RELATIONS BETWEEN ERGODICITY AND MEAN DRIFT FOR MARKOV CHAINS1

If {X_n} is an irreducible aperiodic Markov chain on a state apace denote the mean one step change of position, or “drift”, of {X_n} at j. The main result of this note is to show that, when |µ(j)| is bounded, {X_n} admits a stationary distribution {π_j}if and only if for some N > 0 and some state i, lim inf ∑when this holds, the limit infimum is in fact . Many of the known sufficient or necessary criteria for the existence of a stationary distribution can then be derived easily from this and related results.     

How Does Performance Management Affect Workers? Beyond Human Resource Management and Its Critique

While performance management (PM) is pervasive across contemporary workplaces, extant research into how performance management affects workers is often indirect or scattered across disciplinary silos. This paper reviews and synthesizes this research, identifies key gaps and explores ‘recognition theory’ as a nascent framework that can further develop this important body of knowledge. The paper develops in three main stages. The first stage reviews ‘mainstream’ human resource management (HRM) research. While this research analyses workers’ reactions to performance management in some depth, its focus on serving organizational goals marginalizes extra‐organizational impacts. The second stage reviews more critical HRM research, which interprets performance management as a disciplinary, coercive or inequitable management device. While this literature adds an important focus on organizational power, there is scope to analyse further how PM affects workers’ well‐being. To develop this strand of PM research, the third stage turns to the emerging field of recognition theory independently developed by Axel Honneth and Christophe Dejours. The authors focus especially on recognition theory's exploration of how (in)adequate acknowledgement of workers’ contributions can significantly affect their well‐being at the level of self‐conception. Although recognition theory is inherently critical, the paper argues that it can advance both mainstream and critical performance management research, and also inform broader inquiry into recognition and identity at work.     

Total quality management and information technology

Total Quality Management is now well-established as a central method for improving the efficiency and competitive position of firms and organizations. Developed in a manufacturing environment, it has shown the same attributes in administrative and service industries. In this paper we examine the potential for use in the Information Technology sector. Little impact has been made by TQM in the IT industry. We use the famous 14 points of W. Edwards Deming to discuss the potential and the pitfalls awaiting such implementation. The particular nature of the industry means that the techniques are not transparently transportable, and we develop some initial ideas for using TQM in situations which are traditionally seen as one-off and not susceptible to process analysis and measurement. We illustrate the idea of designing in quality by focusing on software design for reduction of maintenance costs and also touch on the role of the IT sector in corporate TQM development.     

THE IMPACT OF METHOD CHOICE ON META-ANALYSIS

This paper focusses mainly on three crucial choices identified in recent meta-analyses, namely (a) the effect of using approximate statistical techniques rather than exact methods, (fa) the effect of using fixed or random effect models, and (c) the effect of publication bias on the meta-analysis result. The paper considers their impact on a set of over thirty studies of passive smoking and lung cancer in non-smokers, and addresses other issues such as the role of study comparability, the choice of raw or adjusted data when using published summary statistics, and the effect of biases such as misclassification of subjects and study quality. The paper concludes that, at least in this example, different conclusions might be drawn from metaanalyses based on fixed or random effect models; that exact methods might increase estimated confidence interval widths by 5–20% over standard approximate (logit and Mantel-Haenszel) methods, and that these methods themselves differ by this order of magnitude; that taking study quality into account changes some results, and also improves homogeneity; that the use of unadjusted or author-adjusted data makes limited difference; that there appears to be obvious publication bias favouring observed raised relative risks; and that the choice of studies for inclusion is the single most critical choice made by the modeller.     

Non-Gaussian Conditional Linear AR(1) Models

This paper gives a general formulation of a non-Gaussian conditional linear AR(1) model subsuming most of the non-Gaussian AR(1) models that have appeared in the literature. It derives some general results giving properties for the stationary process mean, variance and correlation structure, and conditions for stationarity. These results highlight similarities with and differences from the Gaussian AR(1) model, and unify many separate results appearing in the literature. Examples illustrate the wide range of properties that can appear under the conditional linear autoregressive assumption. These results are used in analysing three real datasets, illustrating general methods of estimation, model diagnostics and model selection. In particular, the theoretical results can be used to develop diagnostics for deciding if a time series can be modelled by some linear autoregressive model, and for selecting among several candidate models.     

Examples of the application of matrix factors of a design matrix

When the elements of a design matrix are rational numbers and the variances of the observations are rational multiples of a common real constant, the covariances being zero, the design matrix may be factorised into a product of matrices which have usefil statistical interpretations. The main factor matrices have integer elements, while the other factor matrices are diagonal with rational elements. Weights which are rational numbers, and missing observations, are readily accommodated. A computer is usually needed to find the factors. This paper shows how, once the factors have been found, they may be employed for any suitable set of observations without further need for such assistance.     

‘Talkin' Jockney’? Variation and change in Glaswegian accent1

This paper presents an analysis of language variation and change in a socially stratified corpus of Glaswegian collected in 1997. Eight consonantal variables in read and spontaneous speech from 32 speakers were analysed separately and then together using multivariate analysis. Our results show that middle‐class speakers, with weaker network ties and more opportunities for mobility and contact with English English speakers, are maintaining traditional Scottish features. Working‐class adolescents, with more limited mobility and belonging to close‐knit networks, are changing their vernacular by using ‘non‐local’ features such as TH‐fronting and reducing expected Scottish features such as postvocalic /r/. We argue that local context is the key to understanding the findings. Mobility and network structures are involved, but must be taken in conjunction with the recent history of structural changes to Glasgow and the resulting construction of local class‐based language ideologies which continue to be relevant in the city today.