A new model and class of designs for mixture experiments with process variables

Experiments that involve the blending of several components are known as mixture experiments. In some mixture experiments, the response depends not only on the proportion of the mixture components, but also on the processing conditions, A new combined model is proposed which is based on Taylor series approximation and is intended to be a compromise between standard mixture models and standard response surface models. Cost and/or time constraints often limit the size of industrial experiments. With this in mind, we present a new class of designs that will accommodate the fitting of the new combined model.     

Optimal estimation in rotation patterns

The aim of this paper is to examine the setting of surveys repeated over time when the elements in the sample are rotated in a predesigned way. On each occasion the best linear unbiased estimator (BLUE) of the current population mean, built on all past responses, is to be found. The most straightforward approach would be to compute the estimator as a solution of a least squares problem with linear restrictions. However, this method has certain drawbacks related to the fact that the size of the response data set increases over time. We follow a different approach based on finding linear recurrence relationships between optimal estimators obtained on successive occasions. Most of the original disadvantages are then corrected. In this context we present the solution to the BLUE estimation problem for some—sufficiently regular—classes of rotation patterns.     

How to Maintain High Retention Rates in Long-Term Research on Addiction: A Case Report

High dropout rates threaten the validity of longitudinal studies involving participants suffering from substance use disorder (SUD). The objective of this article is to evaluate strategies to improve retention. The data analysis was based on an ongoing study of long-term recovery of neurocognitive and psychosocial functions. We used traditional strategies to enhance the contact rate. Common factors in psychosocial treatments and biweekly Short Message Service (SMS; text messaging) monitoring were added to further increase the contact rate. The participants (n = 146) were recruited from treatment facilities for SUD. Assessments were measured at 3 months, 6 months, 9 months, and 12 months. This study was successful in retaining a cohort of participants who are typically characterized as having a negative prognosis because of their unstable living arrangements and incomes. Compared to those using GHB, addictive medications, and opiates, the retention rate was the lowest among participants with severe alcohol use disorder (42%).     

Power differentials and identity formation: Images of self and other on the Russian–Georgian boundary

This article looks at how Russia places herself in relation to one of her southern neighbours, Georgia, and vice versa. Russia and Georgia have recently been engaged in a short but full-fledged war, hence their interrelationship has been intensely debated in both countries. Both Russia and Georgia are, as it were, poised ‘between East and West’.

As a starting point, therefore, we hypothesized that Russians would present themselves as a European nation while they would orientalise the Georgians. Conversely, the Georgians would define themselves into and the Russians out of Europe. We found, however, that identity construction on the Russian-Georgian border is not symmetrical. While the Georigan discourse basically confirmed our assumption, in Russia the dominant discourse is that Russia and Georgia are closely related, fraternal peoples. This shows the importance of power relations in the study of reciprocal identity formations.

While hegemonic discourses often are discourses of exclusion our study shows that a discourse of inclusion—in our case a rhetoric of ‘brotherhood’—often may be a more effective technique of domination. The prevailing discourse in the weaker group, on the other hand, will focus more on cultural distance towards the more powerful Other.     

Metabolism and colonization. Modes of production and the physical exchange between societies and nature

Written by a sociologist and a biologist, this paper attempts an interdisciplinary approach to describing the basic exchange relations between human societies and their natural environments. One type of exchange relation is termed ‘metabolism’ and related to the biological metabolism member organisms of societies require. A historical overview (part 1) demonstrates this exchange relation in terms of mass throughput per inhabitant to have grown in the course of human cultural evolution—without necessarily increasing the quality of life of those concerned—to the twentyfold it now amounts to in industrial societies (as is demonstrated empirically for Austria in part 3). A strategy of ‘contraction of physical metabolism’ (reduction of physical growth irrespective of ‘economic’ growth) of industrial societies is proposed as a strategic means of survival and possible ways to this goal are discussed quantitatively. The other exchange relation termed ‘colonization’ refers to treatments of natural environments that purposively change some components to render better exploitability (for the purpose of social metabolism), while still relying upon their basic self‐regenerating qualities. It is sketched how colonization strategies developed historically, and it is demonstrated empirically that industrial societies now use about 50% of the available plant biomass (the energetic basis of all animal life) upon their territories for human purposes (part 4). Part 5 classifies different ‘paradigms’ for judging the ‘harmfulness’ of social interventions into the environment and outlines the logic of an information system that would enable society to generate an awareness of its own interventions into nature. On the whole the paper presents a theoretical as well as an empirical attempt to view societies as physical systems (among other physical systems on this planet) and confront sociology with the paradigmatic task to analyze the social regulation of these physical processes.     

On step fixed-charge transportation problem

The fixed-charge problem is a nonlinear programming problem of practical interest in business and industry. One of its variations is the fixed-charge transportation problem (FCTP) where fixed cost is incurred for every route that is used in the solution, along with the variable cost that is proportional to the amount shipped. That cost structure causes the value of the objective function Z to also behave like a step function. Each time we open or close a route the objective function jumps a step. The step fixed-charge transportation problem (SFCTP) is a variation of the FCTP where the fixed cost is in the form of a step function dependent on the load in a given route. While the value of the objective function Z in the FCTP is a step function, the introduction of the step fixed cost in the SFCTP results in the objective function Z being itself a step function with many more steps. Fixed-charge problems are usually solved using sophisticated analytical or computer software. This paper discusses the theory of SFCTP and presents a computationally simple heuristic algorithm for solving small SFCTPs.     

A psychosocial and biomedical comparison between men in six contrasting service occupations

A study of psychosocial job factors and physiological states was conducted using 150 men, aged between 25 and 60, working in six widely different occupations—freight handlers, aircraft mechanics, air traffic controllers, physicians, waiters and symphony musicians. The aim of the study was to characterize any differences in physiological parameters in relation to differences in psychosocial job characteristics. Analyses of variance were used to test differences between groups. The results indicated that waiters, the group reporting the most unfavourable working conditions, tended to have many cardiovascular risk factors. Physicians, who had relatively few cardiovascular risk factors, reported high decision latitude and many emotional reactions in their daily life and had relatively high cortisol and low prolactin levels in plasma.     

The exact distribution function of the ratio of two dependent quadratic forms

A closed-form representation of the distribution function of the ratio of two linear combinations of Chi-squared variables is derived. The ratio is of the following form R = (X + aY)/(bY + Z), where X, Y, Z are independent Chi-square variables and a, b > 0. Two methods of obtaining the distribution function of this ratio are used. The exact density function of such a ratio is then obtained by differentiation. Two numerical examples are provided.     

On the difference in inference and prediction between the joint and independent f-error models for seemingly unrelated regressions

We consider likelihood and Bayesian inferences for seemingly unrelated (linear) regressions for the joint niultivariate terror (e.g. Zellner, 1976) and the independent t-error (e.g. Maronna, 1976) models. For likelihood inference, the scale matrix and the shape parameter for the joint terror model cannot be consistently estimated because of the lack of adequate information to identify the latter. The joint terror model also yields the same MLEs for the regression coefficients and the scale matrix as for the independent normal error model. which are not robust against outliers. Further, linear hypotheses with respect

to the regression coefficients also give rise to the same mill distributions AS for the independent normal error model, though the MLE has a non-normal limiting distribution. In contrast to the striking similarities between the joint t-error and the independent normal error models, the independent f-error model yields AiLEs that are lubust against uuthers. Since the MLE of the shape parameter reflects the tails of the data distributions, this model extends the independent normal error model for modeling data distributions with relatively t hicker tails. These differences are also discussed with respect to the posterior and predictive distributions for Bayesian inference.