Dynamische Probleml?sekompetenz

This article is about the importance and validity of dynamic problem solving (DPS) in predicting technical problem solving performances. In addition to DPS, fluid intelligence and technical knowledge were included in the study. Assessing DPS and fluid intelligence simultaneously also allowed for testing the empirical distinction between both constructs. Results are based on data of a sample of car mechatronics (n?=?129) and electronics technicians (n?=?88) and showed that DPS and fluid intelligence were empirically separable. In the first sample (electronics technicians) DPS explained technical problem solving performance best, although fluid intelligence showed incremental validity too. In the second sample (car mechatronics) fluid intelligence was weakly related to the criterion whereas DPS was not at all. When technical knowledge was added to the prediction, both DPS and fluid intelligence did not predict technical problem solving performance. However, in the sample of the electronics technicians fluid intelligence indirectly affected the criterion through technical knowledge. A discussion of the somewhat contradictory results, the different results patterns in the two samples and a theoretical explanation of the results are provided.     

Trust in results

Educational Assessment, Evaluation and Accountability -     

Learning from text: knowing the test format enhanced metacognitive monitoring

In an experiment with 56 young adults, the hypothesis was tested that information about the format of an anticipated test improves metacognitive monitoring. Half of the participants were informed about the format of the test before they started studying a text about human genetics. The other half of the sample received the same information after studying the text. All participants then answered 15 true-false inference items about the contents of the text and judged their confidence in the correctness of each answer. Whereas experimental and control group did not differ in the number of correct answers, the confidence judgments in the experimental group were more accurate and discriminated better between correct and incorrect answers than the control participants’ judgments. Furthermore, the informed participants’ discrimination performance correlated positively with their domain-related prior knowledge. The results extend earlier findings concerning the role of the test format for monitoring processes.     

Learning from scientific texts: personalizing the text increases transfer performance and task involvement

In an experiment with 65 high-school students, we tested the hypothesis that personalizing learning materials would increase students’ learning performance and motivation to study the learning materials. Students studied either a 915-word standard text on the anatomy and functionality of the human eye or a personalized version of the same text in which 60 definite articles (e.g., “the eye”) were replaced with 60 second-person possessive pronouns (e.g., “your eye”). Afterwards, participants answered comprehension and transfer questions. One week later, the participants were asked to restudy the text and to answer the same questions again with the aim to improve their performance. In the personalized text condition, students showed higher transfer performance, spent more time on restudying the text, and reported being more motivated than students in the standard text condition. However, only duration of restudying (not self-reported motivation) mediated the effect of personalization on transfer performance.     

Measuring teaching quality: some key issues

Educational Assessment, Evaluation and Accountability -     

Analgesics use in competitive triathletes: its relationship to doping and on predicting its usage

The two major objectives of this study were (i) to assess variables that predict the use of analgesics in competitive athletes and (ii) to test whether the use of analgesics is associated with the use of doping. A questionnaire primarily addressing the use of analgesics and doping was distributed among 2,997 triathletes. Binary logistic regression analysis was performed to predict the use of analgesics. Moreover, the randomised response technique (RRT) was used to estimate the prevalence of doping in order to assess whether users of analgesics have a higher potential risk for doping than non-users. Statistical power analyses were performed to determine sample size. The bootstrap method was used to assess the statistical significance of the prevalence difference for doping between users and non-users of analgesics. Four variables from a pool of 16 variables were identified that predict the use of analgesics. These were: “version of questionnaire (English)”, “gender (female)”, “behaviour in case of pain (continue training)”, and “hours of training per week (>12 h/week)”. The 12-month prevalence estimate for the use of doping substances (overall estimate 13.0%) was significantly higher in athletes that used analgesics (20.4%) than in those athletes who did not use analgesics (12.4%). The results of this study revealed that athletes who use analgesics prior to competition may be especially prone to using doping substances. The predictors of analgesic use found in the study may be of importance to prepare education material and prevention models against the misuse of drugs in athletes.     

A learning optimizer for a federated database management system

Zusammenfassung Anfrageoptimierer in modernen Datenbanksystemen (DBS) verwenden ein Kostenmodell, um für eine Anfrage einen effizienten Zugriffsplan aus der Menge aller m?glichen Zugriffspl?ne auszuw?hlen. Die Genauigkeit, mit der diese Kosten gesch?tzt werden, h?ngt stark davon ab, wie genau Statistiken über die Datenverteilung die ihnen zugrunde liegenden Daten modellieren. Veraltete Statistiken, oder ungültige Annahmen im statistischen Modell, führen h?ufig zur Auswahl suboptimaler Zugriffspl?ne und dadurch zu Defiziten bei der Ausführungsgeschwindigkeit von Datenbankanfragen. F?derierte Systeme müssen auf dem koordinierenden DBS Statistiken über verteilte Daten unterhalten, um effiziente Zugriffspl?ne zu entwickeln, wenn Daten aus verschiedenen Datenquellen zusammengeführt werden. Fehlerhafte Statistiken verursachen in diesem Fall schlechte Lastenverteilung sowie zus?tzlichen Kommunikationsaufwand und Netzlast. Die Synchronisation dieser zus?tzlichen Statistiken mit den Statistiken der entfernten Datenquellen ist problematisch, da Datenquellen v?llig unabh?ngig voneinander sind und ihre Statistiken eventuell nicht externalisieren oder inkompatible bzw. andere statistische Modelle verwenden als das koordinierende DBS. Im folgenden zeigen wir eine Erweiterung der Feedback-Schleife von DB2s lernendem Optimierer (LEO) auf f?derierte Architekturen, die es LEO erm?glicht, automatisch Statistiken über Daten auf entfernten Datenquellen zu lernen und zu verwalten. Wir diskutieren mehrere M?glichkeiten, Feedback für f?derierte Anfragen zu erhalten und stellen einen L?sung vor, die lokales Feedback verwendet, und einen Stichprobenprozess (statistical sampling) zu steuren, der die ben?tigten entfernten Daten basierend auf dem Feedback effizient ermittelt. Mittels einer detaillierten Leistungsanalyse der vorgestellten M?glichkeiten im Rahmen einer Fallstudie belegen wir, dass der potenzielle Geschwindigkeitszuwachs für Datenbankanfragen sehr hoch ist, w?hrend die Zusatzbelastung durch LEO in moderaten Grenzen liegt. CR Subject Classification H.2.4,H.2.5,D.3.4     

Music in the air: estimating the social return to cultural amenities

If being around smart people makes us smarter and more productive, what can regions do to attract smart people? This paper considers endogenous cultural amenities as a location factor for high-skilled workers. To overcome selection in the provision of cultural amenities, we exploit variation in contemporaneous cultural amenities that is explained by the path-dependence of historical agglomerations of the cultural activities. To assess spillovers from high-skilled workers attracted by cultural amenities, we use a 1% sample drawn from the population of all West German workers under social security during the period 1975–2010. This panel of individual observations allows us to compare wages of similar individuals who work in locations with different levels of high-skilled workers who are attracted by cultural amenities. To account for non-random selection of workers among cities, we use individual-location fixed effects. Our results show that cultural amenities are an important factor in the location decision of high-skilled workers. The positive effect of the local share of high-skilled workers on unskilled, skilled and high-skilled wages indicates strong and productive spillovers.