Table 4.

Moderation of Mentoring Effects (Random Effects Models)

Moderator	Level of Moderator								Meta-Regression
	Absent				Present
	k	SMD	L	U	k	SMD	L	U	B	SE
Mentee Selection
Individual Risk	22	0.20	0.05	0.35	16	0.23	0.11	0.35	0.03	0.09
Environmental Risk	28	0.20	0.09	0.31	8	0.23	-0.06	0.51	0.03	0.12
Other Interventions	23	0.20	0.06	0.34	23	0.31	0.13	0.49	0.07	0.10
Motivations of Mentors
Civic Duty	11	0.24	0.00	0.47	21	0.22	0.09	0.35	0.02	0.11
Professional Development	20	0.16	0.05	0.27	16	0.42	0.16	0.68	0.21*	0.11
Quality and Fidelity Checks
Quality Check	14	0.20	0.06	0.35	20	0.21	0.05	0.38	-0.00	0.10
Fidelity Check	27	0.20	0.09	0.30	6	0.29	-0.15	0.73	0.05	0.14
Key Processes
Modeling/Identification	28	0.24	0.08	0.40	11	0.32	0.08	0.56	0.06	0.12
Emotional Support	12	0.11	0.00	0.23	27	0.32	0.14	0.50	0.22*	0.12
Teaching	11	0.12	-0.01	0.24	30	0.29	0.15	0.44	0.15	0.10
Advocacy	32	0.13	-0.05	0.31	10	0.39	0.06	0.72	0.17*	0.09

Notes:

^*p <.05, one-tailed

Random effects models of standardized mean differences (SMD) are the sources of the significance tests for the SMDs within levels of each moderator. The meta-regression models are mixed effects models using full maximum likelihood estimation.

k = number of studies, SMD = standardized mean difference, L = lower limit of the 95% confidence interval for the SMD, U = upper limit of the 95% confidence interval for the SMD