Estimating true changes when categorical panel data are affected by uncorrelated and correlated classification errors - An application to unemployment data
F. Bassi et al., Estimating true changes when categorical panel data are affected by uncorrelated and correlated classification errors - An application to unemployment data, SOCIOL METH, 29(2), 2000, pp. 230-268
Conclusions about changes in categorical characteristics based on observed
panel data can be incorrect when (even a small amount of) measurement error
is present. Random measurement errors, referred to as independent classifi
cation errors, usually lead to over-estimation of the total amount of gross
change, whereas systematic, correlated errors usually cause underestimatio
n of the transitions. Furthermore, the patterns of true change may be serio
usly distorted by independent or systematic classification errors. Latent c
lass models and directed log-linear analysis are excellent tools to correct
for both independent and correlated measurement errors. An extensive examp
le on labor market stares taken from the Survey of Income and Program Parti
cipation panel is presented.