MULTIVARIATE DECOMPOSITION FOR HAZARD RATE MODELS

We develop a regression decomposition technique for hazard rate models, where the difference in observed rates is decomposed into components attributable to group differences in characteristics and group differences in effects. The baseline hazard is specified using a piecewise constant exponential model, which leads to convenient estimation based on a Poisson regression model fit to person-period, or split-episode data. This specification allows for a flexible representation of the baseline hazard and provides a straightforward way to introduce time-varying covariates and time-varying effects. We provide computational details underlying the method and apply the technique to the decomposition of the black-white difference in first premarital birth rates into components reflecting characteristics and effect contributions of several predictors, as well as the effect contribution attributable to race differences in the baseline hazard .     

Immigrant fertility patterns and differentials in Australia, 1971–1976

In this paper the fertility patterns and differentials among various immigrant groups in Australia are examined. Official vital statistics for the period 1971–76 are used.

Fertility ratios standardized by age and marital status suggest that the overall fertility of foreign-born women was higher in both 1971 and 1976; however, some evidence -of convergence towards an ‘Australian’ norm was found.

Four distinct patterns of fertility were noted. In two of them, Arab and South European, marital fertility was substantially higher but non-marital fertility quite low. The North-West European pattern was closest to that of the native-born; however, in the East European pattern fertility was lowest. Component analysis showed that most of the differences between the total fertility rate of Australians and those of the other groups reflect the significantly higher marital fertility rates and proportions married among the foreign-born groups.     

Morality and Nuclear Energy: Perceptions of Risks and Benefits,Personal Norms,and Willingness to Take Action Related to Nuclear Energy

We examined factors underlying people's willingness to take action in favor of or against nuclear energy from a moral perspective. We conducted a questionnaire study among a sample of the Dutch population (N = 123). As expected, perceptions of risks and benefits were related to personal norms (PN), that is, feelings of moral obligation toward taking action in favor of or against nuclear energy. In turn, PN predicted willingness to take action. Furthermore, PN mediated the relationships between perceptions of risk and benefits and willingness to take action. In line with our hypothesis, beliefs about the risks and benefits of nuclear energy were less powerful in explaining PN for supporters compared to PN of opponents. Also, beliefs on risks and benefits and PN explained significantly more variance in willingness to take action of opponents than of supporters. Our results suggest that a moral framework is useful to explain willingness to take action in favor of and against nuclear energy, and that people are more likely to protest in favor of or against nuclear energy when PN are strong.     

Research on stalking: what do we know and where do we go?

Findings of the article in this two-volume series on stalking are reviewed. Building on the findings of the National Violence Against Women Survey (Tjaden & Thoennes, 1998), this series of studies adds to the literature in defining and measuring stalking behaviors. Repeated stalking victimization is reported by up to 62% of young adults, although frequencies depend on the sample and the precise definition used. Self-definitions of stalking victimization may be quite different from legal definitions. Although the majority of legally defined stalkers (where victim fear is a key component) are men stalking women, studies utilizing other definitions find many more women as stalkers, with no significant gender differences in many studies. In many cases, it does appear that stalking is one part of a larger pattern of relationship physical and psychological abuse. Data on emotional reactions and coping strategies of victims are also reviewed, along with findings on characteristics of stalkers.     

Adoption Issues: Preparation of Psychologists and an Evaluation of the Need for Continuing Education

This study is an investigation of the level of preparation psychologists have in dealing with adoption issues and an assessment of the need for further education in treating triad members. A questionnaire was mailed to 497 psychologists randomly selected from the National Register of Health Service Providers in Psychology who received doctoral degrees between the years of 1990 and 1996. Two hundred and ten psychologists, all of whom were currently treating clients, participated in the study. Psychologists completed a Likert rating of their sense of preparation to treat triad members. The largest group (51%) rated themselves as Somewhat prepared, with the second largest group (23%) rating themselves as Not very prepared. Ninety percent reported they needed more education in adoption, and 81% reported interest in taking a continuing education course in adoption. Of the 210 participants, only 67 reported taking courses that dealt with adoption as part of their formal education. Of those psychologists who reported taking courses that involved adoption, they averaged only 1.3 courses during their undergraduate education, and 1.5 courses during graduate education, while the remaining 143 psychologists received no education in this area. When one considers the proportion of triad members seen clinically, this study suggests psychologists need more education and background information about the effects of the adoption process.     

A methodological strategy for a one-number census in the UK

As a result of lessons learnt from the 1991 census, a research programme was set up to seek improvements in census methodology. Underenumeration has been placed top of the agenda in this programme, and every effort is being made to achieve as high a coverage as possible in the 2001 census. In recognition, however, that 100% coverage will never be achieved, the one-number census (ONC) project was established to measure the degree of underenumeration in the 2001 census and, if possible, to adjust fully the outputs from the census for that undercount. A key component of this adjustment process is a census coverage survey (CCS). This paper presents an overview of the ONC project, focusing on the design and analysis methodology for the CCS. It also presents results that allow the reader to evaluate the robustness of this methodology.     

Immigrant incorporation in the garment industry of Los Angeles

This study expands immigrant social network theory and examined employment patterns in the garment industry in Los Angeles, California, among Latino workers employed by Asian immigrant entrepreneurs. The study determined that a large percentage of immigrant employees found their jobs through the immigrant economy. Entrepreneurship increased the supply of local jobs and expanded the economy at destination at no expense to natives. Immigrant entrepreneurs bought firms from nonimmigrant owners or started new ones with an immigrant labor supply. Massey's index is flawed due to its exclusion of the role of entrepreneurs. Migration networks facilitate entrepreneurship, but some ethnic groups have fewer entrepreneurs, such as Mexicans and Central Americans. A 1993 Los Angeles survey identified 3642 garment factories in its county. Mean employment was 27.1 persons. The garment industry was the 4th largest industry in the area in 1996, with 98,700 employees. It represented 6% of all wage and salary employees in the City and 5.5% of the immigrant labor force in the County in 1990. 93% of garment workers in 1990 were immigrants. It is estimated that 51% of garment factory owners were Asians; most employees were Latinos. Census figures on sewing machine operators indicated 47.3% of owners were Whites and 42.45 were Asians. 53.3% of employees were other ethnic groups, 14.5% were Asians, and 32.2% were Whites. It is estimated that 47.2% of total employment was due to the immigration economy. 71.5% of the total employment in the garment industry was in the immigrant sector.     

Residential Building Codes,Affordability, and Health Protection: A Risk-Tradeoff Approach

Residential building codes intended to promote health and safety may produce unintended countervailing risks by adding to the cost of construction. Higher construction costs increase the price of new homes and may increase health and safety risks through income and stock effects. The income effect arises because households that purchase a new home have less income remaining for spending on other goods that contribute to health and safety. The stock effect arises because suppression of new-home construction leads to slower replacement of less safe housing units. These countervailing risks are not presently considered in code debates. We demonstrate the feasibility of estimating the approximate magnitude of countervailing risks by combining the income effect with three relatively well understood and significant home-health risks. We estimate that a code change that increases the nationwide cost of constructing and maintaining homes by $150 (0.1% of the average cost to build a single-family home) would induce offsetting risks yielding between 2 and 60 premature fatalities or, including morbidity effects, between 20 and 800 lost quality-adjusted life years (both discounted at 3%) each year the code provision remains in effect. To provide a net health benefit, the code change would need to reduce risk by at least this amount. Future research should refine these estimates, incorporate quantitative uncertainty analysis, and apply a full risk-tradeoff approach to real-world case studies of proposed code changes.