Latent class analysis has been used to model measurement error, to identify flawed survey questions and to estimate mode effects. Using data from a survey of University of Maryland alumni together with alumni records, we evaluate this technique to determine its usefulness for detecting bad questions in the survey context. Two sets of latent class analysis models are applied in this evaluation: latent class models with three indicators and latent class models with two indicators under different assumptions about prevalence and error rates. Our results indicated that the latent class analysis approach produced good qualitative results for the latent class models-the item that the model deemed the worst was the worst according to the true scores. However, the approach yielded weaker quantitative estimates of the error rates for a given item.
A detailed investigation was conducted to understand the contamination characteristics of a selected set of potentially toxic metals in Shanghai. The amount of Pb, Zn, Cu, Cr, Cd and Ni were determined from 273 soil/dust samples collected within urban area. The results indicated that concentration of all metals except Ni in soils was significant, and metal pollution was even severer in roadside dust. A series of metal spatial distribution maps were created through geostatistical analysis, and the pollution hotspots tended to associate with city core area, major road junctions, and the regions close to industrial zones. In attempt of identifying the source of metals through geostatistical and multivariate statistical analyses, it was concluded as follows: Pb, Zn and Cu mainly originated from traffic contaminants; soil Ni was associated with natural concentration; Cd largely came from point-sourced industrial pollution; and Cr, Ni in dust were mainly related to atmospheric deposition. Human activities have led to high accumulation of potentially toxic metals in urban soils and roadside dust of Shanghai.