Disaggregated Effects of Device on Score Comparability

The use of tablets for large‐scale testing programs has transitioned from concept to reality for many state testing programs. This study extended previous research on score comparability between tablets and computers with high school students to compare score distributions across devices for reading, math, and science and to evaluate device effects for gender and ethnicity subgroups. Results indicated no significant differences between tablets and computers for math and science. For reading, a small device effect favoring tablets was found for the middle to lower part of the score distribution. This effect seemed to be driven by increases in performance for male students when testing on tablets. No interactions of device with ethnicity were observed. Consistent with previous research, this study provides additional evidence for a relatively high degree of comparability between tablets and computers.     

Persistence to Graduation among At‐Risk Criminal Justice and Criminology Students: Experiences in Intervening with a Sophomore Cohort∗

A Seminar targeting academically at‐risk sophomores was developed and introduced in 2002 to address weak connectedness to the university and poor academic abilities. We measured changes in approaches taken to college studies by those at‐risk sophomores who enrolled in the course, as well as employed a longitudinal, quasi‐experimental design to track this experimental group (n = 29) and the remainder of the at‐risk cohort as a comparison group (n = 36). Our preliminary findings suggest that the Seminar has had a variety of positive outcomes, the most impressive impact being that the experimental group had dramatically better retention and graduation rates than the comparison group.     

The Rise of Creative Writing Programmes

Creative writing courses are growing in universities and outside them. Publishers and agents even turn to them now as sources of promise and talent. This article describes a particularly successful MA in creative writing at Goldsmiths College in London and makes a case for the popularity and the usefulness to universities, to aspiring writers and to publishers of such courses.     

Integrierte Entwicklung von Automotive-Software mit AutoFocus

Zur Beherrschung der komplexen vernetzten und verteilten Funktionen von Automotive-Software ist eine Beschreibung des zu erstellenden Systems auf verschiedenen Abstraktionsebenen und schrittweise überg?nge zwischen diesen Ebenen notwendig. Neben der Definition geeigneter Ebenen werden zur Unterstützung echtzeitkritischer Systemanteile ein einheitliches Berechnungsmodell, ebenenspezifische Beschreibungstechniken, sowie methodische Regeln für diese Abstraktionsebenen eingeführt und in den Werkzeugprototypen AutoFocus integriert.     

Dimensions of variability in embedded operating systems

Design, implementation, and re-engineering of operating systems are still an ambitious undertaking. Despite, or even because, of the long history of theory and practice in this field, adapting existing systems to environments of different conditions and requirements as originally specified or assumed, in terms of functional and/or non-functional respects, is anything but simple. Especially this is true for the embedded systems domain which, on the one hand, calls for highly specialized and application-aware system abstractions and, on the other hand, cares a great deal for easily reusable implementations of these abstractions. The latter aspect becomes more and more important as embedded systems technology is faced with an innovation cycle decreasing in length. Software for embedded systems needs to be designed for variability, and this is in particular true for the operating systems of this domain. The paper discusses dimensions of variability that need to be considered in the development of embedded operating systems and presents approaches that aid construction and maintenance of evolutionary operating systems. CR subject classification  C.3; D.2.11; D.2.13; D.4.7     

Efficient reverse k-nearest neighbor estimation

The reverse k-nearest neighbor (RkNN) problem, i.e. finding all objects in a data set the k-nearest neighbors of which include a specified query object, has received increasing attention recently. Many industrial and scientific applications call for solutions of the RkNN problem in arbitrary metric spaces where the data objects are not Euclidean and only a metric distance function is given for specifying object similarity. Usually, these applications need a solution for the generalized problem where the value of k is not known in advance and may change from query to query. In addition, many applications require a fast approximate answer of RkNN-queries. For these scenarios, it is important to generate a fast answer with high recall. In this paper, we propose the first approach for efficient approximative RkNN search in arbitrary metric spaces where the value of k is specified at query time. Our approach uses the advantages of existing metric index structures but proposes to use an approximation of the nearest-neighbor-distances in order to prune the search space. We show that our method scales significantly better than existing non-approximative approaches while producing an approximation of the true query result with a high recall.