Coverage errors
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.
To produce good survey estimates, it is necessary that a survey sample adequately represent the survey population. To ensure proper coverage, SLID weights are adjusted using census population projections as control totals. The slippage rate is a measure of the percentage difference between these census projections and the survey estimate using weights prior to the application of this slippage related adjustment. More precisely, slippage is computed as
where Class C is the group or class for which we want to calculate slippage rates. For example at a detailed level the groups are based on province, sex and age group.
CPC is the census population projection for class C
wkc is the survey weight for kth responding unit in class C
SC is the set of responding sample households in class C
Slippage rates for household surveys are generally positive because of frame under coverage. Figure 4.1 shows slippage rates at the person level by panel. At lower geographic levels, the slippage rate varies more. Table 4.1 shows the person level slippage rates by province. We also computed slippage rates at the household level (Table 4.2 ). For household slippage rates for previous reference years, see Figure 4.2.
Figure 4.1 Person-level slippage rate (%)
Table 4.1 National and provincial Person-level slippage rates by panel for 2006
Figure 4.2 Household slippage rate (%)
Table 4.2 Household level slippage rates by province and household size (%)
- Date modified: