Statistical Models for Monolingual and Bilingual Information Retrieval

This work reviews information retrieval systems developed at ITC-irst which were evaluated through several tracks of CLEF, during the last three years. The presentation tries to follow the progress made over time in developing new statistical models first for monolingual information retrieval, then for cross-language information retrieval. Besides describing the underlying theory, performance of monolingual and bilingual information retrieval models are reported, respectively, on Italian monolingual tracks and Italian-English bilingual tracks of CLEF. Monolingual systems by ITC-irst performed consistently well in all the official evaluations, while the bilingual system ranked in CLEF 2002 just behind competitors using commercial machine translation engines. However, by experimentally comparing our statistical topic translation model against a state-of-the-art commercial system, no statistically significant difference in retrieval performance could be measured on a larger set of queries.     

Combined Approach to Multi-Informant Data Using Latent Factors and Latent Classes: Trifactor Mixture Model

Although collecting data from multiple informants is highly recommended, methods to model the congruence and incongruence between informants are limited. Bauer and colleagues suggested the trifactor model that decomposes the variances into common factor, informant perspective factors, and item-specific factors. This study extends their work to the trifactor mixture model that combines the trifactor model and the mixture model. This combined approach allows researchers to investigate the common and unique perspectives of multiple informants on targets using latent factors and simultaneously take into account potential heterogeneity of targets using latent classes. We demonstrate this model using student self-rated and teacher-rated academic behaviors (N = 24,094). Model specification and testing procedures are explicated in detail. Methodological and practical issues in conducting the trifactor mixture analysis are discussed.     

The influence of test‐based accountability policies on early elementary teachers: School climate,environmental stress,and teacher stress

This study examined the potential influence of test‐based accountability policies on school environment and teacher stress among early elementary teachers. Structural equation modeling of data from 541 kindergarten through second grade teachers across three states found that use of student performance on high‐stakes tests to evaluate teachers indirectly was related to teachers’ professional investment via test stress in the environment. Although students in kindergarten through second grade do not take high‐stakes assessments, early elementary teachers reported high levels of stress associated with test‐based accountability policies. This study provides data across multiple states that test‐based accountability policies may have negative influences on school environment and teacher stress among early elementary teachers. Implications for practice and research are discussed.     

Transfers of user process innovations to process equipment producers: A study of Dutch high-tech firms

A detailed survey of 498 high technology small and medium-sized enterprises in the Netherlands shows process innovation by user firms to be common practice. Fifty-four percent of these firms reported developing entirely novel process equipment or software for their own use and/or modifying these, both at significant private expense. Twenty-five percent of the user innovations in our sample were transferred to commercializing producer firms. Many transfers were made without any direct compensation. Very importantly from the perspective of effective diffusion of user innovations, innovations with higher commercial potential - and more general appeal for users - are much more likely to be transferred to producers. The pattern we document of frequent innovation by individual user firms at substantial cost, followed in many cases by voluntary, no-charge information spillovers to producers, suggests that “open source economics” may be a general pattern in the economy.     

“Pyramiding: Efficient search for rare subjects”

The need to economically identify rare subjects within large, poorly mapped search spaces is a frequently encountered problem for social scientists and managers. It is notoriously difficult, for example, to identify “the best new CEO for our company,” or the “best three lead users to participate in our product development project.” Mass screening of entire populations or samples becomes steadily more expensive as the number of acceptable solutions within the search space becomes rarer.The search strategy of “pyramiding” is a potential solution to this problem under many conditions. Pyramiding is a search process based upon the idea that people with a strong interest in a topic or field tend to know people more expert than themselves. In this paper we report upon four experiments empirically exploring the efficiency of pyramiding searches relative to mass screening. We find that pyramiding on average identified the most expert individual in a group on a specific topic with only 28.4% of the group interviewed - a great efficiency gain relative to mass screening. Further, pyramiding identified one of the top 3 experts in a population after interviewing only 15.9% of the group on average. We discuss conditions under which the pyramiding search method is likely to be efficient relative to screening.     

A theoretical framework for productive education

having held various posts specializing in educational administration in different private institutions in California. Author of several articles dealing with the issue of productive work combined with academic study.     

Visual-spatial strength in dyslexia: rapid discrimination of impossible figures

It has often been conjectured that visual-spatial talents accompany dyslexia. But do individuals with dyslexia have unusual visual-spatial talents? Individuals with dyslexia were compared to controls on two computer-based visual-spatial tasks in two studies. In accordance with a diverging ability hypothesis, it was predicted that the dyslexia group would perform better on a global task, identifying impossible figures, and worse on a feature-oriented task, matching figures. These hypotheses were partially supported. On the global task, the dyslexia group was faster but no more accurate than the control group at identifying impossible objects. On the feature-oriented task, the control group outperformed the dyslexia group. The results indicate that the diverging ability hypothesis of dyslexia merits further investigation.     

An SEM Approach for the Evaluation of Intervention Effects Using Pre-Post-Post Designs

A directly applicable latent variable modeling procedure for classical item analysis is outlined. The method allows one to point and interval estimate item difficulty, item correlations, and item-total correlations for composites consisting of categorical items. The approach is readily employed in empirical research and as a by-product permits examining the latent structure of tentative versions of multiple-component measuring instruments. The discussed procedure is straightforwardly utilized with the increasingly popular latent variable modeling software Mplus, and is illustrated on a numerical example.     

Mediation Analysis in a Latent Growth Curve Modeling Framework

This article presents several longitudinal mediation models in the framework of latent growth curve modeling and provides a detailed account of how such models can be constructed. Logical and statistical challenges that might arise when such analyses are conducted are also discussed. Specifically, we discuss how the initial status (intercept) and change (slope) of the putative mediator variable can be appropriately included in the causal chain between the independent and dependent variables in longitudinal mediation models. We further address whether the slope of the dependent variable should be controlled for the dependent variable's intercept to improve the conceptual relevance of the mediation models. The models proposed are illustrated by analyzing a longitudinal data set. We conclude that for certain research questions in developmental science, a multiple mediation model where the dependent variable's slope is controlled for its intercept can be considered an adequate analytical model. However, such models also show several limitations.