An illustration of variable precision rough sets model:an analysis of the findings of the UK Monopolies and Mergers Commission

This paper introduces a new technique in the investigation of limited-dependent variable models. This paper illustrates that variable precision rough set theory (VPRS), allied with the use of a modern method of classification, or discretisation of data, can out-perform the more standard approaches that are employed in economics, such as a probit model. These approaches and certain inductive decision tree methods are compared (through a Monte Carlo simulation approach) in the analysis of the decisions reached by the UK Monopolies and Mergers Committee. We show that, particularly in small samples, the VPRS model can improve on more traditional models, both in-sample, and particularly in out-of-sample prediction. A similar improvement in out-of-sample prediction over the decision tree methods is also shown.

The goal-dependent automaticity of drinking habits

In recent treatments of habitual social behaviour, habits are conceptualized as a form of goal-directed automatic behaviour that are mentally represented as goal-action links. Three experiments tested this conceptualization in the context of students' drinking (alcohol consumption)habits. Participants were randomly assigned to conditions where either a goal related to drinking behaviour (socializing) was activated, or an unrelated goal was activated. In addition, participants' drinking habits were measured. The dependent variable in Experiments 1 and 2 was readiness to drink, operationalized by speed of responding to the action concept 'drinking' in a verb verification task. Experiment 3 used the uptake of a voucher to measure drinking behaviour. Findings supported the view that when habits are established, simply activating a goal related to the focal behaviour automatically elicits that behaviour. These findings are consistent with a goal-dependent conception of habit. Possibilities for interventions designed to attenuate undesirable habitual behaviours are considered.

Probing three-way interactions in moderated multiple regression:Development and application of a slope difference test

Researchers often use 3-way interactions in moderated multiple regression analysis to test the joint effect of 3 independent variables on a dependent variable. However, further probing of significant interaction terms varies considerably and is sometimes error prone. The authors developed a significance test for slope differences in 3-way interactions and illustrate its importance for testing psychological hypotheses. Monte Carlo simulations revealed that sample size, magnitude of the slope difference, and data reliability affected test power. Application of the test to published data yielded detection of some slope differences that were undetected by alternative probing techniques and led to changes of results and conclusions. The authors conclude by discussing the test's applicability for psychological research. Copyright 2006 by the American Psychological Association.

People, technology, processes and risk knowledge sharing

The present global economic crisis creates doubts about the good use of accumulated experience and knowledge in managing risk in financial services. Typically, risk management practice does not use knowledge management (KM) to improve and to develop new answers to the threats. A key reason is that it is not clear how to break down the “organizational silos” view of risk management (RM) that is commonly taken. As a result, there has been relatively little work on finding the relationships between RM and KM. We have been doing research for the last couple of years on the identification of relationships between these two disciplines. At ECKM 2007 we presented a general review of the literature(s) and some hypotheses for starting research on KM and its relationship to the perceived value of enterprise risk management. This article presents findings based on our preliminary analyses, concentrating on those factors affecting the perceived quality of risk knowledge sharing. These come from a questionnaire survey of RM employees in organisations in the financial services sector, which yielded 121 responses. We have included five explanatory variables for the perceived quality of risk knowledge sharing. These comprised two variables relating to people (organizational capacity for work coordination and perceived quality of communication among groups), one relating to process (perceived quality of risk control) and two related to technology (web channel functionality and RM information system functionality). Our findings so far are that four of these five variables have a significant positive association with the perceived quality of risk knowledge sharing: contrary to expectations, web channel functionality did not have a significant association. Indeed, in some of our exploratory regression studies its coefficient (although not significant) was negative. In stepwise regression, the variable organizational capacity for work coordination accounted for by far the largest part of the variation in the dependent variable perceived quality of risk knowledge sharing. The “people” variables thus appear to have the greatest influence on the perceived quality of risk knowledge sharing, even in a sector that relies heavily on technology and on quantitative approaches to decision making. We have also found similar results with the dependent variable perceived value of Enterprise Risk Management (ERM) implementation.

Statnote 27: Principal components analysis (PCA)

In Statnotes 24 and 25, multiple linear regression, a statistical method that examines the relationship between a single dependent variable (Y) and two or more independent variables (X), was described. The principle objective of such an analysis was to determine which of the X variables had a significant influence on Y and to construct an equation that predicts Y from the X variables. ‘Principal components analysis’ (PCA) and ‘factor analysis’ (FA) are also methods of examining the relationships between different variables but they differ from multiple regression in that no distinction is made between the dependent and independent variables, all variables being essentially treated the same. Originally, PCA and FA were regarded as distinct methods but in recent times they have been combined into a single analysis, PCA often being the first stage of a FA. The basic objective of a PCA/FA is to examine the relationships between the variables or the ‘structure’ of the variables and to determine whether these relationships can be explained by a smaller number of ‘factors’. This statnote describes the use of PCA/FA in the analysis of the differences between the DNA profiles of different MRSA strains introduced in Statnote 26.

The accuracy of parameter estimation from noisy data, with application to resonance peak estimation in distributed Brillouin sensing

Distributed Brillouin sensing of strain and temperature works by making spatially resolved measurements of the position of the measurand-dependent extremum of the resonance curve associated with the scattering process in the weakly nonlinear regime. Typically, measurements of backscattered Stokes intensity (the dependent variable) are made at a number of predetermined fixed frequencies covering the design measurand range of the apparatus and combined to yield an estimate of the position of the extremum. The measurand can then be found because its relationship to the position of the extremum is assumed known. We present analytical expressions relating the relative error in the extremum position to experimental errors in the dependent variable. This is done for two cases: (i) a simple non-parametric estimate of the mean based on moments and (ii) the case in which a least squares technique is used to fit a Lorentzian to the data. The question of statistical bias in the estimates is discussed and in the second case we go further and present for the first time a general method by which the probability density function (PDF) of errors in the fitted parameters can be obtained in closed form in terms of the PDFs of the errors in the noisy data.

Remote sensing applications to flood hydrology in Belize

The research compares the usefullness of four remote sensing information sources, these being LANDSAT photographic prints, LANDSAT computer compatible tapes, Metric Camera and SIR-A photographic prints. These sources provide evaluations of the catchment characteristics of the Belize and Sibun river basins in Central America. Map evaluations at 1:250,000 scale are compared to the results of the same scale, remotely sensed information sources. The values of catchment characteristics for both maps and LANDSAT prints are used in multiple regression analysis, providing flood flow formulae, after investigations to provide a suitable dependent variable discharge series are made for short term records. The use of all remotely sensed information sources in providing evaluations of catchment characteristics is discussed. LANDSAT prints and computer compatible tapes of a post flood scene are used to estimate flood distributions and volumes. These are compared to values obtained from unit hydrograph analysis, using the dependent discharge series and evaluate the probable losses from the Belize river to the floodplain, thereby assessing the accuracy of LANDSAT estimates. Information relating to flood behaviour is discussed in terms of basic image presentation as well as image processing. A cost analysis of the purchase and use of all materials is provided. Conclusions of the research indicate that LANDSAT print material may provide information suitable for regression analysis at levels of accuracy as great as those of topographic maps, that the differing information sources are uniquely applicable and that accurate estimates of flood volumes may be determined even by post flood imagery.

Identification of industrial sources of airborne heavy metal pollution in urban areas

The research examines the deposition of airborne particles which contain heavy metals and investigates the methods that can be used to identify their sources. The research focuses on lead and cadmium because these two metals are of growing public and scientific concern on environmental health grounds. The research consists of three distinct parts. The first is the development and evaluation of a new deposition measurement instrument - the deposit cannister - designed specifically for large-scale surveys in urban areas. The deposit cannister is specifically designed to be cheap, robust, and versatile and therefore to permit comprehensive high-density urban surveys. The siting policy reduces contamination from locally resuspended surface-dust. The second part of the research has involved detailed surveys of heavy metal deposition in Walsall, West Midlands, using the new high-density measurement method. The main survey, conducted over a six-week period in November - December 1982, provided 30-day samples of deposition at 250 different sites. The results have been used to examine the magnitude and spatial variability of deposition rates in the case-study area, and to evaluate the performance of the measurement method. The third part of the research has been to conduct a 'source-identification' exercise. The methods used have been Receptor Models - Factor Analysis and Cluster Analysis - and a predictive source-based deposition model. The results indicate that there are six main source processes contributing to deposition of metals in the Walsall area: coal combustion, vehicle emissions, ironfounding, copper refining and two general industrial/urban processes. |A source-based deposition model has been calibrated using facctorscores for one source factor as the dependent variable, rather than metal deposition rates, thus avoiding problems traditionally encountered in calibrating models in complex multi-source areas. Empirical evidence supports the hypothesised associatlon of this factor with emissions of metals from the ironfoundry industry.

Measuring the assessment and counselling provided with the supply of non-prescription asthma reliever medication: a simulated patient study

BACKGROUND: Over one quarter of asthma reliever medications are provided without prescription by community pharmacies in Australia. Evidence that community pharmacies provide these medications with sufficient patient assessment and medication counseling to ensure compliance with the government's Quality Use of Medicines principles is currently lacking. OBJECTIVE: To assess current practice when asthma reliever medication is provided in the community pharmacy setting and to identify factors that correlate with assessment of asthma control. METHODS: Researchers posing as patients visited a sample of Perth metropolitan community pharmacies in May 2007. During the visit, the simulated patient enacted a standardized scenario of someone with moderately controlled asthma who wished to purchase a salbutamol (albuterol) inhaler without prescription. Results of the encounter were recorded immediately after the visit. Regression analysis was performed, with medication use frequency (a marker of asthma control) as the dependent variable. RESULTS: One hundred sixty community pharmacies in the Perth metropolitan area were visited in May 2007. Pharmacists and/or pharmacy assistants provided some form of assessment in 84% of the visits. Counseling was provided to the simulated patients in 24% of the visits. Only 4 pharmacy staff members asked whether the simulated patient knew how to use the inhaler. Significant correlation was found between assessment and/or counseling of reliever use frequency and 3 independent variables: visit length (p < 0.001), number of assessment questions asked (p < 0.001), and the simulated patient who conducted the visit (p < 0.02). CONCLUSIONS: Both patient assessment and medication counseling were suboptimal compared with recommended practice when nonprescription asthma reliever medication was supplied in the community pharmacy setting. Pharmacy and pharmacist demographic variables do not appear to affect assessment of asthma control. This research indicates the need for substantial improvements in practice in order to provide reliever medication in line with Quality Use of Medication principles of ensuring safe and effective use of medication.

Provision of primary care to patients with chronic cough in the community pharmacy setting

BACKGROUND: Community pharmacies are at the forefront of primary care providers and have an important role in the referral of patients to a medical practitioner for review when necessary. Chronic cough is a common disorder in the community and requires medical assessment. The proficiency of community pharmacy staff to refer patients with chronic cough is currently unknown. OBJECTIVE: To assess the ability of community pharmacy staff to recognize and medically refer patients with a chronic nonproductive cough. METHODS: Following ethics approval, a simulated patient study of 156 community pharmacies in Perth, Western Australia, was conducted over a 3-month period. Simulated patients presented to the pharmacy requesting treatment for a cough. The simulated patient required a referral based on a designated scenario. Demographic details, assessment questions, and advice provided were recorded by the simulated patient immediately postvisit. A logistic regression analysis was performed, with referral for medical assessment as the dependent variable. RESULTS: Of the 155 community pharmacies included in the analysis, 38% provided appropriate medical referral. Cough suppressants were provided as therapy in 72% of all visits. Predictors of medical referral were assessment of symptom duration, medical history, current medications being taken, frequency of reliever use, and the position of the pharmacy staff member conducting the consultation. A third of community pharmacies provided appropriate primary care by recommending medical referral advice to patients with chronic cough. The majority of pharmacy staff members acquired information from the patient that suggested a need for medical referral, yet did not provide referral advice. CONCLUSIONS: Appropriate medical referral is more likely when adequate assessment is undertaken and when a pharmacist is directly involved in the consultation. This highlights the need for pharmacies to ensure that processes are in place for patients to access the pharmacist.

Personality across working life:the longitudinal and reciprocal influences of personality on work

Summary: This paper focuses on the role of personality at different stages of people's working lives. We begin by reviewing the research in industrial, work, and organizational (IWO) psychology regarding the longitudinal and dynamic influences of personality as an independent variable at different career stages, structuring our review around a framework of people's working lives and careers over time. Next, we review recent studies in the personality and developmental psychology domain regarding the influence of changing life roles on personality. In this domain, personality also serves as a dependent variable. By blending these two domains, it becomes clear that the study of reciprocal effects of work and personality might open a new angle in IWO psychology's long-standing tradition of personality research. To this end, we outline various implications for conceptual development (e.g., trait stability) and empirical research (e.g., personality and work incongruence). Finally, we discuss some methodological and statistical considerations for research in this new research domain. In the end, our review should enrich the way that IWO psychologists understand personality at work, focusing away from its unidirectional predictivist influence on job performance toward a more complex longitudinal reciprocal interplay of personality and working life. © 2013 John Wiley & Sons, Ltd.

Factors preventing myopia progression with orthokeratology correction

PURPOSE: To examine which baseline measurements constitute predictive factors for axial length growth over 2 years in children wearing orthokeratology contact lenses (OK) and single-vision spectacles (SV). METHODS: Sixty-one children were prospectively assigned to wear either OK (n = 31) or SV (n = 30) for 2 years. The primary outcome measure (dependent variable) was axial length change at 2 years relative to baseline. Other measurements (independent variables) were age, age of myopia onset, gender, myopia progression 2 years before baseline and baseline myopia, anterior chamber depth, corneal power and shape (p value), and iris and pupil diameters as well as parental refraction. The contribution of all independent variables to the 2-year change in axial length was assessed using univariate and multivariate regression analyses. RESULTS: After univariate analyses, smaller increases in axial length were found in the OK group compared to the SV group in children who were older, had earlier onset of myopia, were female, had lower rate of myopia progression before baseline, had less myopia at baseline, had longer anterior chamber depth, had greater corneal power, had more prolate corneal shape, had larger iris diameter, had larger pupil sizes, and had lower levels of parental myopia (all p < 0.05). In multivariate analyses, older age and greater corneal power were associated with smaller increases in axial length in the OK group (both p < 0.05), whereas in SV wearers, smaller iris diameter was associated with smaller increases in axial length (p = 0.021). CONCLUSIONS: Orthokeratology is a successful treatment option in controlling axial elongation compared to SV in children of older age, had earlier onset of myopia, were female, had lower rate of myopia progression before baseline, had lower myopia at baseline, had longer anterior chamber depth, had greater corneal power, had more prolate corneal shape, had larger iris and pupil diameters, and had lower levels of parental myopia. © American Academy of Optometry.

The accuracy of parameter estimation from noisy data, with application to resonance peak estimation in distributed Brillouin sensing

Distributed Brillouin sensing of strain and temperature works by making spatially resolved measurements of the position of the measurand-dependent extremum of the resonance curve associated with the scattering process in the weakly nonlinear regime. Typically, measurements of backscattered Stokes intensity (the dependent variable) are made at a number of predetermined fixed frequencies covering the design measurand range of the apparatus and combined to yield an estimate of the position of the extremum. The measurand can then be found because its relationship to the position of the extremum is assumed known. We present analytical expressions relating the relative error in the extremum position to experimental errors in the dependent variable. This is done for two cases: (i) a simple non-parametric estimate of the mean based on moments and (ii) the case in which a least squares technique is used to fit a Lorentzian to the data. The question of statistical bias in the estimates is discussed and in the second case we go further and present for the first time a general method by which the probability density function (PDF) of errors in the fitted parameters can be obtained in closed form in terms of the PDFs of the errors in the noisy data.

Competition, agency and productivity

This article tests a set of hypotheses relating to agency and Schumpeterian views on how competition affects performance. A survey data set of Australian workplaces is used, with the change in labour productivity as the dependent variable. The results show strong support for the idea that intense competition raises productivity growth in managerial workplaces, but not in non-managerial workplaces (i.e. where the principal owner also works). Testing the agency theories in more detail, we find no evidence that the number of competitors, the price elasticity of demand or a proxy for bankruptcy (pre-tax losses) are the mechanisms behind the process. For non-managerial workplaces the results indicate support for the idea that greater demand uncertainty reduces productivity growth. In contrast, for managerial workplaces, greater demand uncertainty tends to raise productivity growth.

Handling varying amounts of missing data when classifying mental-health risk levels

One of the main challenges of classifying clinical data is determining how to handle missing features. Most research favours imputing of missing values or neglecting records that include missing data, both of which can degrade accuracy when missing values exceed a certain level. In this research we propose a methodology to handle data sets with a large percentage of missing values and with high variability in which particular data are missing. Feature selection is effected by picking variables sequentially in order of maximum correlation with the dependent variable and minimum correlation with variables already selected. Classification models are generated individually for each test case based on its particular feature set and the matching data values available in the training population. The method was applied to real patients' anonymous mental-health data where the task was to predict the suicide risk judgement clinicians would give for each patient's data, with eleven possible outcome classes: zero to ten, representing no risk to maximum risk. The results compare favourably with alternative methods and have the advantage of ensuring explanations of risk are based only on the data given, not imputed data. This is important for clinical decision support systems using human expertise for modelling and explaining predictions.