John G. Cragg

Making Good Inferences from Bad Data

Errors in variables can seriously distort inference when they are not taken into account explicitly. Coefficient values, their significance, and whether some explanatory variables should instead be used as instruments are largely a matter of interpretation unless further information is available. Higher moments of the observable variables impose restrictions that allow testing for identification and specification and estimating the parameters of the standard errors-in-variables model. The argument is developed partly through examples illustrating the points. Errors in variables can seriously distort inference when they are not taken into account explicitly. Coefficient values, their significance, and whether some explanatory variables should instead be used as instruments are largely a matter of interpretation unless further information is available. Higher moments of the observable variables impose restrictions that allow testing for identification and specification and estimating the parameters of the standard errors-in-variables model. The argument is developed partly through examples illustrating the points.

Those Returning to Income Assistance

This paper presents descriptive information about returnees to Income Assistance in British Columbia (IA). IA returnees are defined as those who had a spell on IA over the previous two years, dropped off IA for one or more months (and hence were at risk of returning), and who subsequently returned to IA. Factual information on the IA returnee population is a potentially important input into policy discussions of how the self sufficiency prospects of those on welfare might be improved, leading to desired reductions in welfare caseloads and expenditures. There has been little previous research on IA returnees in Canada. More is known about repeat recipients of Unemployment Insurance (UI). (See Corak 1993.) Our data are described in section II. Section III provides information about the returnee population, including comparisons with the population of those leaving IA. Our evidence suggests that it is a mistake to focus IA policy reforms mostly on single women with children. Important overlaps between the IA and UI programs are also documented. In section III we present estimates of the probability of return to IA for various groups. (This paper draws on a 1993 preliminary study by Bruce, Bailey, Cragg, Nakamura and Warburton.)