Table 4.

Results of moderator analyses investigating the effect of differences in intervention study and methodological characteristics on work engagement.

Outcome	k	n (int)	n (con)	Intervention effects			p‐value	Heterogeneity within each subgroup
				g	SE	95%‐CI		Q	df	p	I2
Intervention type
Health promotion	4	806	815	0.14	0.06	0.02–0.26	0.03	4.24	3	0.24	29.23
Job resources	2	88	173	0.40	0.22	−0.04–0.84	0.08	2.84	1	0.09	64.75
Leadership training	4	371	337	0.14	0.08	−0.01–0.30	0.07	1.72	3	0.63	0.00
Personal resources	4	493	609	1.00	0.61	−0.20–2.20	0.10	42.08	3	0.00	92.87
					Heterogeneity between			3.18	3	0.36
Intervention style
Group	8	828	797	0.51	0.20	0.12–0.90	0.01	43.28	7	0.00	83.83
Group and individual	3	536	479	0.07	0.06	−0.05–0.20	0.26	0.98	2	0.61	0.00
Individual	1	38	97	0.63	0.19	0.25–1.01	0.00	0.00	0	1.000	0.00
Online and individual	2	356	561	0.17	0.11	−0.05–0.38	0.14	2.27	1	0.13	55.88
					Heterogeneity between			10.89	3	0.01
Type of organisation
Private	5	535	573	0.24	0.09	0.07–0.41	0.01	5.99	4	0.20	33.21
Public	4	549	676	0.12	0.10	−0.03–0.38	0.90	8.41	3	0.04	64.31
					Heterogeneity between			1.76	2	0.42

Notes. k = number of studies included in the analysis; n(con) = number of participants in the control group; n(int) = number of participants in the intervention group; g = average effect size according to Hedges' g; SE = standard error of the average effect size; 95%‐CI, LL‐UL = the minimum and maximum limits of the 95% confidence interval; Q = statistical test used for the estimation of heterogeneity; I² = proportion of effect size variance that can be attributed to moderator variables (%).