A simple procedure for testing linear hypotheses about the parameters of a nonlinear model using weighted least squares

Suppose the same nonlinear function involving k parameters is fit to each of t populations. Suppose further it is of interest to compare a specific parameter of the models across the populations. Such comparisons can be expressed as linear hypotheses about the parameters of the nonlinear models. A weighted linear least squares (WLLS) procedure is proposed to test these linear hypotheses. The advantages and disadvantages of the WLLS procedure are discussed. This procedure is also compared to a nonlinear least squares procedure for testing these hypotheses in nonlinear models.     

Gaze Cueing Effect in a Face-to-Face Situation

People are faster to detect a target looked at, rather than ignored or not cued, by another person. This gaze cueing effect (GCE) reflects joint attention processes which play a crucial role in social interactions. To date, GCE has only been measured with on-screen experiments. This only approximates the real situation of joint attention which involves a triadic interaction between oneself, another person, and an external object. Here, we present an innovative experimental setup to investigate GCE in a face-to-face situation. We show a robust GCE, which is of the same size range as in on-screen experiments. This study fosters an ecological approach of joint attention.     

Part I––Comparing Noncancer Chronic Human Health Reference Values: An Analysis of Science Policy Choices

The goal of this study was to systematically evaluate the choices made in deriving a chronic oral noncancer human health reference value (HHRV) for a given chemical by different organizations, specifically those from the U.S. Environmental Protection Agency, Health Canada, RIVM (the Netherlands), and the U.S. Agency for Toxic Substances and Disease Registry. This analysis presents a methodological approach for comparing both the HHRVs and the specific choices made in the process of deriving an HHRV across these organizations. Overall, across the 96 unique chemicals and 171 two‐way organizational comparisons, the HHRV agreed approximately 26% of the time. A qualitative method for identifying the primary factors influencing these HHRV differences was also developed, using arrays of HHRVs across organizations for the same chemical. The primary factors identified were disagreement on the critical or principal study and differential application of the total uncertainty factor across organizations. Of the cases where the total UF was the primary factor influencing HHRV disagreement, the database UF had the greatest influence.     

Part II: Quantitative Evaluation of Choices Used in Setting Noncancer Chronic Human Health Reference Values Across Organizations

Environmental and public health organizations, including the World Health Organization (WHO) and the U.S. Environmental Protection Agency (USEPA), develop human health reference values (HHRV) that set “safe” levels of exposure to noncarcinogens. Here, we systematically analyze chronic HHRVs from four organizations: USEPA, Health Canada, RIVM (the Netherlands), and the U.S. Agency for Toxic Substances and Disease Registry. This study is an extension of our earlier work and both closely examines the choices made in setting HHRVs and presents a quantitative method for identifying the primary factors influencing HHRV agreement or disagreement.⁽¹⁾ We evaluated 171 organizational comparisons, developing a quantitative method for identifying the factors to which HHRV agreement (that is, when both organizations considering the same data set the identical HHRV values) is most sensitive. To conduct this analysis, a Bayesian belief network was built using expert judgment, including the specific science policy choices analysis made in the context of setting an HHRV. Based on a sensitivity of findings analysis, HHRV agreement is most sensitive to the point of departure value, followed by the total uncertainty factor (UF), critical study, critical effect, animal model, and point of departure approach. This analysis also considered the specific impacts of individual UFs, with the database UF and the subchronic‐to‐chronic UF being identified as primary factors impacting the total UF differences observed across organizations. The sensitivity of findings analysis results were strengthened and confirmed by frequency analyses evaluating which choices most often disagreed when the HHRV and the total UF disagreed.     

Parameter estimation in dynamic models

In this paper we study the parameter estimation of a first-order dynamic model for intervention and transfer function analysis, A new parameterization is proposed to avoid the “overshoot” problem in nonlinear estimation in a frequently used parameterization (Box and Tiao, 1975).     

Voluntary and involuntary driving cessation in later life

This study explores the decision-making process of driving cessation in later life, with a focus on voluntariness. The sample included 83 former drivers from the Baltimore Epidemiologic Catchment Area Study. A majority of participants (83%) reportedly stopped driving by their own decision. However, many voluntary driving retirees reported external factors such as financial difficulty, anxiety about driving, or lack of access to a car as main reasons for driving cessation. These findings imply that distinction between voluntary and involuntary driving cessation is ambiguous and that factors beyond health status, including financial strain, play a role in the transition to non-driving.     

European Public Perceptions of Food Risk: Cross-National and Methodological Comparisons

This article presents a comparative study of public perceptions of food risk across 25 European member states. A secondary data analysis is conducted on a Eurobarometer survey fielded to nationally representative samples in 2005. The survey included closed questions as well as free associations to map risk perceptions. Taking a quantitative approach, we find that people in a majority of European countries express similar levels of concern about food risks. However, outside this majority a North-South divide is evident, with the Northern countries worrying less than the Southern countries. Multilevel modeling shows that cross-national differences in individual respondents' level of worry are in part attributable to shared country effects and to generalized risk sensitivity about a range of personal risks. On the underlying structure of food risk concerns, factor analysis points to three dimensions described by groupings of risks related to adulteration and contamination, health effects, and production and hygiene. A qualitative analysis of respondents' free associations about problems and risks with food identifies three major themes that are consistent with the quantitative results. However, the free associations also point toward greater cross-national diversity and to striking variations in the range and importance of food risks. Overall, the picture is of a public that frames food risks in a wider context of beliefs about the links between diet and health. We conclude with some implications for research on food risk perceptions in particular and risk perception studies in general.     

A modification of the Fisher-Cornish approximation for the student t percentiles

An asymptotic expansion of the Student t distribution is derived by expanding the standardized Student t distribution in terms of the normal distribution. This expansion is inverted to obtain corresponding asymptotic expansions for the Student t percentiles as functions of the standard normal percentiles0 Using the first two, three or four terms of these expansions, we get approximations of the Student t percentiles which are generally more accurate than the approximations given by Fisher and Cornish(1960) and Koehler (1983).An approximation of the distribution function obtained from this expansion is compared with the approximations discussed by Ling (1978) andfound to be more accurate for moderate degrees of freedom.