Ein interaktionsbasiertes Modell für den objektorientierten Integrations- und Regressionstest

Zusammenfassung.   Bei der iterativen, inkrementellen Software-Entwicklung werden Klassen sukzessive spezifiziert, implementiert bzw. ge?ndert oder erweitert, klassengetestet und zu einem ausführbaren Anwendungssystem integriert. Integrations- und Regressionstest stellen dabei besonders wichtige und h?ufig ausgeführte Testaktivit?ten dar, die zudem aufgrund der vielf?ltigen Interaktionsm?glichkeiten in objektorientierten Programmen sehr komplex sind. Integrations- und Regressionstest werden in der Literatur stets separat behandelt, obwohl sie durchaus ?hnliche Teilaktivit?ten beinhalten. Bestehende Ans?tze zum objektorientierten Integrationstest betrachten entweder strukturelle oder verhaltensorientierte Systembeschreibungen. In diesem Beitrag stellen wir das Klassen-Botschaftsdiagramm (KBD) vor, ein Struktur und Verhalten objektorientierter Programme gleicherma?en berücksichtigendes interaktionsbasiertes (Test-)Modell. Wir skizzieren Algorithmen zur ?nderungsanalyse, Ableitung einer Integrationsstrategie und Auswahl von Regressionstestf?llen nach der Modifikation bestehender Klassen. Abschlie?end werden die Ergebnisse einiger mit Smalltalk-80 Klassen durchgeführter Experimente zur Evaluierung des Verfahrens vorgestellt. Eingegangen am 14. Dezember 1999 / Angenommen am 19. Juni 2000     

Exploring educators’ professional learning ecologies in a blended learning environment

In recent years the concept of learning ecology has been interpreted in different ways to highlight its potential for learning and professional development. However, there are several aspects that still deserve to be understood such as, eg, the role that the different contexts come to play within a learning ecology. For this purpose, this paper presents and discusses the results of a study exploring the learning ecologies of educators in a blended programme delivered in higher education. The participants in the study were 85 professionals in the field of education aged between 28 and 58 years. The research design adopted here can be defined as a convergent mixed method where both quantitative and qualitative data were collected through an e-portfolio. The study found that educators in the sample had different postures, of whom some more aware of their professional roles or Self than others. Moreover, different professional identities seemed to be linked to different learning ecologies, where elements like formal or informal education, cultural artefacts, personal experiences, colleagues or work experiences played a different role. Formal education emerged as having a questionable function, while colleagues came up as a crucial resource.     

An Analysis of College Choice Information Provided on Graduate Program Websites: Implications for Improving Applicant Diversity in Educational Measurement

There has long been a concern about the lack of representation of ethnic minorities in the field of educational measurement. As previous research has shown that graduate programs primarily rely on their websites for recruiting efforts, the objective of this study was to conduct a content analysis of all U.S. educational measurement program websites to evaluate the availability of college choice information found to be useful for underrepresented ethnic minority applicants. In terms of program climate, results revealed that less than 10% of programs directly encouraged ethnic minorities to apply or included an antidiscrimination statement with regard to application review on their websites. Moreover, only a few program websites indicated the availability of flexible programming—previously found to be important for underrepresented ethnic minority students—such as part‐time options (16%), evening courses (10%), and online course/program availability (8%). Recommendations for how measurement programs can improve their websites to include desirable college choice information for underrepresented ethnic minority applicants are discussed.     

Measuring scientific performance of public research units for strategic change

This research presents a new metrics to measure and assess the scientific performance of public research institutes, which improves models based on standard multivariate techniques. These models called Research Lab Evaluation (RELEV) adjusted are successfully applied to Italian public research institutes, operating in five scientific fields. In addition, the paper presents a comparison between this method and the Data Envelopment Analysis to show some analogies and differences in the results.     

Utility and Reliability of an App for the System for Observing Play and Recreation in Communities (iSOPARC®)

The app for the System for Observing Play and Recreation in Communities (iSOPARC®) was developed to enhance System for Observing Play and Recreation in Communities data collection and management. The study aim was to examine the usability and inter-rater reliability of iSOPARC®. Trained observers collected data in 16 park areas in two Latin America cities (Curitiba, Brazil; Bogotá, Colombia) using both the original System for Observing Play and Recreation in Communities (mechanical counter) and new iSOPARC® (electronic counter) methods. A total of 8,707 park users (3,958 females and 4,749 males) were observed. Reliabilities among observers were determined using the intraclass correlation coefficient and relative overall agreement (%) in three ways: (a) mechanical versus mechanical, (b) mechanical versus electronic, and (c) electronic versus electronic. When both used mechanical counters comparisons overall agreement was over 78% (intraclass correlation coefficient = .982; p < .001) for variables in all analysis. When one used a mechanical counter and the other used electronic (iSOPARC®) comparisons overall agreement was higher for age (84%; p < .001) than for physical activity (83%; p < .001) variables and average intraclass correlation coefficient was .965. Using both used iSOPARC® comparisons overall agreement was 81% to physical activity and 83% to age group. Results suggest iSOPARC® can be used reliably for examining variables previously validated in System for Observing Play and Recreation in Communities.     

Interrupting commemoration: thinking with art,thinking through the strictures of Argentina's Espacio para la memoria

Recently, a few buildings within the Espacio para la memoria in Buenos Aires have been designated as a UNESCO Centre where, amongst other educational activities, evidentiary materials of the past repression are to be stored and displayed. Another building in the complex houses a Community Centre operated by the Mothers of the Plaza de Mayo, where the mandate is to learn about and continue the revolutionary ideals of those disappeared during the repression. Amid these commemorative projects, designating a space for exhibiting art presents a significant opportunity for posing difficult questions, which go beyond the terms of information and idealization. Through a close reading of Diana Dowek's art exhibit, ‘A Long March’, the paper invites us to consider how art can draw out a particular haunting temporality that holds the present open to provocative questions that are difficult to pose and sustain amid the strictures of this commemorative site.     

A Longitudinal Analysis of Doctoral Graduate Supply in the Educational Measurement Field

For nearly three decades, researchers have been concerned that the educational measurement field is not producing enough graduate‐level specialists to meet the growing demand driven by the increased use of educational assessments in the United States. This study examined the supply‐side aspect of the proposed labor shortage by relying on data from the National Science Foundation's Survey of Earned Doctorates collected between 1997 and 2016. Over the 20 years examined, measurement programs produced 3,124 doctoral graduates, and across this time span, the annual production of graduates nearly doubled. This supply expansion can largely be attributed to the increase in the number of international graduates, which outpaced the annual growth rate of domestic PhD recipients by 156%. Moreover, 85% of graduates were found to either self‐identify as White or Asian. Less than 10 Hispanic and no more than 20 Black graduates were produced in any of the years examined. Of the 76% of graduates that reported having a job offer or accepted a position upon graduation, most entered the academy despite the overall average starting salary ($59,484) being considerably lower than the starting salary for their counterparts entering industry ($84,918), government ($69,970), or other educational institutions ($81,428).