Data Driven Similarity Measures for <Emphasis Type="Italic">k</Emphasis>-Means Like Clustering Algorithms

We present an optimization approach that generates k-means like clustering algorithms. The batch k-means and the incremental k-means are two well known versions of the classical k-means clustering algorithm (Duda et al. 2000). To benefit from the speed of the batch version and the accuracy of the incremental version we combine the two in a ping–pong fashion. We use a distance-like function that combines the squared Euclidean distance with relative entropy. In the extreme cases our algorithm recovers the classical k-means clustering algorithm and generalizes the Divisive Information Theoretic clustering algorithm recently reported independently by Berkhin and Becher (2002) and Dhillon1 et al. (2002). Results of numerical experiments that demonstrate the viability of our approach are reported.This research was supported in part by the US Department of Defense, the United States–Israel Binational Science Foundation (BSF), and Northrop Grumman Mission Systems (NG/MS).     

Muscular adaptations in low- versus high-load resistance training: A meta-analysis

There has been much debate as to optimal loading strategies for maximising the adaptive response to resistance exercise. The purpose of this paper therefore was to conduct a meta-analysis of randomised controlled trials to compare the effects of low-load (≤60% 1 repetition maximum [RM]) versus high-load (≥65% 1 RM) training in enhancing post-exercise muscular adaptations. The strength analysis comprised 251 subjects and 32 effect sizes (ESs), nested within 20 treatment groups and 9 studies. The hypertrophy analysis comprised 191 subjects and 34 ESs, nested with 17 treatment groups and 8 studies. There was a trend for strength outcomes to be greater with high loads compared to low loads (difference = 1.07 ± 0.60; CI: ?0.18, 2.32; p = 0.09). The mean ES for low loads was 1.23 ± 0.43 (CI: 0.32, 2.13). The mean ES for high loads was 2.30 ± 0.43 (CI: 1.41, 3.19). There was a trend for hypertrophy outcomes to be greater with high loads compared to low loads (difference = 0.43 ± 0.24; CI: ?0.05, 0.92; p = 0.076). The mean ES for low loads was 0.39 ± 0.17 (CI: 0.05, 0.73). The mean ES for high loads was 0.82 ± 0.17 (CI: 0.49, 1.16). In conclusion, training with loads ≤50% 1 RM was found to promote substantial increases in muscle strength and hypertrophy in untrained individuals, but a trend was noted for superiority of heavy loading with respect to these outcome measures with null findings likely attributed to a relatively small number of studies on the topic.     

Measuring Multidimensional Subjective Well-Being with the I COPPE Scale in a Hispanic Sample

The purpose of this study was to provide initial validity evidence for measuring multidimensional subjective well-being in a Hispanic sample with the Interpersonal, Community, Occupational, Physical, Psychological, Economic (I COPPE) Scale. Participants were 641 English-speaking adults who self-identified as Hispanic. Bi-factor analyses were used to evaluate (a) the a priori measurement theory for responses to the I COPPE Scale and (b) convergent relationships of the seven I COPPE subjective well-being factors with scores from established comparison instruments. There was evidence that (a) the a priori hypothesized measurement theory for responses to the I COPPE Scale emerged in an exploratory bi-factor analysis and (b) the I COPPE subjective well-being factors exhibited convergent relationships with scores from comparison instruments. Use of the I COPPE Scale to derive multidimensional measures of subjective well-being may be of potential utility to future research in the interdisciplinary study of human movement and in a diversity of populations in which health disparities may exist.     

Mode 2 and the Tension Between Excellence and Utility: The Case of a Policy-Relevant Research Field in Sweden

This paper investigates the impact of changing science policy doctrines on the development of an academic field, working life research. Working life research is an interdisciplinary field of study in which researchers and stakeholders collaborated to produce relevant knowledge. The development of the field, we argue, was both facilitated and justified by the, at the time dominant, science policy orthodoxy in Sweden, sector research. Sector research science policy doctrine favoured stakeholder-driven research agendas in the fields relevant to the sector. This approach to agenda setting was highly contested by Swedish universities and left scientists vulnerable to the fallout from any conflicts arising among the stakeholder groupings that were part of the governance arrangement. Our case shows that working life research was in part a victim of the struggle between science and policy over who sets the agenda for science in Sweden. In this struggle, each side chose to use ‘scientific quality’ as a proxy for furth ing its respective interests and visions for how science should be governed. The paper argues that this case is of interest to the continued elaboration of the Mode 2 thesis and the debate about ‘relevant science’. We find that the close association with stakeholders and the concomitant dependence it created left working life research unable to defend itself against its critics and that this state of affairs was particularly problematic for social science research on working life.     

Fundamental Issues Regarding the Nature of Technology

Science & Education - Science and technology are so intertwined that technoscience has been argued to more accurately reflect the progress of science and its impact on society, and most...     

Being a teacher: altruistic and narcissistic expectations of pre-service teachers

The article focuses on investigating pre-service teachers’ expectations of their future teaching career, in particular concerning teacher–student interrelations. In an attempt to comprehend why people choose teaching as a professional career, a conceptual model titled ‘Teachers’ altruistic-narcissistic classroom expectations’ was designed and tested. The model puts forward for consideration the idea that pre-service teachers view their future relations with students as being based on four basic psychological foundations: genuine altruism, paternalistic altruism, benevolent narcissism, and genuine narcissism. A study to test for the validity of this model, using Facet Theory and smallest space analysis as the methodological approach is reported. One hundred and sixty student–teachers participated in the study. The findings provided evidence in support of the model’s validity. It is argued that altruism and narcissism conjointly may be regarded as factors motivating people to opt for teaching as a career, and that altruistic and narcissistic expectations can predict teachers’ classroom behavior.     

Distributed media cooperation for enhanced video communication

INTRODUCTION According to yearly statistics the demand for multimedia traffic over computer networks is con- stantly on the increase (Nielsen//NetRatings, http:// www.nielsen-netratings.com/). Therefore, it is inevi- table that situations where different media streams will need to share communication resources will be- come commonplace. In such cases, it would be of interest to optimize the overall performance of the network. In other words, resources should be allo- cated across the com…     

Ko wai Au? Who am I? Examining the multiple identities of Māori youth

In this paper, we present preliminary findings from a unique collaborative research project involving six Deaf Māori rangatahi (youth) in Tāmaki Makaurau (Auckland), Aotearoa New Zealand. This study utilized kaupapa whānau (research family) protocols, established in consultation with two cultural advisory groups within New Zealand and the young people themselves, combined with elements of photovoice methodology, to explore the identities of these youth. Emerging findings highlight the complex nature of these youth’s cultural identity as well as specific issues related to access to and participation within te ao Māori (the Māori world). Specific and critical reflections on the research process are also included.     

Where the boys aren't: non-cognitive skills, returns to school and the gender gap in higher education

Nearly 60 percent of college students today are women. Using longitudinal data on a nationally representative cohort of eighth grade students in 1988, I examine two potential explanations for the differential attendance rates of men and women—returns to schooling and non-cognitive skills. The attendance gap is roughly five percentage points for all high school graduates. Conditional on attendance, however, there are few differences in type of college, enrollment status or selectivity of institution. The majority of the attendance gap can be explained by differences in the characteristics of men and women, despite some gender differences in the determinants of college attendance. I find that higher non-cognitive skills and college premiums among women account for nearly 90 percent of the gender gap in higher education. Interestingly, non-cognitive factors continue to influence college enrollment after controlling for high school achievement.