Probabilistic text understanding

We discuss a new framework for text understanding. Three major design decisions characterize this approach. First, we take the problem of text understanding to be a particular case of the general problem of abductive inference. Second, we use probability theory to handle the uncertainty which arises in this abductive inference process. Finally, all aspects of natural language processing are treated in the same framework, allowing us to integrate syntactic, semantic and pragmatic constraints. In order to apply probability theory to this problem, we have developed a probabilistic model of text understanding. To make it practical to use this model, we have devised a way of incrementally constructing and evaluating belief networks. We have written a program,wimp3, to experiment with this framework. To evaluate this program, we have developed a simple single-blind testing method.     

The effect of variability in the fertility schedule on numbers of kin

"The kinship formulas developed by Goodman, Keyfitz, and Pullum (1974) are extended to encompass populations in which fertility varies among women. An expression is derived to determine the amount by which the number of sisters in a heterogeneous population exceeds that in a population with homogeneous fertility. This expression, which is a function of the variance in the gross reproduction rate of the population, can readily be applied to numbers of other kin, such as aunts and cousins. Several trial calculations indicate that calculations of average numbers of sisters based on an assumption of uniform fertility could result in a underestimate of about 13 percent." (SUMMARY IN FRE)     

Co‐producing innovation or innovating co‐production? Responding to the contact needs of non‐resident parents in the Republic of Ireland

The empowerment and participation of children and families in both the design and the delivery of services oriented towards them is now an accepted norm, if not always the accepted practice. Yet, challenges still remain, particularly where parents are separated and contact issues are problematic. Although contact centres are common in some Western countries including Australia, New Zealand, the UK, France, and the USA, this is not the case in the Republic of Ireland. The relatively recent introduction of divorce, increases in separation and childbirth outside of marriage have led to the need for service planners to respond to the growing issue of non‐resident contact with children. Using Bovaird's co‐production framework, the purpose of this paper is to outline and discuss specific findings relating to service user involvement emanating from a formative evaluation of an innovative supported contact service in Ireland. What emerges from the findings is a picture of co‐production where the broad parameters of the service are professionally defined and where the specifics of service user involvement in delivery vary from family to family but within the bounds of professional control.     

Beyond Self‐Reports: Changes in Biomarkers as Predictors of Mortality

The proliferation of biosocial surveys has increased the importance of weighing the costs and benefits of adding biomarker collection to population‐based surveys. A crucial question is whether biomarkers offer incremental value beyond self‐reported measures, which are easier to collect and impose less respondent burden. We use longitudinal data from a nationally representative sample of older Taiwanese (aged 54+ in 2000, examined in 2000 and 2006 with mortality follow‐up through 2011) to address that question with respect to predicting all‐cause mortality. A summary measure of biomarkers improves mortality prediction (as measured by the area under the receiver operating characteristic curve) compared with self‐reports alone, but individual biomarkers perform better than the summary score. We find that incorporating change in biomarkers over a six‐year period yields a small improvement in mortality prediction compared with one‐time measurement. But, is the incremental value worth the costs?     

The Best Predictors of Survival: Do They Vary by Age,Sex, and Race?

We consider a broad set of variables used by social scientists and clinicians to identify the leading predictors of five‐year survival among American adults. We address a question not considered in earlier research: Do the strongest predictors of survival vary by age, sex or race/ethnicity? The analysis uses hazard models with 30 well‐established predictors to examine five‐year survival in the National Health and Nutrition Examination Survey. We find that the simple measure of self‐assessed health and self‐reported measures of functional ability and disability are the strongest predictors in all demographic groups, and are generally ranked considerably higher than biomarkers. Among the biomarkers, serum albumin is highly ranked in most demographic groups, whereas clinical measures of cardiovascular and metabolic function are consistently among the weakest predictors. Despite these similarities, there is substantial variation in the leading predictors across demographic groups, most notably by race and ethnicity     

Determinants of Mortality at Older Ages: The Role of Biological Markers of Chronic Disease

Researchers have had a longstanding interest in understanding the determinants of mortality. This article examines the impact of a broad array of biological markers, together with self‐reports of physical and mental health status, on the probability of dying for older adults. The estimates are derived from logistic regression models based on data from a national survey in Taiwan. The analysis confirms previous studies demonstrating the effects of clinical measures related to metabolic syndrome on mortality and identifies detrimental effects of neuroendocrine and immune‐system markers. The results reveal that biomarkers provide independent explanatory power in the presence of self‐reported health measures. The associations between biomarkers and mortality found here provide new avenues for projecting future mortality and elucidating differences in longevity across populations.     

Foundations of British sociology 1880–1930: contexts and biographies

‘This paper provides an overview of aspects of the history of British sociology. In particular, it tries to answer critical historical work by among others, Perry Anderson and Philip Abrams, which sought to explain the supposed indigenous ‘failure’ to develop academic sociology in Britain before the 1960s. It is argued that a narrowly academic reading of the history of sociology cannot do justice to its role in the service of social administration and public enlightenment and may exaggerate the degree to which sociology from its foundations was conceived as a purely intellectual discipline. The paper points to a thriving sociological culture in Britain in the generation before the First World War, though it was one in which many contributions came from philosophers, natural scientists and political economists rather then self‐proclaimed ‘sociologists’. It ends with a brief review of Patrick Geddes and Victor Branford, a founder of the Sociological Society and editor of the Sociological Review, whose biographies and eclectic social and international interests tell us something about the personalities and political interests of early British sociological pioneers.’     

Evaluation of the "Lose Your Excuse" public service advertising campaign for tweens to save energy

This study evaluates the 2008-2009 "Lose your Excuse" public service advertising (PSA) campaign on energy efficiency targeting 8- to 12-year-olds, intended to increase knowledge, foster proactive attitudes, and change energy usage behaviors. Baseline and two follow-up surveys were conducted with online samples representative of the national population of households with kids with online access. Almost half (47%) of the tweens recognized at least one ad from the campaign. Ad recognition was positively associated with knowledge, proactive attitudes, and energy-saving behavior. Propensity score analysis confirmed a small but measurable and statistically significant effect on energy-saving behavior. The discussion section compares these results to public health campaigns in terms of ghost awareness, reach, and effect size.     

From the new left to the new enlightenment: the methodological implications of public attention on private lives

Research on Rajneeshpuram was partially supported by the Center for the Study of Women in Society and a Summer Research Grant from University of Oregon. Research on the fate of Santa Barbara activists was partially supported by grants from the Society for the Psychological Study of Social Issues and the Graduate Division of the University of California at Santa Barbara.