Table 3. Pearson’s R correlation between the predicted outcomes in the test set using the trained regression models and the clinical scores of the test set, for both the data-driven and manually chosen models.
| Pearson’s R correlation | Relative mean error (%) | 95% Confidence Interval | |
|---|---|---|---|
| AHA | 0.706** | 0.145 | (0.488, 0.841) |
| BRIEF | 0.482 | -0.303 | (0.124, 0.729) |
| SDQ | 0.795** | -0.419 | (0.456, 0.932) |
| TVPS | 0.617* | 0.099 | (0.252, 0.827) |
| WR | 0.545* | -0.686 | (0.161, 0.786) |
| VOC | 0.682** | 0.343 | (0.354, 0.860) |
Correlations in bold have a statistical significance of p < 0.008. Asterisked correlations were found to be statistically significant after multiple comparisons:
* p < 0.008
** p < 0.0016.
AHA, Assisting Hand Assessment; BRIEF, Behaviour Rating Inventory of Executive Function; SDQ, Strengths and Difficulties Questionnaire; TVPS, Test of Visual Perception Skills; VOC, vocabulary; WR, Word reasoning