Figure 1.
Three possible critical period effects.
Figure 2.
Illustration of the difference between correlation coefficients and slopes.
Relationships on the same row were generated by the same underlying function ( and
, respectively) but are characterised by different correlation coefficients (
and
, respectively). The inverse is true for relationships in the same column.
Table 1.
Descriptive statistics for the extracted data for the North America and Israel studies.
Table 2.
Correlation coefficients for the relationship between AOA and GJT based on the extracted data for the North America and Israel studies.
Figure 3.
Scatterplot of the AOA–GJT relationship in the North America study.
The trend line is a non-parametric scatterplot smoother. The scatterplot itself is a near-perfect replication of DK et al.'s Fig. 1.
Figure 4.
Scatterplot of the AOA–GJT relationship in the Israel study.
The trend line is a non-parametric scatterplot smoother. The scatterplot itself is a near-perfect replication of DK et al.'s Fig. 5.
Table 3.
Linear regression models containing no breakpoints.
Figure 5.
Regression lines for the North America data.
Solid: regression with breakpoint at aoa 18 (dashed lines represent its 95% confidence interval); dot-dash: regression without breakpoint.
Figure 6.
Regression lines for the Israel data.
Solid: regression with breakpoint at aoa 18 (dashed lines represent its 95% confidence interval); dot-dash (hardly visible due to near-complete overlap): regression without breakpoint.
Table 4.
Linear regression models containing breakpoints at AOA 18.
Figure 7.
Deviances for regression models assuming breakpoints as a function of the position of the breakpoints.
Figure 8.
Regression lines for the North America data.
Solid: regression with breakpoint at aoa 16 (dashed lines represent its 95% confidence interval); dot-dash: regression without breakpoint.
Figure 9.
Regression lines for the Israel data.
Solid: regression with breakpoint at aoa 6 (dashed lines represent its 95% confidence interval); dot-dash (hardly visible due to near-complete overlap): regression without breakpoint.
Table 5.
Regression model for the North America data without a breakpoint at AOA 16.
Table 6.
Regression model for the North America data with a breakpoint at AOA 16.
Table 7.
Regression model for the Israel data without a breakpoint at AOA 6.
Table 8.
Regression model for the Israel data with a breakpoint at AOA 6.
Table 9.
Partial correlation coefficients for the relationship between AOA and GJT with AAT controlled for.
Table 10.
Correlation coefficients for the relationship between aptitude and GJT.