There exists a continuous need to promote better Science Technology Engineering and Mathematics (STEM) education at the younger students. To satisfy this need hands-on laboratory assignments and inquiry learning projects are widely accepted as appropriate approaches. One key issue for both approaches is the effective and adaptive data logging. This article describes the development of educational datalogger devices, using open source software and hardware which can be used to collect, present and save data for both offline and online analysis. The novelty of the proposed devices lies on the fact the presented implementations are not dedicated devices bind to specific features but they can be seen as educational datalogger platforms which are expandable and adaptive to students’ needs in a minimum cost since they are based in open source solutions.

The aim of this work is the comparison of two well-known examination methods, the first consisted of multiple-choice questions (MCQs) and the second based on constructed-response questions (CRQs). During this research MCQ and CRQ tests were created for examining the undergraduate engineering module of “project management” and were given to a group of students. Computers and a special software were used for supporting the process. During the first part the examinees had to answer a set of CRQs. Afterwards, they had to answer a set of MCQs. Both sets covered the same topics and had the same level of difficulty. The second method (MCQs) is more objective in terms of grading, though it may conceal an error in the final formulation of the score when a student gives an answer based on an instinctive feeling. For eliminating this fact a set of MCQs pairs was composed taking care that each question of the pair addressed the same topic in a way that similarity would not be evident to a student who did not possess adequate knowledge. By applying a suitable scoring rule to the MCQs, very similar results are obtained when comparing these two examination methods

This paper presents a novel scheme for the estimation of large earthquake event occurrence based on radial basis function (RBF) neural network (NN) models. The input vector to the network is composed of different seismicity rates between main events, which are easy to calculate in a reliable manner. Training of the NNs is performed using the powerful fuzzy means training algorithm, which, in this case, is modified to incorporate a leave-one-out training procedure. This helps the algorithm to account for the limited number of training data, which is a common problem when trying to model earthquakes with data-driven techniques. Additionally, the proposed training algorithm is combined with the Reasenberg clustering technique, which is used to remove aftershock events from the catalog prior to processing the data with the NN. In order to evaluate the performance of the resulting framework, the method is applied on the California earthquake catalog. The results show that the produced RBF model can successfully estimate interevent times between significant seismic events, thus resulting to a predictive tool for earthquake occurrence. A comparison with a different NN architecture, namely, multilayer perceptron networks, highlights the superiority of the proposed approach.