Search

In 2001/2002 a study was undertaken to establish whether significant differences existed between the levels of development of several entrepreneurial attributes, as perceived by undergraduate business students from three universities in three different countries. The rationale was that, if entrepreneurial attributes could be identified as more developed in one country than in another, solutions could be provided for developing these attributes in others. The primary objective of this study is to investigate and compare the levels of development of entrepreneurial attributes of undergraduate business students in the present study (2010) to the levels of development reported by undergraduate business students in the 2001/2002 study. Cronbach alpha coefficients were calculated to assess the reliability of the measuring instrument and ttests to establish significant differences. Cohen’s d statistics were calculated to establish practical significance. The findings suggest that the educational environment and entrepreneurship education policy of the Dutch university participating in this study could provide solutions as to how entrepreneurial attributes among students could be developed further.

Previous studies suggest that there is a significant variation in both shape and physical properties from filament to filament in a multifilament superconducting Bi-2223/Ag tape. We are investigating this variation for a tape with a high critical current density (J/sub c/). For this study, we extracted all the filaments from a specially prepared high-quality tape having J/sub c/ (0 T, 77 K) = 45 kA/cm/sup 2/ with 19 disconnected filaments. By attaching current leads and voltage taps to the filaments, we are able to measure electrical properties on a scale of 200-300 /spl mu/m. The experiments show a large variation in filament properties. The conductivity of a filament at room temperature is proportional to its critical current density J/sub c/ at 77 K (self-field) and filaments with a higher transition temperature T/sub c/ do also show a higher J/sub c/. There is also a variation in the magnetic field behavior of the filaments. SEM pictures of the filaments also show large variations. The filaments from the center of the tape, which are thinner and have a larger aspect ratio (and thus a larger silver-filament interface), show better grain alignment and fewer impurities than filaments from the edges.

Previously, Bayesian methods have been proposed for neural networks to solve regression and classification problems. These methods claim to overcome some difficulties encountered in the standard approach such as overfitting. However, an implementation of the full Bayesian approach to neural networks as suggested in the literature applied to classification problems is not easy. In fact we are not aware of applications of the full approach to real world classification problems. In this paper we discuss how the Bayesian framework can improve the predictive performance of neural networks. We demonstrate the effects of this approach by an implementation of the full Bayesian framework applied to two real world classification problems. We also discuss the idea of calibration to measure the predictive performance.

A comparison has been made of the infrared spectra of alumina with molybdenum oxide-alumina in both the oxidized and reduced forms. In the case of molybdena-alumina prepared via adsorption of gaseous MoO2(OH)2, the spectra show that a practically complete monolayer of Mo6+ oxide covers the alumina. After reduction with hydrogen the hydroxyls of the carrier appear. From the reversibility of reduction and oxidation under mild conditions it has been established that the reduced oxide is present as an interrupted monolayer.