Table 5. Independent-Groups Post-test Comparisons: Dichotomous Moderator Analyses.

Moderator	Coded values	k0	k1	${\widehat{\mu }}_{{\delta }_0}$	${\widehat{\mu }}_{{\delta }_1}$	QB	QW	${\widehat{\sigma }}_{\delta }$
Publication status	0 = unpublished, 1 = published	15	114	.44***	.46***	0.0	443.8***	.302
Funding	0 = no funding, 1 = funding	35	94	.60***	.41***	5.1*	399.0***	.283
SES	0 = low, 1 = moderate or high	4	6	.37†	.27	0.1	25.7**	.314
Random allocation	0 = absent, 1 = present	23	106	.58***	.43***	2.6	440.1***	.301
Behavioral target	0 = physical activity only, 1 = multiple health behaviors	55	74	.57***	.38***	7.4**	426.8***	.295
Supervised exercise	0 = absent, 1 = present	79	50	.41***	.53***	2.5	441.0***	.302
Tailoring	0 = absent, 1 = present	125	4	.45***	.66***	1.2	418.8***	.295
Mediated delivery	0 = face-to-face only, 1 = mail/telephone/mass media	97	32	.47***	.41***	0.6	443.1***	.304
Intervention social context	0 = individual, 1 = group	43	85	.48***	.45***	0.1	439.6***	.303
Social cognitive theory	0 = absent, 1 = present	97	32	.45***	.46***	0.0	439.9***	.303
Transtheoretical model	0 = absent, 1 = present	117	12	.48***	.22*	4.7*	427.8***	.295
Behavioral–any	0 = no behavioral strategy, 1 = at least 1 behavioral strategy	30	99	.28***	.51***	8.8**	412.7***	.289
Behavioral–pure	0 = cognitive or no behavioral, 1 = behavioral and no cognitivea	72	57	.40***	.53***	3.1†	438.0***	.301
Cognitive–any	0 = no cognitive strategy, 1 = at least 1 cognitive strategy	81	48	.46***	.44***	0.1	442.0***	.303
Cognitive–pure	0 = behavioral or no cognitive, 1 = cognitive and no behavioralb	123	6	.47***	.10	5.6	437.8***	.302
Barriers management	0 = absent, 1 = present	101	28	.45***	.46***	0.0	441.7***	.303
Consequences	0 = absent, 1 = present	111	18	.45***	.49***	0.2	430.4***	.299
Contracting	0 = absent, 1 = present	118	11	.44***	.60***	1.6	440.2***	.303
Exercise prescription	0 = absent, 1 = present	90	39	.42***	.53***	2.1	438.8***	.301
Feedback	0 = absent, 1 = present	99	30	.47***	.42***	0.4	442.4***	.304
Fitness testing	0 = absent, 1 = present	97	32	.43***	.55***	1.8	444.0***	.303
Goal setting	0 = absent, 1 = present	92	37	.47***	.43***	0.2	443.8***	.305
Problem solving	0 = absent, 1 = present	98	31	.44***	.50***	0.5	436.3***	.301
Self-monitoring	0 = absent, 1 = present	87	42	.40***	.56***	5.1*	397.1***	.281
Stimulus/cues	0 = absent, 1 = present	121	8	.45***	.56***	0.5	442.2***	.301
Recommended specific exer. form	0 = other than walking, 1 = walking	115	14	.46***	.45***	0.0	442.8***	.302
Recommended intensity	0 = low, 1 = moderate or high	5	40	.39†	.60***	0.9	142.5***	.396
Days after intervention	0 = none, 1 = at least one	43	80	.45***	.46***	0.0	435.6***	.314

Note. k_j = number of (possibly dependent) ES estimates in group coded j. Heterogeneity statistics: Q_B = between groups (distributed as chi-square on df = 1 under H₀: μ_δ0 = μ_δ1), Q_W = combined within groups (distributed as chi-square on df = k₀ + k₁ − 2 under H₀: ${\sigma }_{{\text{δ}}_{\text{0}}}^{\text{2}}={\text{σ}}_{{\text{δ}}_{\text{1}}}^2=0)$. Weighted method of moments used to estimate between-studies variance component ${\sigma }_{\text{δ}}^2$. Analysis reported if k₀ ≥3 and k₁ ≥3.

^†p < .10.

^*p < .05.

^**p < .01.

^***p < .001 (for ${\widehat{\mu }}_{{\text{δ}}_{\text{0}}},\hspace{0.17em}{\widehat{\mu }}_{{\text{δ}}_{\text{1}}}$, Q_B and Q_W).

^a0=no behavioral strategy or at least 1 cognitive strategy, 1=at least 1 behavioral strategy and no cognitive strategy.

^b0=no cognitive strategy or at least 1 behavioral strategy, 1=at least 1 cognitive strategy and no behavioral strategy.