workplace: (Modified) Canadian Workplace and Employee Survey

In robsurvey: Robust Survey Statistics Estimation

(Modified) Canadian Workplace and Employee Survey

Description

The workplace data are from Fuller (2009, pp. 366–367).

Usage

data(workplace)

Format

A data.frame with a sample of 142 workplaces on the following variables

ID

identifier variable [integer].

weight

sampling weight [double].

employment

employment total [double].

payroll

payroll total (1000 dollars)[double].

strat

stratum identifier[integer].

fpc

finite population correction [integer].

Details

The workplace data represent a sample of workplaces in the retail sector in a Canadian province. The data are not those collected by Statistics Canada, but have been generated by Fuller (2009, Example 3.1.1) to display similar characteristics to the original 1999 Canadian Workplace and Employee Survey (WES).

Sampling design of the 1999 WES

The WES target population is defined as all workplaces operating in Canada with paid employees. The sampling frame is stratified by industry, geographic region, and size (size is defined using estimated employment). A sample of workplaces has been drawn independently in each stratum using simple random sample without replacement (the stratum-specific sample sizes are determined by Neyman allocation). Several strata containing very large workplaces were sampled exhaustively; see Patak et al (1998). The original sampling weights were adjusted for nonresponse.

Remarks by Fuller (2009, p. 365)

The original weights of WES were about 2200 for the stratum of small workplaces, about 750 for medium-sized, and about 35 for large workspaces.

Source

The data workplace is from Table 6.3 in Fuller (2009, pp. 366–367).

References

Fuller, W. A. (2009). Sampling Statistics, Hoboken (NJ): John Wiley and Sons. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1002/9780470523551")}

Patak, Z., Hidiroglou, M. and Lavallée, P. (1998). The methodology of the Workplace and Employee Survey. Proceedings of the Survey Research Methods Section, American Statistical Association, 83–91.

Examples

head(workplace)

library("survey")
# Survey design for stratified simple random sampling without replacement
dn <- if (packageVersion("survey") >= "4.2") {
        # survey design with pre-calibrated weights
        svydesign(ids = ~ID, strata = ~strat, fpc = ~fpc, weights = ~weight,
                  data = workplace, calibrate.formula = ~-1 + strat)
    } else {
        # legacy mode
        svydesign(ids = ~ID, strata = ~strat, fpc = ~fpc, weights = ~weight,
                  data = workplace)
    }

robsurvey documentation built on Sept. 11, 2024, 6:35 p.m.