Table 16.1
Adjusted patterns of association of public and private sector employment with speed of closure of trajectory of work-to-retirement transition during 1998 to 2001, for three nested models, Canada
(Persons who began their transitions to retirement during 1996 to 1997)

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.

Categories of the two-year class-of-worker variable Degrees of Freedom Model One logit = (speedcls is 2)/ (speedcls is 1,3,4) Model Two logit = (speedcls is 3)/ (speedcls is 1,4) Model Three logit = (speedcls is 1)/ (speedcls is 4)
A1 B2 C3
Parameter estimates and Wald chi-squares4
Employed public sector in 1996 and 1997 (1) 1 0.1843 (0.4) -0.3859 (1.3) 0.4732 (3.4)
Employed private sector in 1996 and 1997 (2) 1 0.2337 (1.9) -0.3356 (2.6) 0.5235 (9.4)
Self-employed in 1996 and 1997 (3) 1 0.2217 (1.6) 0.5458 (7.2) 0.8392 (16.2)
Changed sector from 1996 to 1997 (4) 1 -0.2764 (1.8) -0.3045 (1.7) -0.6626 (14.0)
Other (5) 0 ref. ref. ref.
  Odds ratios
1 vs 5 1.729 0.421 5.189
2 vs 5 1.816 0.442 5.456
3 vs 5 1.795 1.068 7.482
4 vs 5 1.091 0.456 1.666
1. PROC LOGISTIC is modeling the probability that speedcls is 2. The name "speedcls" refers to the four categories of speed of closure shown in Chart 16.2.
2. PROC LOGISTIC is modeling the probability that speedcls is 3.
3. PROC LOGISTIC is modeling the probability that speedcls is 4.
4. Wald chi-squares are shown in parentheses. The standard errors underlying the chi-squares are only approximate; because their underlying estimates are not fully adjusted for the complexity of the survey design. The approximate adjustment involves standardizing the original survey weights so that their average is 1.0. Tests using more appropriate adjustment via bootstrap computations indicate that when the Wald chi-squares are 6.0 or greater, it can be considered that the parameter estimate is statistically significant at the 5% level or better, in the event that bootstrap standard errors were computed. When the Wald chi-squares are between 3.0 and 6.0, they can be considered to be statistically significant at a level between 5% and 15%.
Source: Survey of Labour and Income Dynamics, longitudinal file for Panel 2.
Table source: Statistics Canada, 2008, New Frontiers of Research on Retirement, catalogue number 75-511-XWE.