Table 2
Least squares regression results for most likely thresholds (T) of age-at-arrival impacts on not having graduated from high school, men and women

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

Symbols Next Previous

Least squares regression results for most likely thresholds (T) of age-at-arrival impacts on not having graduated from high school, men and women: Part 1


Table summary
This table displays the results of least squares regression results for most likely thresholds (t) of age-at-arrival impacts on not having graduated from high school men, women and t = 7, calculated using t = 7, t = 8, t = 10 and coefficient units of measure (appearing as column headers).
  Men Women
T = 7 T = 7 T = 8 T = 10 T = 10 T = 8
  coefficient
Variables  
Age at arrival 0.0001  Note 3 -0.0018  Note 3 -0.0008  Note 3 0.0022  Note 3 0.0012  Note 3 -0.0005  Note 3
Threshold (T) -0.0138  Note 3 -0.0081  Note 3 0.0025  Note 3 -0.0123  Note 3 -0.0100  Note 3 -0.0107  Note 3
T × age at arrival 0.0112  Note 1 0.0106  Note 1 0.0127  Note 1 0.0168  Note 1 0.0152  Note 1 0.0163  Note 1
Mother tongue Note ...: not applicable 0.1120  Note 1 0.1430  Note 1 Note ...: not applicable 0.1120  Note 1 0.2200  Note 1
Constant 0.1500  Note 1 0.0930  Note 1 0.0592  Note 1 0.1310  Note 1 0.0725  Note 1 0.0218  Note 1
... not applicable

Least squares regression results for most likely thresholds (T) of age-at-arrival impacts on not having graduated from high school, men and women: Part 2


Table summary
This table displays the results of least squares regression results for most likely thresholds (t) of age-at-arrival impacts on not having graduated from high school men and women, calculated using t = 7, t = 8 and t = 10 units of measure (appearing as column headers).
  Men Women
T = 7 T = 7 T = 8 T = 10 T = 10 T = 8
Indicators for region of origin none none eleven none none eleven
Diagnostic statistics  
Adjusted R-squared 0.0980  Note 2 0.2840  Note 2 0.7900  Note 2 0.1680  Note 2 0.3580  Note 2 0.7330  Note 2
F test - p value 0.0010  Note 2 0.0000  Note 2 0.0000  Note 2 0.0000  Note 2 0.0000  Note 2 0.0000  Note 2
1.
Marginal significance is less than or equal to 0.05.
2.
Marginal significance is greater than 0.05 and less than or equal to 0.10.
3.
Marginal significance is greater than 0.10.
Note(s):
Derivation by author from Statistics Canada, 2006 Census using analytical files described in the text. T refers to the threshold used in the estimation of a piece-wise linear least squares model. Results presented are those for models with highest adjusted R-squared among all possible thresholds from 5 to 12 years. Sample size is 198 observations on weighted averages by age at arrival from 0 to 17 for each gender, representing 50,253 men and 51, 631 women. All regressions use analytical weights. Standard errors are robust to hetroscedasticity. Region of origin controls refer to indicator variables for the eleven regions categorized by linguistic distance from English and French as described in the text, with the omitted category for the sake of the regression analysis being the group of English-speaking countries other than the United Kingdom or the United States. P value for the F test refers to the null hypothesis that all regressors except the constant equal 0.
Date modified: