Latent Class Analysis of Complex Sample Survey Data

Abstract

High fruit and vegetable intake is associated with decreased cancer risk. However, dietary recall data from national surveys suggest that, on any given day, intake falls below the recommended minima of three daily servings of vegetables and two daily servings of fruit. There is no single widely accepted measure of "usual" intake. One approach is to regard the distribution of intake as a mixture of "regular" (relatively frequent) and "nonregular" (relatively infrequent) consumers, using an indicator of whether an individual consumed the food of interest on the recall day. We use a new approach to summarizing dietary data, latent class analysis (LCA), to estimate "usual" intake of vegetables. The data consist of four 24-hour dietary recalls from the 1985 Continuing Survey of Intakes by Individuals collected from 1,028 women. Traditional LCA based on simple random sampling was extended to complex survey data by introducing sample weights into the latent class estimation algorithm and by accounting for the complex sample design through the use of jackknife standard errors. A two-class model showed that 18% do not regularly consume vegetables, compared to an unweighted estimate of 33%. Simulations showed that ignoring sample weights resulted in biased parameter estimates and that jackknife variances were slightly conservative but provided satisfactory con dence interval coverage. Using a survey-wide estimate of the design effect for variance estimation is not accurate for LCA. The methods proposed in this article are readily implemented for the analysis of complex sample survey data.

Culture and Public Goods: The Case of Religion and the Voluntary Provision of Environmental Quality

“…Although latent class analysis (LCA) has been applied to several social issues (see, for example, Patterson et al 2002;and Biemer and Wiesen, 2002), it is still a fairly novel methodology in the economics literature (see Boxall and Adamowicz, 2002;Greene and Hensher, 2003;Provencher and Bishop, 2004;Clark, Etile, Postel-Vinay, Senik and Van der Straeten, 2005;Scarpa and Thiene, 2005;Morey, Thacher, and Breffle, 2006). Both on theoretical and practical grounds, however, latent class analysis is a promising approach to the study of social and cultural capital.…”

Use of Dietary Indexes among Children in Developed Countries

“…Those methods may include: 1) applying specific weights to each component scale of the index on the basis of relative importance (24,25,27); 2) grouping index components according to the domain of the dietary behavior they assess (e.g. diet composition, beliefs, behaviors) (106); 3) conducting proper statistical tests to evaluate the discriminating characteristics and abilities of the index, such as sensitivity, specificity, c-statistic values, positive/negative predictive values, and others; and 4) performing additional tests beyond multivariate regression analyses to reveal any potential heterogeneity in index performance (106), including latent class analysis (106,121), structural equation modeling (122), and data mining (123).…”

Fitting Log-Linear Models to Contingency Tables From Surveys With Complex Sampling Designs: An Investigation of the Clogg-Eliason Approach

“…Leaving aside consideration of the standard errors, there seems no strong reason to prefer one set of estimates to the other. One possible argument in favour of the PML estimator, following Patterson et al (2002) and mentioned in section 6.2, is that the PML estimator is 'estimating' a well-defined population quantity if the model is false, whereas the CE estimator is then estimating a quantity dependent on the arbitrariness of the sampling scheme.…”