Consistent selection of the actual model in regression analysis

SUMMARY In regression analysis, a best subset of regressors is usually selected by minimizing Mallows's C statistic or some other equivalent criterion, such as the Akaike lambda information criterion or cross-validation. It is known that the resulting procedure suffers from a lack of consistency that can lead to a model with too many variables. For this reason, corrections have been proposed that yield consistent procedures. The object of this paper is to show that these corrected criteria, although asymptotically consistent, are usually too conservative for finite sample sizes. The paper also proposes a new correction of Mallows's statistic that yields better results. A simulation study is conducted that shows that the proposed criterion performs well in a variety of situations.     

Estimating the concentration of the Langevin distribution

We propose an estimator of the concentration parameter of the Langevin distribution that is based on the spatial median. We derive its asymptotic distribution, compute its asymptotic efficiency, and show that it possesses some interesting stability properties under contamination.     

Traditional society, political society and Québec sociology: 1945–1980

L'analyse de la production sociologique au Québec, de 1945 à 1980, ne saurait être entreprise sans faire référence à la transformation des conditions generates d'enonciation qu'a connue la société québécoise durant la période. On peut schématiquement regrouper cette sociographie en quatre courants distincts. Le premier, de 1945 à 1960, se donne comme un savoir critique dc l'adaptation illustrant l'inadéquation de la référence au traditionalisme. Le début des années soixante marque le passage à la dominance d'une forme de la représentation présentant désormais le Québec comme une société politique. Une sociologie fonctionnelle de l'adaptation se développe dés lors, liée à la mise en place de L'État keynésien. Les années soixante-dix voient au contraire l'affirmation d'une sociologie de l'émancipation. Divisée en deux courants antagonistes, culturaliste et politiste, cette sociologie travaille à la redéfinition d'une société québécoise dorénavant soumise aux pleines déterminations de la modernité. The modernity of sociological production in Québec from 1945 to 1980 cannot be attempted without referring to the transformation in the general conditions of Québec's expression of itself as a society during this period. We can regroup this sociography schematically into four distinct approaches: the first, from 1945 to 1960, presented itself as a form of critical knowledge of adaptation which illustrated the inadequacy of references to traditionalism. The beginning of the 1960s marked the transition towards the predominance of a representational form which thereafter presented Québec as a political society. In the third tendency a functionalist sociology, linked to the consolidation of the Keynesian state, took root. The 1970s saw in contradistinction the affirmation of a sociology of emanicipation. Divided between two mutually antagonistic approaches, culturalist and political, Québec sociology is working towards the redefinition of a Québecois society henceforth caught in the determining web of modernity.     

Accuracy of the judgment of facial expression of emotions as a function of sex and level of education

The purpose of this study is to examine the recognition of facial expressions of six emotions as a function of sex and level of education (high school, college, university) of the subjects. Three hundred French-speaking citizens of Quebec had to judge which emotion was expressed in various facial stimuli presented on slides. Results show that overall, the recognition of emotions was very good. However, there were significant and strong differences between emotions and sex and levels of education did not have strong effects on the results.This research was supported by grant EQ-1717 from Fonds FCAC (Gouvernment du Quebec).     

Laggards or Leaders: Conservers of Traditional Agricultural Knowledge in Bolivia

Many sustainable agricultural practices are based on local and traditional farming knowledge. This article examines the conservation and loss of three traditional practices in the Bolivian Altiplano that agronomic research has shown increase the resiliency of small farmers in the face of climate‐related risks. These practices are the use of manure, the use of local forecasts and risk‐management strategies, and the preservation of crop biodiversity. Although these practices are widely used today, farmers have been steadily abandoning them during the past decade. This article examines the characteristics of those who maintain and those who abandon traditional practices to see if the abandonment of local knowledge can be explained by the adoption‐diffusion literature. This research does not support the adoption‐diffusion literature; although the factors related to conservation are slightly different for each practice, the findings do not support the idea that young, educated, and wealthier farmers are more likely to reject local knowledge. Instead, off‐farm activities such as migration, employment, and trade seem to be related to the decline in local practices as each affects the availability of labor and the availability of people to learn these practices.     

Reliability and validity of the French version of the 18-item Karasek job content questionnaire

Previous studies have shown that job strain, a combination of high psychological demands and low decision latitude, may be involved in the development of cardiovascular diseases and other health outcomes. In 1985, Karasek recommended a standard questionnaire to measure psychological demands (nine items) and job decision latitude (nine items). The internal consistency, factorial validity, discriminant validity, and 1-year stability of the French version of this instrument were evaluated in a population of 8263 white collar workers. Participants filled out the questionnaire during working hours in the context of a larger cardiovascular study. A subgroup of the study population completed the questionnaire again 1 year later (n = 953). Internal consistency was adequate: .74 for men and .73 for women for psychological demands and .83 for men and .81 for women for decision latitude. The results of the factor analysis were consistent with the two dimensions expected from the theory, although some items had high loading on two factors. Variations in the means of psychological demand and decision latitude scores and variations in the prevalence of high job strain by gender, job category, and age support the discriminant validity. These results were generally comparable to those observed with the English version. The percentages of workers remaining, after 1 year, in the same category of psychological demands, decision latitude, and job strain were 86.6, 86.1 and 75.2% respectively. These data provide evidence of the reliability and validity of the French version of these psychological demand and decision latitude scales     

Les métiers flous du développement rural

Along with the notion of development, new jobs have emerged characterized by: a vague, overall definition; an ambiguity as to roles (especially in dealings with locally elected officials); makeshift statuses; and occupational itineraries that cannot be reduced to civil service careers. As this examination of their work over three decades shows, the autonomy of rural development agents has been related to a “phase of invention”, when the switch was made from rural planning to development. Role ambiguity still prevails and has even spread, owing to the social conditions of employment, to occupational itineraries. Given the cofinancing of public authorities’ “territorial actions”, we see a couple forming between elected officials, who want to take hold in an intercommunal territory, and development agents, who “create their jobs” by capturing subsidies. Over this 30-year period, a peculiar image of “fuzzy jobs” has emerged.     

Home, Home Ownership and the Search for Ontological Security

The central focus of this paper is the notion that the home can provide a locale in which people can work at attaining a sense of ontological security in a world that at times is experienced as threatening and uncontrollable. The paper builds on and develops the ideas of Giddens and Saunders on ontological security and seeks to break down and operationalise the concept and explore it through a set of empirical data drawn from interviews with a group of older New Zealand home owners. The extent to which home and home life meets the conditions for the maintenance of ontological security is assessed through an exploration of home as the site of constancy in the social and material environment; home as a spatial context in which the day to day routines of human existence are performed; home as a site free from the surveillance that is part of the contemporary world which allows for a sense of control that is missing in other locales; and home as a secure base around which identities are constructed. The paper also argues that meanings of home are context specific and thus the data need to be seen in relation to New Zealanders' long standing pre-occupation with land and home ownership. The paper concludes by speculating on how meanings of home may be changing.