Variance estimation in complex survey sampling for generalized linear models

Summary.  Complex survey sampling is often used to sample a fraction of a large finite population. In general, the survey is conducted so that each unit (e.g. subject) in the sample has a different probability of being selected into the sample. For generalizability of the sample to the population, both the design and the probability of being selected into the sample must be incorporated in the analysis. In this paper we focus on non-standard regression models for complex survey data. In our motivating example, which is based on data from the Medical Expenditure Panel Survey, the outcome variable is the subject's 'total health care expenditures in the year 2002'. Previous analyses of medical cost data suggest that the variance is approximately equal to the mean raised to the power of 1.5, which is a non-standard variance function. Currently, the regression parameters for this model cannot be easily estimated in standard statistical software packages. We propose a simple two-step method to obtain consistent regression parameter and variance estimates; the method proposed can be implemented within any standard sample survey package. The approach is applicable to complex sample surveys with any number of stages.     

Reputational Capital and Academic Pay

This article examines how research productivity, administrative service, and teaching affect reputational capital in the market for academic economists. Also, we investigate the issue of the durability of reputational capital, estimating the penalties associated with gaps in research output. Our results reveal that (1) the market makes a distinction between the quantity and the quality of an individual's research output, (2) a smaller number of frequently cited papers enhances reputational capital more than a greater number of less frequently cited papers, (3) reputational capital does not depreciate rapidly during research breaks, and (4) the market does not discount co-authored papers.     

Understanding Interest in the Second Year of Life

Infants (n = 24, mean age 13 months and n = 24, mean age 19 months) were tested on an extension of the method introduced by Tomasello and Haberl (2003) to examine the understanding of another person’s interest in a novel object. Four objects were presented serially. For two objects, infants played with an experimenter. The infant played with one object alone, and the experimenter played with one object alone. Finally, all four objects were presented together, and the experimenter excitedly asked for one without indicating which. Results showed that younger infants tended to chose the object that they had not yet played with, whereas older infants were significantly more likely to choose the object that the experimenter had not yet played with. These results are discussed in the context of research on the development of understanding diversity of simple object‐directed attitudes in the second year of life.     

Metabolic rift or metabolic shift? dialectics,nature, and the world-historical method

In the flowering of Red-Green Thought over the past two decades, metabolic rift thinking is surely one of its most colorful varieties. The metabolic rift has captured the imagination of critical environmental scholars, becoming a shorthand for capitalism’s troubled relations in the web of life. This article pursues an entwined critique and reconstruction: of metabolic rift thinking and the possibilities for a post-Cartesian perspective on historical change, the world-ecology conversation. Far from dismissing metabolic rift thinking, my intention is to affirm its dialectical core. At stake is not merely the mode of explanation within environmental sociology. The impasse of metabolic rift thinking is suggestive of wider problems across the environmental social sciences, now confronted by a double challenge. One of course is the widespread—and reasonable—sense of urgency to evolve modes of thought appropriate to an era of deepening biospheric instability. The second is the widely recognized—but inadequately internalized—understanding that humans are part of nature.     

Modified goodness-of-fit test for the laplace distribution

In this paper critical values of the Anderson-Darling and Cra[macute]er-von Mises are given for the Laplace distribution for maximum likelihood estimation of the location and scale parameters. The power of these tests is investigated for a number of alternative distributions. The Anderson-Darling test yields higher power generally than the Cra[macute]er-von Mises test.     

Examining the relationship between gender and drug-using behaviors in adolescents: the use of diagnostic assessments and biochemical analyses of urine samples

The present study examined the relationship between gender and drug-use among adolescents using diagnostic assessments and biochemical analyses of urine samples. The data were collected in the context of a referral and assessment program for adolescents suspected of using drugs, using the Adolescent Substance Battery [1]. A six-month random sample of 20 percent of adolescents assessed for drug use were targeted for biochemical assay. Compliance for urine delivery was relatively high at 91 percent. Urine samples were quantitatively screened for pharmaceuticals using a standard thin-layer chromatography (TLC) technique [2]. Statistical significance was found in the relationship between gender and marijuana use (p < .05). This study confirms that gender and drug-using behaviors among adolescents need additional research and evaluation.     

Mother–Grandmother and Mother–Father Coparenting Across Time Among Mexican‐Origin Adolescent Mothers and Their Families

The current study examined mothers' reports of coparenting relationship dynamics (i.e., conflict, communication) within and between mother–father and mother–grandmother subsystems from 10 months postpartum to 5 years postpartum among 178 Mexican‐origin teen mothers (M age = 16.78 years; SD = 1.00). Within subsystems, more frequent mother–father coparenting conflict was associated with less frequent mother–father coparenting communication from 10 months to 5 years postpartum, and more frequent mother–father coparenting communication was associated with less frequent mother–father conflict from 3 to 4 years postpartum. Furthermore, more frequent mother–grandmother coparenting communication was associated with less frequent mother–grandmother conflict from 10 months to 2 years postpartum. Regarding relations across subsystems, more frequent mother–father coparenting conflict was associated with more frequent mother–grandmother conflict from 10 months to 2 years postpartum, as well as from 3 to 4 years postpartum. Findings have implications for future interventions focused on coparenting relationships within the context of adolescent parenthood.     

The journalist as the audience: Evaluating the U.S. Navy’s “Sailor for a Day” campaign

This study demonstrates how focusing on the journalist as the audience of a media relations campaign can result in significant positive media coverage. Specifically, we examined the U.S. Navy’s “Sailor for a Day” campaign that invited media to experience a day in the life of a U.S. Navy sailor aboard an aircraft carrier. We propose that the “Sailor for a Day” campaign successfully influenced journalist perceptions and media coverage because 1) the pitch tied to programming needs; 2) uncommon access was granted; and 3) the journalist experience was a prominent feature of the media pitch.