Cluster measurement and benchmarking - A recent Welsh approach

Drawing on extensive academic research and theory on clusters and their analysis, the methodology employed in this pilot study (sponsored by the Welsh Assembly Government’s Economic Research Grants Assessment Board) seeks to create a framework for reviewing and monitoring clusters in Wales on an ongoing basis, and generate the information necessary for successful cluster development policy to occur. The multi-method framework developed and tested in the pilot study is designed to map existing Welsh sectors with cluster characteristics, uncover existing linkages, and better understand areas of strength and weakness. The approach adopted relies on synthesising both quantitative and qualitative evidence. Statistical measures, including the size of potential clusters, are united with other evidence on input-output derived inter-linkages within clusters and to other sectors in Wales and the UK, as well as the export and import intensity of the cluster. Multi Sector Qualitative Analysis is then designed for competencies/capacity, risk factors, markets, types and crucially, the perceived strengths of cluster structures and relationships. The approach outlined above can, with the refinements recommended through the review process, provide policy-makers with a valuable tool for reviewing and monitoring individual sectors and ameliorating problems in sectors likely to decline further.

Reliability measures for local nodes assessment in classification trees

Most of the modem developments with classification trees are aimed at improving their predictive capacity. This article considers a curiously neglected aspect of classification trees, namely the reliability of predictions that come from a given classification tree. In the sense that a node of a tree represents a point in the predictor space in the limit, the aim of this article is the development of localized assessment of the reliability of prediction rules. A classification tree may be used either to provide a probability forecast, where for each node the membership probabilities for each class constitutes the prediction, or a true classification where each new observation is predictively assigned to a unique class. Correspondingly, two types of reliability measure will be derived-namely, prediction reliability and classification reliability. We use bootstrapping methods as the main tool to construct these measures. We also provide a suite of graphical displays by which they may be easily appreciated. In addition to providing some estimate of the reliability of specific forecasts of each type, these measures can also be used to guide future data collection to improve the effectiveness of the tree model. The motivating example we give has a binary response, namely the presence or absence of a species of Eucalypt, Eucalyptus cloeziana, at a given sampling location in response to a suite of environmental covariates, (although the methods are not restricted to binary response data).

Population pharmacokinetics and association between A77 1726 plasma concentrations and disease activity measures following administration of leflunomide to people with rheumatoid arthritis

Aims To investigate the concentration-effect relationship and pharmacokinetics of leflunomide in patients with rheumatoid arthritis (RA). Methods Data were collected from 23 RA patients on leflunomide therapy (as sole disease modifying antirheumatic drug (DMARD)) for at least 3 months. Main measures were A77 1726 (active metabolite of leflunomide) plasma concentrations and disease activity measures including pain, duration/intensity of morning stiffness, and SF-36 survey. A population estimate was sought for apparent clearance (CL/F ) and volume of distribution was fixed (0.155 l kg(-1)). Factors screened for influence on CL/F were weight, age, gender and estimated creatinine clearance. Results Significantly higher A77 1726 concentrations were seen in patients with less swollen joints and with higher SF-36 mental summary scores than in those with measures indicating more active disease (P < 0.05); concentration-effect trends were seen with five other disease activity measures. Statistical analysis of all disease activity measures showed that mean A77 1726 concentrations in groups with greater control of disease activity were significantly higher than those in whom the measures indicated less desirable control (P < 0.05). There was large between subject variability in the dose-concentration relationship. A steady-state infusion model best described the pharmacokinetic data. Inclusion of age as a covariate decreased interindividual variability (P < 0.01), but this would not be clinically important in terms of dosage changes. Final parameter estimate (% CV interindividual variability) for CL/F was 0.0184 l h(-1) (50%) (95% CI 0.0146, 0.0222). Residual (unexplained) variability (% CV) was 8.5%. Conclusions This study of leflunomide in patients using the drug clinically indicated a concentration-effect relationship. From our data, a plasma A77 1726 concentration of 50 mg l(-1) is more likely to indicate someone with less active disease than is a concentration around 30 mg l(-1). The marked variability in pharmacokinetics suggests a place for individualized dosing of leflunomide in RA therapy.

Assessment of statistical change criteria used to define significant change in neuropsychological test performance following cardiac surgery

Objective: This paper compares four techniques used to assess change in neuropsychological test scores before and after coronary artery bypass graft surgery (CABG), and includes a rationale for the classification of a patient as overall impaired. Methods: A total of 55 patients were tested before and after surgery on the MicroCog neuropsychological test battery. A matched control group underwent the same testing regime to generate test–retest reliabilities and practice effects. Two techniques designed to assess statistical change were used: the Reliable Change Index (RCI), modified for practice, and the Standardised Regression-based (SRB) technique. These were compared against two fixed cutoff techniques (standard deviation and 20% change methods). Results: The incidence of decline across test scores varied markedly depending on which technique was used to describe change. The SRB method identified more patients as declined on most measures. In comparison, the two fixed cutoff techniques displayed relatively reduced sensitivity in the detection of change. Conclusions: Overall change in an individual can be described provided the investigators choose a rational cutoff based on likely spread of scores due to chance. A cutoff value of ≥20% of test scores used provided acceptable probability based on the number of tests commonly encountered. Investigators must also choose a test battery that minimises shared variance among test scores.

Evaluation of the indirect measures of rock brittleness and fracture toughness in rock cutting

Performance prediction models for partial face mechanical excavators, when developed in laboratory conditions, depend on relating the results of a set of rock property tests and indices to specific cutting energy (SE) for various rock types. There exist some studies in the literature aiming to correlate the geotechnical properties of intact rocks with the SE, especially for massive and widely jointed rock environments. However, those including direct and/or indirect measures of rock fracture parameters such as rock brittleness and fracture toughness, along with the other rock parameters expressing different aspects of rock behavior under drag tools (picks), are rather limited. With this study, it was aimed to investigate the relationships between the indirect measures of rock brittleness and fracture toughness and the SE depending on the results of a new and two previous linear rock cutting programmes. Relationships between the SE, rock strength parameters, and the rock index tests have also been investigated in this study. Sandstone samples taken from the different fields around Ankara, Turkey were used in the new testing programme. Detailed mineralogical analyses, petrographic studies, and rock mechanics and rock cutting tests were performed on these selected sandstone specimens. The assessment of rock cuttability was based on the SE. Three different brittleness indices (B1, B2, and B4) were calculated for sandstones samples, whereas a toughness index (T-i), being developed by Atkinson et al.(1), was employed to represent the indirect rock fracture toughness. The relationships between the SE and the large amounts of new data obtained from the mineralogical analyses, petrographic studies, rock mechanics, and linear rock cutting tests were evaluated by using bivariate correlation and curve fitting techniques, variance analysis, and Student's t-test. Rock cutting and rock property testing data that came from well-known studies of McFeat-Smith and Fowell(2) and Roxborough and Philips(3) have also been employed in statistical analyses together with the new data. Laboratory tests and subsequent analyses revealed that there were close correlations between the SE and B4 whereas no statistically significant correlation has been found between the SE and T-i. Uniaxial compressive and Brazilian tensile strengths and Shore scleroscope hardness of sandstones also exhibited strong relationships with the SE. NCB cone indenter test had the greatest influence on the SE among the other engineering properties of rocks, confirming the previous studies in rock cutting and mechanical excavation. Therefore, it was recommended to employ easy-to-use index tests of NCB cone indenter and Shore scleroscope in the estimation of laboratory SE of sandstones ranging from very low to high strengths in the absence of a rock cutting rig to measure it until the easy-to-use universal measures of the rock brittleness and especially the rock fracture toughness, being an intrinsic rock property, are developed.