion¶

Related Claims¶

• Digital story applications have a high positive average effect on students' academic achievement compared to other teaching methods (ES = 0.750) — related
• Publication-bias checks give mixed results: Egger's test indicates funnel-plot asymmetry while Trim and Fill and Fail-Safe N indicate no bias — related
• No publication bias was detected among the studies included in the meta-analysis — related
• Publication-bias checks indicate the pooled effect is unlikely to result from publication bias — related
• Publication-bias analyses indicate a low, negligible risk of publication bias in the synthesized effect sizes — related
• The meta-analytic effect of learning stations is resistant to publication bias — related
• Publication bias in the meta-analysis is low, as indicated by a fail-safe N of 4860 and funnel plot inspection. — related
• Publication-bias analysis indicates 51 unpublished studies would be needed to reverse the meta-analysis findings — related
• The pooled effect of multiple intelligence practices is resistant to publication bias — related
• The meta-analysis shows no publication bias problem by Rosenthal's fail-safe N criterion — related
• The meta-analysis is resistant to publication bias by Rosenthal's fail-safe N criterion — related
• Publication-bias checks (funnel plot symmetry and Rosenthal Safe N of 6801) indicate no publication bias in the synthesis — related
• The meta-analysis results are resistant to publication bias, per fail-safe N of 1791 — related