Further exploratory analysis of split-plot experiments to study certain stratified effects

Designed experiments are a key component in many companies' improvement strategies. Because completely randomized experiments are not always reasonable from a cost or physical perspective, split-plot experiments are prevalent. The recommended analysis accounts for the different sources of variation affecting whole-plot and split-plot error. However experiments on industrial processes must be run and, consequently analyzed quite differently from ones run in a controlled environment. Such experiments are typically subject to a wide array of uncontrolled, and barely understood, variation. In particular, it is important to examine the experimental results for additional, unanticipated sources of variation. In this paper, we consider how unanticipated, stratified effects may influence a split-plot experiment and discuss further exploratory analysis to indicate the presence of stratified effects. Examples of such experiments are provided, additional tests are suggested and discussed in light of their power, and recommendations given.     

AP(E)L as a guidance and support mechanism in the Diploma in Social Work

In order to contextualize the multi-interpretative paradigms that characterize the present HE, we propose a model in which the Assessment of Prior and Experiential Learning, AP(E)L, focuses more on assessment and less on accreditation, with a four phase application of the interrelationship between AP(E)L and Learning Development for pre-entry, entry, on-programme and exit of a student's time within a programme of study. In turn, this brings into focus the way experience has now become a contested space that reflects paradigm shifts in the focus of learning and the way learning is to be acknowledged in the new universities. This account details the application of this model within the social work subject area at the University of East London. In many ways, this reflects not only the pre-supposition, central to the AP(E)L policy, that learning should be recognized wherever it takes place, but also the shift from interior intellectual systems to a more open epistemology that has been shaped by the relationship of knowledge producers and knowledge users in specific markets.     

Preschool-Age Adopted Chinese Girls’ Behaviors That Were Most Concerning to Their Mothers

Children's behavioral difficulties are a major source of parental concern. The goals of the current study were to (1) describe preschool-age adopted Chinese girls’ behaviors that were most concerning to their adoptive mothers and (2) examine the relation between children's concerning behaviors and parent/child demographics. Qualitative data from 422 mothers on 480 adopted Chinese girls (M = 44.2 months, SD = 15.2) were collected using three sequenced methods (i.e., open-ended question, follow-ups, and a focus group discussion). These girls were adopted at 3 to 60 months of age (M = 13.8, SD = 8.0) and had lived in the adoptive homes for at least 6 months (M = 30.2, SD = 14.4). Data analysis showed that 242 (57.3%) mothers (representing 274 girls) reported one to three concerning behaviors per child (total = 323 concerns), including behaviors that indicated attachment problems (107 or 33.1%), poor social skills (46 or 14.2%), language/speech problems (42 or 13%), sleep problems (29 or 9%), health/physical problems (24 or 7.4%), and problems in other areas (e.g., potty training) (75 or 23.2%). Further analysis showed that the mothers with prior experience of raising adopted children, but not biological children, were more likely to report concerning attachment behaviors than first-time mothers.     

Does poor weight gain identify children in need?

To determine the socioeconomic variables in children with slow growth (failure to thrive), a case–control study was carried out in two economically deprived areas of Newcastle upon Tyne. Three hundred and six children, aged 12–24 months, registered at two health authority clinics, were studied by reviewing their growth charts. Sixty-three had poor growth (weight deviating downwards across two major centiles from the maximum centile attained at 4–8 weeks of age). Five children were excluded because of known organic disease. A matched control group of normally growing children was selected from the same clinics. Study children and controls were examined and their parents interviewed on a range of socioeconomic, health and interactional factors. A health visitor assessment and a record of social work contact was obtained. The slowly growing children differed significantly from controls in being lighter and shorter, while their mothers spent less on food, needed more help with the child, had a poorer home environment and expressed greater concern over their child's health. Four different patterns of poor growth were identified by inspection of the charts: early onset (before 6 months but sustained), late onset (after 6 months), temporary and recurrent. The ‘recurrent’ group was associated with parenting difficulties and the ‘late onset’ group with adverse socioeconomic factors. Slow growth patterns in children under 2 years may be used as a screening tool to identify ‘children in need’ and should be sought on a population basis. Those identified can then be assessed further to ascertain the area where underlying difficulties lie, for intervention to follow as appropriate.     

Validation of a longitudinally measured surrogate marker for a time-to-event endpoint

The objective of this paper is to extend the surrogate endpoint validation methodology proposed by Buyse et al. (2000) to the case of a longitudinally measured surrogate marker when the endpoint of interest is time to some key clinical event. A joint model for longitudinal and event time data is required. To this end, the model formulation of Henderson et al. (2000) is adopted. The methodology is applied to a set of two randomized clinical trials in advanced prostate cancer to evaluate the usefulness of prostate-specific antigen (PSA) level as a surrogate for survival.     

Dicing with the unknown

There are many things that I am uncertain about, says Tony O'Hagan. Some are merely unknown to me, while others are unknowable. This article is about different kinds of uncertainty, and how the distinction between them impinges on the foundations of Probability and Statistics.     

The Moderating Role of Power Distance on the Reaction of Consumers to the CEO as a Spokesperson During a Product Harm Crisis: Insights From China and South Korea

During a crisis the corporate message is not the only issue facing the company. The role of the spokesperson is an under-researched area which is examined in this paper. In studies conducted in South Korea and China we examine the reaction of consumers to the CEO as a spokesperson during a product harm crisis. We find in both countries that consumer responses to the CEO was contingent on the consumers' level of power distance. When consumers had high levels of power distance they had higher future purchase intentions when compared with consumers who had low levels of power distance when the CEO was the spokesperson during the crisis. In addition, in a study conducted in South Korea we find that higher levels of power distance generate increased levels of brand trust when the CEO is the spokesperson, which in turn increases future purchase intentions. Our studies have important theoretical and managerial implications which are discussed in the paper.