The effect of non-normality on estimating the variance of the combined ratio estimate in complex surveys

The performance of the balanced half-sample, jackknife and linearization methods for estimating the variance of the combined ratio estimate is studied by means of a computer simulation using artificially generated non-normally distributed populations.

The results of this investigation demonstrate that the variance estimates for the combined ratio estimate may be highly biased and unstable when the underlying distributions are non-normal. This is particularly true when the number of observations available from each stratum is small. The jack-     

A Diagnostic Test for Normality Within the Power Exponential Family

This article develops the locally uniformly most powerful unbiased Lagrange multiplier test of normality of regression disturbances within the family of power exponential distributions. The small sample power properties of the test are compared in a Monte Carlo study with 6 well-known tests across 12 alternative nonnormal distributions. In addition, the finite sample power properties for nonnormal alternatives within the power exponential family are summarized by estimating response surfaces. The results suggest that the proposed text is computationally convenient and possesses relatively attractive power properties even against alternatives outside the power exponential family.     

An Evaluation of Hispanic Population Estimates*

Objective. Estimates of the Hispanic population have traditionally been based on historical trends, ratios, or some variant of the cohort‐component method. In this article, we describe and test a methodology in which estimates of the Hispanic population are based on symptomatic indicators of population change such as births, deaths, and school enrollments. Methods. Using a variety of techniques, we develop Hispanic population estimates for counties in Florida. We evaluate the accuracy of those estimates by comparing them with 2000 Census counts. Results. Hispanic population estimates have larger errors than estimates of total population; errors vary considerably by population size and growth rate; some techniques perform better than others in places with particular population characteristics; and averages often perform better than individual techniques. Conclusions. In many circumstances, symptomatic data series can provide more accurate estimates of the Hispanic population than more commonly used techniques.     

The Couch Project

Carl Couch had a distinctive project. He was a man of distinctive character. In this commentary, I first describe the Couch project and then discuss how the project was interwoven with the character of its author. I then discuss “Oh, What Webs Those Phantoms Spin” in the context of Carl's larger project and close with a brief discussion of New Iowa School research after Carl Couch.     

Modeling Social Processes: Some Lessons from Sports

Sports events can provide helpful models for thinking about social processes. DiMaggio's hitting streak in baseball, for example, raises serious questions about how the causes of relatively uncommon social events are typically studied. The hitting streak also directs our attention to considering models that generate a distribution of outcomes in a population of social units. The series of games leading to a football championship is relevant for approaching the role of complex chains of events in leading to a given outcome (where each individual sequence itself has a low probability of occurring). A model of changing templates suggests a more realistic way of thinking about the linkages over time from an initial starting point to a given outcome. Another sports example, based on the application of a binomial model to the outcome of a series of games, suggests that we should think of chance as a dependent variable that is influenced by—among other factors—the nature of the social structure. Also examined are the special problems in explaining any specific outcome.     

Local Control Strategy: Simple Analyses of Air Pollution Data Can Reveal Heterogeneity in Longevity Outcomes

Claims from observational studies that use traditional model specification searches often fail to replicate, partially because the available data tend to be biased. There is an urgent need for an alternative statistical analysis strategy, that is not only simple and easily understood but also is more likely to give reliable insights when the available data have not been designed and balanced. The alternative strategy known as local control first generates local, nonparametric effect‐size estimates (fair treatment comparisons) and only then asks whether the observed variation in these local estimates can be predicted from potential confounding factors. Here, we illustrate application of local control to a historical air pollution data set describing a “natural experiment” initiated by the federal Clean Air Act Amendments of 1970. Our reanalysis reveals subgroup heterogeneity in the effects of air quality regulation on elderly longevity (one size does not fit all), and we show that this heterogeneity is largely explained by socioeconomic and environmental confounders other than air quality.     

The influence of anthropogenic resources on multi-scale habitat selection by raccoons

With the continuing spread of urban areas, gaining a greater understanding of the effect of human presence on wildlife species is essential for wildlife managers. We determined the influence of anthropogenic resources on home range size and habitat selection of raccoons (Procyon lotor) during summer (June–August) 1996–2000 for 120 raccoons at three sites exposed to varying levels of urbanization and anthropogenic resources, specifically food. Home range estimates were larger (P < 0.05) at the rural site than the suburban and urban sites for both genders. We used compositional analysis to examine raccoon habitat selection at the second-order home range, second-order core area, and third-order home range scales. Woodland was consistently a highly-selected habitat type for both sexes at every spatial scale. Relative to other habitat types, habitat associated with human-related food (human use areas) was selected most often at the urban site, intermediately at the suburban site, and not selected at the rural site. Spatial scale also affected habitat selection. Human use areas were preferentially selected at the second- and third-order level at the urban site, third-order level only at the suburban site, and at neither level at the rural site. Additionally, intersexual differences in habitat selection were reduced at the urban site, with both sexes preferentially selecting for human use areas as well as woodland habitat. Smaller home ranges in urbanized environments are often attributed to the abundant and concentrated anthropogenic resources associated with human activity, but with little empirical support. Our habitat selection analyses followed our predictions that raccoon foraging is strongly influenced by the artificial distribution and abundance of human-related food. Male and female raccoons in urban areas reduce their foraging patterns and focus their foraging activity on anthropogenic foods.