Mixture models in measurement error problems, with reference to epidemiological studies

Summary. The paper focuses on a Bayesian treatment of measurement error problems and on the question of the specification of the prior distribution of the unknown covariates. It presents a flexible semiparametric model for this distribution based on a mixture of normal distributions with an unknown number of components. Implementation of this prior model as part of a full Bayesian analysis of measurement error problems is described in classical set-ups that are encountered in epidemiological studies: logistic regression between unknown covariates and outcome, with a normal or log-normal error model and a validation group. The feasibility of this combined model is tested and its performance is demonstrated in a simulation study that includes an assessment of the influence of misspecification of the prior distribution of the unknown covariates and a comparison with the semiparametric maximum likelihood method of Roeder, Carroll and Lindsay. Finally, the methodology is illustrated on a data set on coronary heart disease and cholesterol levels in blood.     

Levelling across the life‐span?: Tracing the face vowel in panel data from the North East of England

Apparent time analysis has revealed that the Tyneside face vowel is the site of two intersecting trends: levelling towards the supra‐northern monophthong as well as the gradual incursion of the southern standard closing diphthong. This article investigates the participation of individual speakers across their life‐span in these ongoing changes in the face vowel. We report on a small‐scale panel sample of six speakers who were recorded in 1971 and again, 42 years later, in 2013. The analysis probes the stability of individual speakers’ grammars, relying on longitudinal ethnographic analysis in the community as well as insights gleaned from sociolinguistic interviews about the speakers’ socio‐demographic trajectory and their presentations of self. The article contributes to the growing body of panel research that aims to determine the scope and the limits of linguistic malleability across speakers’ life histories.     

Mental Imagery and Production of Hand Gestures While Speaking in Younger and Older Adults

Speech and conversational hand gestures were recorded during interviews of 23 younger (M = 21 years) and 19 older adults (M = 70 years). Three kinds of questions were used in order to activate either visual images, motor images, or no mental image (abstract topics). On average, the rate of gesture production did not differ in younger and older adults, but it was significantly influenced by imagery conditions. Gesture production was higher in the motor than in the visual imagery condition, and lowest in the abstract condition. A significant interaction between age and imagery conditions influenced the proportion of representational gestures, which were relatively less frequent in older adults, especially in the visual imagery condition. Content analysis of verbal responses showed that imagery values did not differ in younger and older adults, but that concrete words were less frequent in responses to abstract questions than in the two other conditions. The implications of these results concerning the mechanisms of gesture production and the age-related changes in conversational behavior are discussed.     

Engagement philosoviétique et posture sociologique dans l’entre-deux-guerres : le rôle politico-intellectuel de Georges Friedmann

This in-depth biographical study of Georges Friedmann's itinerary during the period between the two World wars draws from private records and the Communist movement's archives. It highlights this sociologist's pivotal role between the Communist world and academic circles in France. In the 1930s, Friedmann helped to legitimate Marxism in the French social sciences, even among Durkheimians who had proven resistant to Karl Marx's and Friedrich Engels's ideas since the end of the 19th century. Friedmann's political and intellectual stance would, however, become untenable in the late 1930s given the closing of the Communist world.     

Changes in Objective and Self-Reported Measures of Physical Capacity After an Intervention in Obese Older Women

The aim of this study was to determine if objective and self-reported measures of physical capacity are two equivalent methods to detect changes following an intervention in obese older women. 36 obese women aged between 55 and 75 years participated in a 3-month study with the aim of improving physical capacity by caloric restriction and/or resistance training. Physical capacity was measured objectively with 10 different tests and self-reported with the SF-36 physical functioning score (SF-36 PF score). Then the performance-to-objective tests were computed using quartiles to provide a baseline global physical capacity score. The mean percentage of change of the 10 tests as well as the SF-36 PF score were also calculated after the study. Body composition was measured by dual-energy X-ray (DXA) absorptiometry. The baseline global physical capacity score and the SF-36 PF score were significantly correlated at baseline (r = 0.43; P < 0.01). Eight out of the 10 objective tests of physical capacity improved after the intervention, while no improvement was observed for the SF-36 PF score. On average, percentage of change in physical capacity was 4.1 ± 5.9% for the SF-36 PF score and 11.1 ± 2.9% for the objective measures. However, no significant correlation was observed between percentage of changes between the two approaches after the intervention (r = 0.32; P = 0.07). Based on these results, the method used to quantify physical capacity after an intervention may have major implications on the outcomes. The methods used should be carefully analyzed in regard to the objective of the study.     

Students and the Consumer Credit Market: Towards a Social Policy Agenda

The rise in access to complex consumer credit arrangements has taken place against a backdrop of a call for increased individual responsibility. Consumers are required to behave in a way which recognizes both their rights and responsibilities. But how much responsibility should they be expected to shoulder in critical areas of complex choice? Students represent a particular group of novice, sometimes vulnerable and often targeted consumers, who may display limited financial capability and responsibility. In addition to arguments for a more nuanced understanding of individual responsibility in different environments, the role of commercial agents and their marketing practices, which can have major implications for social policy, should also be considered. From the perspective of both students and relevant agencies and organizations, this article examines the nature, role and limitations of individual responsibility in managing credit and debt. While the social and economic system confers rights on lenders, their responsibility in marketing remains limited. It is argued that a framework for more responsible marketing of credit is a critical element of social policy acting as a bridge between individual responsibility and regulation. Qualitative research is used to illustrate the argument for marketing's social responsibility from the point of view of students entering credit arrangements to meet short‐term needs but with long‐term implications.