A scaleable document clustering approach for large document corpora

In this paper, the scalability and quality of the contextual document clustering (CDC) approach is demonstrated for large data-sets using the whole Reuters Corpus Volume 1 (RCV1) collection. CDC is a form of distributional clustering, which automatically discovers contexts of narrow scope within a document corpus. These contexts act as attractors for clustering documents that are semantically related to each other. Once clustered, the documents are organized into a minimum spanning tree so that the topical similarity of adjacent documents within this structure can be assessed. The pre-defined categories from three different document category sets are used to assess the quality of CDC in terms of its ability to group and structure semantically related documents given the contexts. Quality is evaluated based on two factors, the category overlap between adjacent documents within a cluster, and how well a representative document categorizes all the other documents within a cluster. As the RCV1 collection was collated in a time ordered fashion, it was possible to assess the stability of clusters formed from documents within one time interval when presented with new unseen documents at subsequent time intervals. We demonstrate that CDC is a powerful and scaleable technique with the ability to create stable clusters of high quality. Additionally, to our knowledge this is the first time that a collection as large as RCV1 has been analyzed in its entirety using a static clustering approach.     

Promoting deep learning in a teacher education programme through self- and peer-assessment and feedback

The incorporation of self- and peer-assessment and feedback has significant potential as a pedagogical strategy to promote deep learning in project based coursework. This study examined the impact of a deeper approach to learning on pre-service teachers’ critical thinking and metacognitive skills. It also examined the impact on student learning outcomes within a project based module with a significant design element. Forty-seven students participated in the pilot of an online peer feedback system. Results suggest that the quality of students’ reflections through peer feedback and overall satisfaction with the module remained high despite students’ citing a preference for instructor feedback. The data also indicate that the incorporation of self- and peer-assessment and feedback resulted in higher quality learning outcomes and enhanced critical thinking skills.     

The development of Canadian student's political conceptions

Two major considerations underlie a set of proposals, outlined in this paper, for a different technology of investigating the psychology of political development in Canadian children and adolescents. First, the paper deals with Hodgetts' (1968) findings, which point out a clear need for further study of the development of political orientations of Canadian students. Second, the paper surveys the typically normative, atheoretical, and methodologically constraining approaches of American investigations of political development in children. Recommendations are advanced to facilitate theory-building and methodological improvement in future investigations. Specific suggestions deal with the construction of an instrument that will elicit scorable information.

---

Résumé Il y a deux considérations principales qui forment la base de propositions présentées dans cet article visant à une technologie différente d'étudier la psychologie du développement politique des enfants et des adolescents canadiens. Premièrement, cet article traite des conclusions de Hodgetts (1968) qui signalent la nécessité d'une étude plus poussée du développement des orientations politiques des étudiants canadiens. Deuxièmement, cet article examine les approches typiquement normatives et nonthéoriques des études américaines du développement politique chez les enfants; cette manière d'aborder le problème est restrictive du point de vue méthodologique. On fait des recommandations afin de faciliter la construction des théories et le progrès méthodologique dans les études à venir. Des suggestions spécifiques touchent à la construction d'un instrument qui obtiendra des renseignements qu'on peut noter.

---

---

OISE     

Trait Understanding or Evaluative Reasoning? An Analysis of Children''s Behavioral Predictions

This study tested the hypothesis that in predicting the future behavior of an actor, older children rely on trait inferences, whereas younger children rely on global, evaluative inferences. Vignettes depicting actors engaging in trait-relevant behaviors were presented to 5- and 6-year-olds (N = 67) and 9- and 10-year-olds (N = 71). For each actor, children made predictions of future behavior, evaluated the goodness and badness of the actor, and rated each actor on a relevant trait. A mediational analysis found that the behavioral predictions of older children were mediated solely by trait ratings, whereas those of younger children were mediated by evaluative ratings. Furthermore, unlike older children, younger children made trait-like predictions only when they made an evaluation of the actor. These results suggest that young children utilize evaluative reasoning when making behavioral predictions, and therefore rely on an inferential process that is distinct from that of older children.