Table 3.
Heterogeneity tests of effect sizes (Hedges's g) for potential moderators.
Potential moderators | Q | N | g | SE |
---|---|---|---|---|
Study designa | 61.830* | |||
Within subjects | 5 | 0.328* | 0.037 | |
Between subjects | 7 | 0.529* | 0.126 | |
Mixed | 38 | 0.759* | 0.041 | |
Sexa | 9.405* | |||
Girls | 5 | 0.909* | 0.143 | |
Boys | 5 | 0.686* | 0.141 | |
Not specified | 40 | 0.499* | 0.028 | |
Agea | 0.000 | |||
0–3 years | 5 | 0.518* | 0.153 | |
4–8 years | 45 | 0.520* | 0.027 | |
Spatial skills outcomesa | 111.263* | |||
Intrinsic–static | 3 | 0.456* | 0.145 | |
Intrinsic–dynamic | 31 | 0.952* | 0.050 | |
Extrinsic–static | 5 | 0.770* | 0.154 | |
Measure that spans cells | 11 | 0.326* | 0.033 | |
Research settingb | 4.229 | |||
Lab | 17 | 0.690* | 0.169 | |
Classroom | 19 | 1.158* | 0.155 | |
Others | 14 | 0.989* | 0.177 | |
Type of trainingb | 1.673 | |||
Video game/play/hands-on operation | 21 | 1.069* | 0.156 | |
Classroom-based course | 17 | 0.993* | 0.171 | |
Spatial task training | 12 | 0.752* | 0.194 |
N, number of effect sizes; SE, standard error;
p < 0.01.
Fixed effect approach is used for the moderator analysis of this variable.
Random effects approach is used for the moderator analysis of this variable.