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     

Discrete‐time survival trees

Tree‐based methods are frequently used in studies with censored survival time. Their structure and ease of interpretability make them useful to identify prognostic factors and to predict conditional survival probabilities given an individual's covariates. The existing methods are tailor‐made to deal with a survival time variable that is measured continuously. However, survival variables measured on a discrete scale are often encountered in practice. The authors propose a new tree construction method specifically adapted to such discrete‐time survival variables. The splitting procedure can be seen as an extension, to the case of right‐censored data, of the entropy criterion for a categorical outcome. The selection of the final tree is made through a pruning algorithm combined with a bootstrap correction. The authors also present a simple way of potentially improving the predictive performance of a single tree through bagging. A simulation study shows that single trees and bagged‐trees perform well compared to a parametric model. A real data example investigating the usefulness of personality dimensions in predicting early onset of cigarette smoking is presented. The Canadian Journal of Statistics 37: 17‐32; 2009 © 2009 Statistical Society of Canada