Table 4.
Results from PET-PEESE indicator based on Stanley and Doucouliagos (2014) approximation to reduce publication selection bias.
| Sample | PET | PEESE | ||
|---|---|---|---|---|
| β0 | β1 | β0 | β1 | |
| Full | 0.41** (0.27, 0.54) | 1.86** | 0.53** (0.43, 0.63) | 4.66** |
| MODEL OF AFFECT | ||||
| Dimensional | 0.38** (0.27, 0.48) | 1.72** | 0.47** (0.38, 0.55) | 4.38** |
| Discrete | 0.55** (0.21, 0.89) | 1.28 | 0.66** (0.46, 0.86) | 3.26 |
Full, the full sample. For PET and PEESE, β0, the intercept (i.e., the corrected estimate of the overall effect); β1, the coefficient for standard error or variance (i.e., the test for funnel plot asymmetry). Numbers given in parentheses are the lower and upper limits of the 95% confidence intervals.
**
p < 0.01.