Résumé : L’administration de la médication en centre d’hébergement (CH) constitue une activité qui amène l’infirmière à résoudre des problèmes complexes dans un contexte d’incertitude. Elle exige un raisonnement clinique rigoureux pour lequel il n’existe pas d’instrument de mesure valide et fidèle. Le but de cette étude était d’élaborer un test de concordance de script (TCS) pour mesurer le raisonnement clinique infirmier relatif à l’usage optimal des médicaments en CH et pour documenter sa validité et sa fidélité. Après l’optimisation, le TCS comprenait 55 items répartis dans 20 vignettes cliniques. Il a été mis à l’essai auprès d’étudiantes en sciences infirmières (n=76) et d’infirmières expertes (n=10). Le test a été rempli une seconde fois (test-retest) par 35 étudiantes. La moyenne des scores obtenue par les étudiantes novices est de 55,6 ± 7,7 tandis que celle obtenue par les infirmières expertes est de 63,7 ± 8,0. Les tests statistiques appliqués ont démontré une différence significative dans la moyenne des deux groupes (p < 0,02). La valeur du coefficient alpha de Cronbach de l’ensemble du TCS est de 0,65. Un coefficient de corrélation intraclasse (CCI) a été calculé pour évaluer la stabilité temporelle de l’instrument de mesure. Son résultat est de 0,618 (p = 0,000). Le TCS conçu dans le cadre de cette étude a besoin d’être modifié afin d’obtenir un meilleur indice de fidélité. Les autres résultats statistiques démontrent que le TCS est capable de distinguer les niveaux d’expertise. Cette étude vient s’ajouter aux autres études ayant évalué le raisonnement clinique et confirme l’importance de suivre les directives fondées sur les données probantes pour développer un TCS. / Abstract : The administration of prescribed medication in long-term care facilities is a task that requires the nurse to solve complex problems in a context of uncertainty. It demands a rigorous clinical reasoning for which there is no valid and reliable measurement tool. The purpose of this study was to develop a script concordance test (SCT) to measure nursing clinical reasoning regarding the optimal use of medication in long-term care facilities and to document its validity and reliability. After optimization, the SCT was made up of 55 items divided in 20 clinical vignettes. It has been tested on nursing students (n = 76) and expert nurses (n = 10). The test was completed a second time (test re-test) by 35 students. The mean score obtained by the novice students is 55.6 ± 7.7 while the one obtained by the expert nurses is 63.7 ± 8.0. The applied statistical tests showed a significant difference in the average of the two groups (p < 0.02). The Cronbach alpha coefficient for the entire SCT is 0.65. An inter-class correlation coefficient (ICC) was calculated to assess the temporal stability of the measuring tool. The result is 0.618 (p = 0.000). The SCT developed as part of this study needs to be modified in order for it to be more reliable. The other statistical results demonstrate that the SCT is able to differentiate between levels of expertise. This study adds to previous studies that evaluated clinical reasoning and confirms the importance of following guidelines that rely on evidence-based data to develop a SCT.

Thesis (MEd (Curriculum sStudies))--University of Stellenbosch, 2007. / The TRAC PAC is a micro computer-based laboratory that allows learners to collect real-time data about a particular event and then displays the information graphically. It was brought to South Africa from the United States of America in an attempt to increase the low number of learners from previously disadvantaged communities entering the Science, Engineering and Technology fields. Anecdotal evidence has shown that the TRAC PAC has not been optimally utilised in classrooms. Subsequently a TRAC laboratory was established at Stellenbosch University and hence this study, which focused on identifying factors that would contribute towards the optimal use of the TRAC PAC. A qualitative case study research method was used, which relied on different techniques to gather data on how the TRAC PAC is used in classrooms and at the TRAC laboratory. The analysis of this data was largely an intuitive process; it relied on the development of categories which provided insights on the advantages and disadvantages of using the TRAC PAC. The thesis concludes with a number of recommendations that can lead to the optimal use of the TRAC PAC. One of the findings of this study was that even grade 12 Physical Science learners lacked certain basic skills such as the ability to take accurate measurements. Addressing this lack amongst South African learners is also a priority of the National Education Department, following its adoption of an outcomes-based education approach. The design of a detailed, well structured series of activities that addresses the required educational outcomes should result in the optimal use of the TRAC PAC.

No description available.

It is generally accepted that the Marginal Value Theorem (MVT) describes optimal foraging strategies. Some research findings, however, indicate that in natural conditions foragers not always behave according to the MVT. To address this inconsistency, in a series of computer simulations, I examined the behaviour of four types of foragers having specific foraging efficiencies and using the MVT and alternative strategies in 16 simulated landscapes in an ideal environment (no intra- and inter-species interactions). I used data on elk (Cervus elaphus) to construct the virtual forager. Contrary to the widely accepted understanding of the MVT, I found that in environments with the same average patch quality and varying average travel times between patches, patch residence times of some foragers were not affected by travel times. I propose a mechanism responsible for this observation and formulate the perfect forager theorem (PFT). I also introduce the concepts of a foraging coefficient (F) and foragers hub (), and formulate a model to describe the relationship between the perfect forager and other forager types. I identify situations where a forager aiming to choose an optimal foraging strategy and maximize its cumulative consumption should not follow the MVT. I describe these situations in a form of a mathematical model. I also demonstrate that the lack of biological realism and environmental noise are not required to explain the deviations from the MVT observed in field research, and explain the importance of scale in optimal foraging behaviour. I also demonstrate that smart foraging, which is a set of rules based on key ecological concepts: the functional response curve (FRC), satisficing, the MVT, and incorporates time limitations, should allow for fitness maximization. Thus, it should be an optimal behavior in the context of natural selection. I also demonstrate the importance of the FRC as a driver for foraging behaviors and argue that animals should focus more on increasing the slope of their FRC than on choosing a specific foraging strategy. Natural selection should, therefore, favor foragers with steep FRC. My findings introduce new concepts in behavioural ecology, have implications for animal ecology and inform wildlife management.