Classrooms in Southern Europe are traditionally ventilated with manual opening of windows. This is an energy-sparing and perfectly appropriate way of ventilating classrooms when weather conditions are warm, however, as outdoor air temperatures drop, teaching staff and students tend to leave windows closed and, as a consequence, ventilation rates fall leading to poor indoor air quality. To safeguard classrooms' indoor air quality and promote energy conservation, understanding 5 the conditions for which manual window-airing is appropriate is of great relevance. Yet, given the stochastic nature of window-airing, it is difficult to get hold of this understanding. The main objective of this paper is to find out when manual window-airing of classrooms is appropriate. To achieve this objective, four free-running classrooms of a Portuguese public secondary school were monitored during a two-year period. Ventilation rates were determined and it is concluded that 10 manual opening of windows provides appropriate ventilation for outdoor running mean temperatures larger than 16 0 C; and, for the studied classrooms, this translates into appropriate ventilation for approximately a quarter of the academic year. Because of the significance of this result, the paper concludes with a review of the ventilation strategy used in the studied classrooms.1 Keywords. Natural ventilation; free-running classrooms; manual window-airing; window opening 15 patterns; indoor air quality; thermal comfort. Highlights-Ventilation rates in four free-running classrooms are determined during two full academic years.-Manual window-airing of classrooms is appropriate with outdoor running mean temperatures 20 larger than 19 0 C.-For outdoor running mean temperatures between 16 and 19 0 C manual window-airing is still appropriate, but depends on indoor air temperature.-When outdoor running mean temperatures are lower than 16 0 C, regardless of indoor air temperature, manual window-airing is inappropriate. 25-For the studied classrooms, manual window-airing provides appropriate indoor air quality and thermal comfort for 25% of the academic year. 45 sociated with undesirable effects on students productivity (Wargocki and Wyon, 2007a,b;Haverinen-Shaughnessy et al., 2011;Bakó-Biró et al., 2012;Petersen et al., 2015), absenteeism (Shendell et al., 2004 and increased health symptoms (Fraga et al., 2008;Bartlett et al., 2004;Rudnick and Milton, 2003;Mendell and Heath, 2005). Still, in spite of these concerns, along the academic year there are conditions for which natural ventilation is perfectly appropriate, and, from the point of view of 50 users, better than mechanical ventilation alternatives.In fact, mechanical systems have problems too; if not properly balanced and maintained, they are frequently a source of thermal discomfort, noise and can even increase indoor air pollution, as discussed in

This paper describes a method for estimating ventilation rates in a window-aired room considering "in-use" conditions and uncertainty in occupancy (CO 2 release) and in measurements of indoor air CO 2 concentration. Estimates are drawn from indoor air mass conservation principles and an Extended Kalman Filter, serving as state observer. The modelling procedure is described and tested against synthetic time series generated from pseudorandom values of occupancy and ventilation rate. Additionally, data collected from a case study (a window-aired classroom) is used to illustrate how the procedure is applied and its practical interest. Test results confirm the state observer's tracking capability and confirm its ability to reconstruct ventilation rates in the presence of sudden changes caused by window opening/ closing and intermittent occupancy. Results for the case study highlight the benefits to designers and ultimately to occupants, derived from knowledge of ventilation rates in window-aired rooms.

Natural ventilation plays an important role on the thermal and energy performance of a building. The present study aims to analyze the natural ventilation conditions of a dwelling in a Mediterranean climate and their impacts on the thermal and energy performance using an advanced building energy simulation tool. Several multi-zone simulations were carried out. In the summer, the simulations were performed under free-floating conditions, whereas in the winter they were carried out under controlled temperature conditions. In the summer, ventilation scenarios with windows opened during certain periods of time and with or without permanent openings in the facades were analyzed. The existence of permanent openings proved to be an important factor of temperature control by lowering the average indoor zone temperatures during the day. Cross-ventilation also showed to be effective. In the winter, we simulated the existence or absence of permanent openings for room ventilation and their surface area. The results showed that the stack effect plays an important role in the ventilation and that in general it outperforms the wind effect. Sizing permanent openings according to the standard guidelines proved to be adequate in providing the expected ventilation rates on an average basis.

The successful implementation of quality management systems in higher education relies on the ability to address topics meaningful to stakeholders. A topic that is moving to the top of many higher education institutions agendas and is meaningful to students, faculty and management is student dropout. Alongside its social and personal consequences, dropout impairs cost efficiency and the institution's image. This paper shows that in spite of the complex web of factors influencing student dropout, simple models for the identification of at-risk of dropout students can be derived and used to support decision making. The paper starts with an introduction to dropout models, next, the difficulty in implementing quality management systems in higher education is addressed; details about a process for the identification of at-risk students are presented.A case study is used to show that it is possible to identify at-risk students using only academic data and administrative records. Finally, the advantage of including an at-risk student identification process within the framework of a higher education quality management system is discussed.

The aims of this study were to evaluate and compare the performance of dental students in interpreting the temporomandibular joint (TMJ) with magnetic resonance imaging (MRI) scans using two learning methods (conventional and digital interactive learning) and to examine the usability of the digital learning object (DLO). The DLO consisted of tutorials about MRI and anatomic and functional aspects of the TMJ. In 2014, dental students in their final year of study who were enrolled in the elective "MRI Interpretation of the TMJ" course comprised the study sample. After exclusions for nonattendance and other reasons, 29 of the initial 37 students participated in the study, for a participation rate of 78%. The participants were divided into two groups: a digital interactive learning group (n=14) and a conventional learning group (n=15). Both methods were assessed by an objective test applied before and after training and classes. Aspects such as support and training requirements, complexity, and consistency of the DLO were also evaluated using the System Usability Scale (SUS). A significant between-group difference in the posttest results was found, with the conventional learning group scoring better than the DLO group, indicated by mean scores of 9.20 and 8.11, respectively, out of 10. However, when the pretest and posttest results were compared, both groups showed significantly improved performance. The SUS score was 89, which represented a high acceptance of the DLO by the users. The students who used the conventional method of learning showed superior performance in interpreting the TMJ using MRI compared to the group that used digital interactive learning.