Le présent mémoire expose les effets de pratiques d’écriture de textes identitaires plurilingues, soutenues par des ateliers d’expression théâtrale plurilingues, sur le rapport à l’écrit d’élèves immigrants allophones nouvellement arrivés en situation de grand retard scolaire au secondaire, et ce, tout au long d’une recherche-action visant le développement de l’écriture. Puis, ces effets sont comparés avec ceux de pratiques traditionnelles d’enseignement de l’écriture. Un groupe expérimental et un groupe contrôle ont participé à cette étude de cas multiples. Des observations participantes et des entretiens individuels semi-dirigés ont permis de dresser le portrait du rapport à l’écrit des huit élèves participant à notre recherche. Dans ce rapport à l’écrit est considéré l’engagement dans la littératie des élèves, un concept clé à prendre en compte en didactique des langues secondes. Nous proposons ainsi l’étude d’une nouvelle notion : le rapport à l’écrit teinté du concept d’engagement (RÉ+). Les résultats de ce mémoire confirment la pertinence de l’approche novatrice mise en place, qui agit sur les quatre dimensions du RÉ+ (affective, praxéologique, conceptuelle et axiologique). Elle permet notamment aux élèves d’associer l’écriture à des sentiments positifs et les amène à poser un regard réflexif sur la langue associé au développement d’habiletés métacognitives. De tels effets sont différents de ceux de pratiques traditionnelles d’enseignement de l’écriture, qui ne paraissent pas en mesure d’engager pleinement, affectivement et cognitivement, les élèves dans les tâches d’écriture. / This study explores the effects of the writing of multilingual identity texts, supported by multilingual drama expression workshops, on the relationship with writing of French as a second language (FSL) learners. Additionally, we compare these effects with traditional teaching practices for writing. The participants were recent immigrants with limited prior schooling educated at the secondary school level. This multiple case study was performed with both an experimental group and a control group. Participant observations and semi-directed interviews were conducted with each of the eight participants in order to describe their relationship with writing. The main contribution of this research project is that the researcher links a key concept in second language teaching, students’ literacy engagement, to their relationship with writing. Thus a new notion is proposed: the FSL students’ relationship with writing enhanced by literacy engagement (designated with the abbreviation “RÉ+” in French). Results confirm the relevance of this innovative approach which has an influence on the four dimensions of RÉ+: affective, praxeological, conceptual and axiological. The benefits of this approach are that students associate writing with positive feelings. Moreover, they adopt an analytical point of view on language learning, which leads to the development of metacognitive abilities. In essence, the effects on RÉ+ are different from those of traditional writing teaching practices, which, in comparison, seem unable to fully engage students emotionally and cognitively in writing tasks.

Les sociétés sont traditionnellement classées en sociétés de personnes et sociétés de capitaux. L’intuitus personae constitue le fondement sur lequel repose cette distinction. Fort présent dans les sociétés de personnes, l’intuitus personae serait, en théorie, absent dans les sociétés de capitaux. Toutefois, cette distinction classique ne correspond plus, aujourd’hui, à la réalité. Nous assistons, en effet, à la résurgence de la considération de la personne dans les sociétés de capitaux à travers le développement de techniques contractuelles dont le but est de contrôler l’accès à ces sociétés. Ce mouvement de personnalisation a été renforcé par le législateur, d’abord, par la création de la société par actions simplifiée, mais aussi par la reconnaissance, dans les sociétés cotées, de la notion de concert ou encore d’un droit de sortie au profit des actionnaires minoritaires. La consécration de la validité de l’exclusion, qu’elle soit conventionnelle ou le résultat d’un mécanisme spécifique du droit des sociétés comme le retrait obligatoire ou le coup d’accordéon, participe également de ce phénomène. Cette présence a des conséquences majeures sur le fonctionnement de ces sociétés, mais aussi sur le droit des sociétés en général. En effet, d’une part, la reconnaissance de la primauté des relations entre actionnaires sur les capitaux offre une nouvelle perspective de la notion d’affectio societatis et du principe majoritaire. Et d’autre part, la mise en oeuvre de l’intuitus personae dans des sociétés dans lesquelles il est censé être absent affecte la cohérence même du droit des sociétés en anéantissant la summa divisio sur laquelle ce droit est bâti et en dénaturant le choix du pluralisme juridique, fait de longue date par le législateur français, qui commande en présence d’une pluralité de structures offertes aux opérateurs, une distinction de chacune d’elles reposant sur des critères précis et invariables. / Companies are traditionally classified into partnerships and joint stock companies. Intuitus personae is the basis for this distinction. Clearly present in partnership companies, intuitus personae is, in theory, absent in joint stock companies. However, this traditional distinction does not correspond today to reality. We are, in fact, witnessing a resurgence of the consideration of the individual in joint stock companies by means of contractual techniques whose aim is to control access to these companies. This resurgence has been further strengthened by the legislator, first by the creation of the status of simplified stock company and also by the recognition in listed companies of the existence of concerted action and a right to voluntary withdrawal benefiting minority stockholders. The admission of the validity of exclusion, either conventional or resulting from mechanisms specific to company law such as a squeeze out or even an “accordion squeeze”, add to the growth of this phenomenon. This presence of intuitus personae has major implications on the functioning of joint stock companies and on company law, in general. Indeed, on the one hand, the primacy of shareholder relationships over capital offers a new perspective of the notion of affectio societatis and the principle of majority rule. On the other hand, the introduction of intuitus personae into companies where this concept should have been absent affects the coherence of company law by destroying the summa divisio basis on which this law is founded and by distorting the legal pluralism long ago adopted by the French legislator which, in the presence of diverse legal structures, necessitates a distinction amongst them based on precise and invariable criteria.

Cette thèse se décompose en deux parties. Dans un premier temps nous nous intéressons à la sélection bayésienne de variables dans un modèle probit mixte.L'objectif est de développer une méthode pour sélectionner quelques variables pertinentes parmi plusieurs dizaines de milliers tout en prenant en compte le design d'une étude, et en particulier le fait que plusieurs jeux de données soient fusionnés. Le modèle de régression probit mixte utilisé fait partie d'un modèle bayésien hiérarchique plus large et le jeu de données est considéré comme un effet aléatoire. Cette méthode est une extension de la méthode de Lee et al. (2003). La première étape consiste à spécifier le modèle ainsi que les distributions a priori, avec notamment l'utilisation de l'a priori conventionnel de Zellner (g-prior) pour le vecteur des coefficients associé aux effets fixes (Zellner, 1986). Dans une seconde étape, nous utilisons un algorithme Metropolis-within-Gibbs couplé à la grouping (ou blocking) technique de Liu (1994) afin de surmonter certaines difficultés d'échantillonnage. Ce choix a des avantages théoriques et computationnels. La méthode développée est appliquée à des jeux de données microarray sur le cancer du sein. Cependant elle a une limite : la matrice de covariance utilisée dans le g-prior doit nécessairement être inversible. Or il y a deux cas pour lesquels cette matrice est singulière : lorsque le nombre de variables sélectionnées dépasse le nombre d'observations, ou lorsque des variables sont combinaisons linéaires d'autres variables. Nous proposons donc une modification de l'a priori de Zellner en y introduisant un paramètre de type ridge, ainsi qu'une manière de choisir les hyper-paramètres associés. L'a priori obtenu est un compromis entre le g-prior classique et l'a priori supposant l'indépendance des coefficients de régression, et se rapproche d'un a priori précédemment proposé par Gupta et Ibrahim (2007).Dans une seconde partie nous développons deux nouvelles méthodes MCMC basées sur des populations de chaînes. Dans le cas de modèles complexes ayant de nombreux paramètres, mais où la vraisemblance des données peut se calculer, l'algorithme Equi-Energy Sampler (EES) introduit par Kou et al. (2006) est apparemment plus efficace que l'algorithme classique du Parallel Tempering (PT) introduit par Geyer (1991). Cependant, il est difficile d'utilisation lorsqu'il est couplé avec un échantillonneur de Gibbs, et nécessite un stockage important de valeurs. Nous proposons un algorithme combinant le PT avec le principe d'échanges entre chaînes ayant des niveaux d'énergie similaires dans le même esprit que l'EES. Cette adaptation appelée Parallel Tempering with Equi-Energy Moves (PTEEM) conserve l'idée originale qui fait la force de l'algorithme EES tout en assurant de bonnes propriétés théoriques et une utilisation facile avec un échantillonneur de Gibbs.Enfin, dans certains cas complexes l'inférence peut être difficile car le calcul de la vraisemblance des données s'avère trop coûteux, voire impossible. De nombreuses méthodes sans vraisemblance ont été développées. Par analogie avec le Parallel Tempering, nous proposons une méthode appelée ABC-Parallel Tempering, basée sur la théorie des MCMC, utilisant une population de chaînes et permettant des échanges entre elles. / This thesis is divided into two main parts. In the first part, we propose a Bayesian variable selection method for probit mixed models. The objective is to select few relevant variables among tens of thousands while taking into account the design of a study, and in particular the fact that several datasets are merged together. The probit mixed model used is considered as part of a larger hierarchical Bayesian model, and the dataset is introduced as a random effect. The proposed method extends a work of Lee et al. (2003). The first step is to specify the model and prior distributions. In particular, we use the g-prior of Zellner (1986) for the fixed regression coefficients. In a second step, we use a Metropolis-within-Gibbs algorithm combined with the grouping (or blocking) technique of Liu (1994). This choice has both theoritical and practical advantages. The method developed is applied to merged microarray datasets of patients with breast cancer. However, this method has a limit: the covariance matrix involved in the g-prior should not be singular. But there are two standard cases in which it is singular: if the number of observations is lower than the number of variables, or if some variables are linear combinations of others. In such situations we propose to modify the g-prior by introducing a ridge parameter, and a simple way to choose the associated hyper-parameters. The prior obtained is a compromise between the conditional independent case of the coefficient regressors and the automatic scaling advantage offered by the g-prior, and can be linked to the work of Gupta and Ibrahim (2007).In the second part, we develop two new population-based MCMC methods. In cases of complex models with several parameters, but whose likelihood can be computed, the Equi-Energy Sampler (EES) of Kou et al. (2006) seems to be more efficient than the Parallel Tempering (PT) algorithm introduced by Geyer (1991). However it is difficult to use in combination with a Gibbs sampler, and it necessitates increased storage. We propose an algorithm combining the PT with the principle of exchange moves between chains with same levels of energy, in the spirit of the EES. This adaptation which we are calling Parallel Tempering with Equi-Energy Move (PTEEM) keeps the original idea of the EES method while ensuring good theoretical properties and a practical use in combination with a Gibbs sampler.Then, in some complex models whose likelihood is analytically or computationally intractable, the inference can be difficult. Several likelihood-free methods (or Approximate Bayesian Computational Methods) have been developed. We propose a new algorithm, the Likelihood Free-Parallel Tempering, based on the MCMC theory and on a population of chains, by using an analogy with the Parallel Tempering algorithm.

Ce travail de recherche propose une solution aux problèmes inverses probabilistes avec des outils de la statistique bayésienne. Le problème inverse considéré est d'estimer la distribution d'une variable aléatoire non observée X à partir d'observations bruitées Y suivant un modèle physique coûteux H. En général, de tels problèmes inverses sont rencontrés dans le traitement des incertitudes. Le cadre bayésien nous permet de prendre en compte les connaissances préalables d'experts en particulier lorsque peu de données sont disponibles. Un algorithme de Metropolis-Hastings-within-Gibbs est proposé pour approcher la distribution a posteriori des paramètres de X avec un processus d'augmentation des données. A cause d'un nombre élevé d'appels, la fonction coûteuse H est remplacée par un émulateur de krigeage (métamodèle). Cette approche implique plusieurs erreurs de natures différentes et, dans ce travail,nous nous attachons à estimer et réduire l'impact de ces erreurs. Le critère DAC a été proposé pour évaluer la pertinence du plan d'expérience (design) et le choix de la loi apriori, en tenant compte des observations. Une autre contribution est la construction du design adaptatif adapté à notre objectif particulier dans le cadre bayésien. La méthodologie principale présentée dans ce travail a été appliquée à un cas d'étude en ingénierie hydraulique. / This thesis provides a probabilistic solution to inverse problems through Bayesian techniques.The inverse problem considered here is to estimate the distribution of a non-observed random variable X from some noisy observed data Y explained by a time-consuming physical model H. In general, such inverse problems are encountered when treating uncertainty in industrial applications. Bayesian inference is favored as it accounts for prior expert knowledge on Xin a small sample size setting. A Metropolis-Hastings-within-Gibbs algorithm is proposed to compute the posterior distribution of the parameters of X through a data augmentation process. Since it requires a high number of calls to the expensive function H, the modelis replaced by a kriging meta-model. This approach involves several errors of different natures and we focus on measuring and reducing the possible impact of those errors. A DAC criterion has been proposed to assess the relevance of the numerical design of experiments and the prior assumption, taking into account the observed data. Another contribution is the construction of adaptive designs of experiments adapted to our particular purpose in the Bayesian framework. The main methodology presented in this thesis has been applied to areal hydraulic engineering case-study.

This project explores research-based methods for creating an e-learning resource to teach an intellectual skill. Intellectual skills are one of the five domains of learning defined by Robert M. Gagné. The intellectual skill taught via the e-learning resource is the rules of Turkish vowel harmony, a fundamental phonological feature of the Turkish language. The purpose of the e-learning resource is to provide novice learners with a set of tools and strategies with which to approach the future study of Turkish. A literature review was conducted in three areas: Gagné’s instructional design theories, Turkish language learning, and best practices for the design of multimedia e-learning. Two rapid prototypes in the form of Moodle course sites were developed. Guidance for improving the prototypes was sought from experts in instructional design, usability, and computer software. The resulting finished e-learning product is a nine-topic Moodle course based on Gagné’s nine events of instruction. The main Moodle course content teaches the rules of Turkish vowel harmony as an intellectual skill which has been broken down into discriminations (the Turkish vowels), concepts (the Turkish vowel groups) and rules (Turkish vowel harmony). Higher order rules are taught in the form of exceptions to the rules of Turkish vowel harmony found in foreign loan-words in Turkish. Practice is provided in applying the rules of Turkish vowel harmony as a set of word attack skills for approaching Turkish language artifacts found online and in the environment. A comprehensive list of resources relating to learning Turkish, learning about the Turkish language, and learning about learning Turkish, is provided to facilitate future extension and application of what was learned in this course. Each lesson is presented in the form of a Moodle book. Each lesson is followed by an ungraded assessment in the form of an Adobe® PDF quiz. The quizzes and accompanying answer keys are designed to provide guided practice, feedback, and self-assessment to help students extend and apply the lesson material. All lessons were beta-tested to ensure usability and reduction of extraneous cognitive load. The project resulted in the development of a Best Practices Checklist for designing e-learning resources to teach intellectual skills. The Best Practices Checklist, which can be found in Appendix B, was used to develop the outline for e-learning resources to teach intellectual skills in other subject areas and was found to be an effective rapid prototyping and instructional design tool. Insight was gained into the significance of prior knowledge for teaching intellectual skills, and on how to calibrate cognitive load in e-learning design relative to the learner’s prior knowledge of the subject matter being taught.

Owing to past injustices, the South African higher education sector is characterised by inequalities of resource allocation and of learning opportunities. Through the National Qualification Framework (NQF), recognition of prior learning (RPL) was established to address the previous inequalities in higher education and training. RPL can be used as a mechanism to offer non-traditional learners such as workers, adult learners, and community workers access to learning programmes in Library and Information Science (LIS) schools. It can also be used for up-skilling within LIS sector, to enable staff to migrate from paraprofessional to professional roles. LIS schools could possibly use this approach to offer experienced but unqualified library workers opportunities for progressive professional development and career growth. Despite it being a national policy and its obvious benefits, very little is known about RPL implementation in LIS schools in South Africa. This study was conducted to investigate the nature of RPL implementation in LIS schools in South Africa and make recommendations for effective and efficient RPL practice in these schools. The study used the questionnaire as the main data collection tool. In addition, document analysis was used to validate the collected data. The results of the study indicated that there were islands of good RPL practice in LIS schools in South Africa specifically with regard to the aspect of RPL assessment process. However, certain weaknesses were identified in other aspects of RPL implementation in LIS schools including the policy environment, training of personnel conducting RPL assessment and the quality management systems. Among other things, the study recommends that RPL quality management system (QMS) should ideally be driven by the head/chair of the school/department. Furthermore, an integrated student management system should be used to monitor the progression of RPL candidates through the formal academic system post RPL. / Information Science / D. Litt. et Phil. (Information Science)

Thesis Abstract: Chapter I: Regional burden sharing of GHG mitigation policies – A Canadian perspective. The distribution of the burden of cost arising from the reduction of greenhouse gas (GHG) emissions is a contentious issue in policy discussions; more so among regional jurisdictions in the federalist countries with decentralized authorities over environmental regulations. In this setting, often the policy discussions are focused on the distribution of regional emission reduction targets that, in turn, entails negotiations over the distribution of the scarcity rents and the regional transfers of wealth. The allocation of regional emission entitlements is thus a key factor that could hinder the political feasibility of a national GHG mitigation policy. In this paper, we build a multi-region computable general equilibrium (CGE) model of the Canadian economy to assess the implications of different burden sharing rules governing the national GHG abatement policy with a cap-and-trade system of emission permits. In addition to assessing the impacts of traditional regional emissions allocation rules that involve intra-regional transfers of wealth, we consider a particular emission allocation that avoids such transfers, which may be a more palatable option given the context of likely fierce negotiations over the issue. Our results indicate to differing outcomes depending on the allocation policy in use. The CGE framework is also able to shed light on the transmission mechanisms that drive the results underlying the policy options. Chapter II: Endogenous technological change and emission allowances. Given the imminent threat of global warming due to GHG emissions, a number of emission mitigation policies have been proposed in the literature. However, they generally suffer from the classical equity-efficiency trade-off. High costs from equity concerns often render environmental policies politically unattractive and thus hard to implement. Recent advancement in the climate policy modeling literature that incorporates endogenous technological change (ETC) into the framework can potentially bring new insights into this debate. Using an inter-temporal, multi-sector CGE approach with ETC incorporated into the framework, this paper builds a model that focuses on the equity-efficiency debate for the policymakers. Canada is chosen as the country of investigation for this purpose. The paper provides a new welfare ranking of four permit allocation policies that address the equity-efficiency trade-off. In a second-best setting with pre-existing distortions, output-based allocation (OBA) of emission permits is compared to three other policy options: (i) an emissions trading system with grandfathered allocation (GFA), (ii) an auction permit trading system where permit revenue is recycled to lower payroll taxes (RPT), and (iii) a hybrid of OBA and R&D subsidy (O-R&D). We find that adapting OBA, as well as O-R&D, is welfare improving over GFA. The implicit output subsidy, entailed in the OBA policy, mitigates against the rising cost effect in the GFA policy. This is reinforced through added investment incentive in R&D when ETC in incorporated into the framework. With O-R&D, since the R&D subsidy corrects for market imperfections in the knowledge accumulation process, the effect is further bolstered, culminating into mitigation of uneven distributional outcome for energy-intensive industries as a whole. Contrary to previous results, we also find that, in terms of the welfare metric, OBA unequivocally improves the distributional outcome across sectors as compared to the RPT policy. Inclusion of ETC also unequivocally generates a higher welfare ranking for all permit policy schemes. Chapter III: Emission permit banking and induced technological change. This paper attempts to undertake an exploratory research by integrating two themes in the emission mitigation policy literature, which include: the inter-temporal emission permit banking and borrowing and the role of induced technological change in emission mitigation. Using a simple optimal control approach, we construct a unified framework that evaluates the optimal path of emissions and the optimal trajectory of permit price when both inter-temporal banking and borrowing of permits and the effects of induced technological change (ITC) are present. We find that ITC leads to a declining emission trajectory over time. The effect of ITC on the optimal permit price path, however, is ambiguous and critically depends on the extent of marginal cost saving that emanates from emission-saving technological innovation.

Thesis Abstract: Chapter I: Regional burden sharing of GHG mitigation policies – A Canadian perspective. The distribution of the burden of cost arising from the reduction of greenhouse gas (GHG) emissions is a contentious issue in policy discussions; more so among regional jurisdictions in the federalist countries with decentralized authorities over environmental regulations. In this setting, often the policy discussions are focused on the distribution of regional emission reduction targets that, in turn, entails negotiations over the distribution of the scarcity rents and the regional transfers of wealth. The allocation of regional emission entitlements is thus a key factor that could hinder the political feasibility of a national GHG mitigation policy. In this paper, we build a multi-region computable general equilibrium (CGE) model of the Canadian economy to assess the implications of different burden sharing rules governing the national GHG abatement policy with a cap-and-trade system of emission permits. In addition to assessing the impacts of traditional regional emissions allocation rules that involve intra-regional transfers of wealth, we consider a particular emission allocation that avoids such transfers, which may be a more palatable option given the context of likely fierce negotiations over the issue. Our results indicate to differing outcomes depending on the allocation policy in use. The CGE framework is also able to shed light on the transmission mechanisms that drive the results underlying the policy options. Chapter II: Endogenous technological change and emission allowances. Given the imminent threat of global warming due to GHG emissions, a number of emission mitigation policies have been proposed in the literature. However, they generally suffer from the classical equity-efficiency trade-off. High costs from equity concerns often render environmental policies politically unattractive and thus hard to implement. Recent advancement in the climate policy modeling literature that incorporates endogenous technological change (ETC) into the framework can potentially bring new insights into this debate. Using an inter-temporal, multi-sector CGE approach with ETC incorporated into the framework, this paper builds a model that focuses on the equity-efficiency debate for the policymakers. Canada is chosen as the country of investigation for this purpose. The paper provides a new welfare ranking of four permit allocation policies that address the equity-efficiency trade-off. In a second-best setting with pre-existing distortions, output-based allocation (OBA) of emission permits is compared to three other policy options: (i) an emissions trading system with grandfathered allocation (GFA), (ii) an auction permit trading system where permit revenue is recycled to lower payroll taxes (RPT), and (iii) a hybrid of OBA and R&D subsidy (O-R&D). We find that adapting OBA, as well as O-R&D, is welfare improving over GFA. The implicit output subsidy, entailed in the OBA policy, mitigates against the rising cost effect in the GFA policy. This is reinforced through added investment incentive in R&D when ETC in incorporated into the framework. With O-R&D, since the R&D subsidy corrects for market imperfections in the knowledge accumulation process, the effect is further bolstered, culminating into mitigation of uneven distributional outcome for energy-intensive industries as a whole. Contrary to previous results, we also find that, in terms of the welfare metric, OBA unequivocally improves the distributional outcome across sectors as compared to the RPT policy. Inclusion of ETC also unequivocally generates a higher welfare ranking for all permit policy schemes. Chapter III: Emission permit banking and induced technological change. This paper attempts to undertake an exploratory research by integrating two themes in the emission mitigation policy literature, which include: the inter-temporal emission permit banking and borrowing and the role of induced technological change in emission mitigation. Using a simple optimal control approach, we construct a unified framework that evaluates the optimal path of emissions and the optimal trajectory of permit price when both inter-temporal banking and borrowing of permits and the effects of induced technological change (ITC) are present. We find that ITC leads to a declining emission trajectory over time. The effect of ITC on the optimal permit price path, however, is ambiguous and critically depends on the extent of marginal cost saving that emanates from emission-saving technological innovation.

El propòsit general de la tesi va ser analitzar les relacions entre el nivell de coneixement previ del tema, els processos de regulació de l'escriptura i la qualitat del text en una mostra d'estudiants universitaris. En l'estudi van participar 24 estudiants de segon any de Magisteri en Educació Infantil, els que van elaborar un text expositiu de comparació i contrast. La recollida de dades va incloure registres de pantalla d'ordinador, auto-informes i documents (notes i text produït).Els resultats de l'estudi son quatre. En primer lloc, els resultats dels autors informes mostren que els dos principals problemes d'escriptura que ambdós grups d'estudiants van esmentar afrontar amb major freqüència durant la composició del text, van ser com elaborar la síntesi comparativa i com organitzar les idees per crear el seu propi text. En segon lloc, en termes descriptius, els resultats indiquen que els estudiants amb alt nivell de coneixements del tema a diferència dels de baix nivell utilitzar una major quantitat de processos de regulació de problemes a nivell superficial del text, especialment en el període inicial i mitjà del procés de composició. A més, s'observa que els estudiants amb major nivell de coneixement utilitzen amb lleugera diferència, una major quantitat de processos de regulació de significat a l'inici i al final de la producció textual, mentre que els dos grups utilitzen una quantitat pràcticament similar durant el període mitjà.En tercer lloc els resultats de l'anàlisi de correlació indiquen que no existeixen relacions significatives entre el nivell de coneixement del tema i l'ús de processos de regulació de l'escriptura, així com tampoc entre el nivell de coneixement del tema i la qualitat del text. En quart lloc, els resultats indiquen que existeixen relacions significatives entre els processos de regulació de problemes superficials en el text i la qualitat del contingut del text produït, en els estudiants d'alt i baix nivell de coneixement previ del tema. A més, es van trobar relacions significatives entre els processos de regulació del significat i la qualitat de l'organització del text, únicament en els estudiants de baix nivell de coneixement previ del tema. Finalment, el treball ofereix discussions dels resultats, assenyala les implicacions educatives i esmenta els aspectes pendents per a la investigació futura en el camp. / El propósito general de la tesis fue analizar las relaciones entre el nivel de conocimiento previo del tema, los procesos de regulación de la escritura y la calidad del texto en una muestra de estudiantes universitarios. En el estudio participaron 24 estudiantes de segundo año de Magisterio en Educación Infantil, quienes elaboraron un texto expositivo de comparación y contraste. La recogida de datos incluyó registros de pantalla de ordenador, auto-informes y documentos (notas y texto producido). Los resultados del estudio son cuatro. En primer lugar, los resultados de los auto-informes muestran que los dos principales problemas de escritura que ambos grupos de estudiantes mencionaron afrontar con mayor frecuencia durante la composición del texto, fueron cómo elaborar la síntesis comparativa y cómo organizar las ideas para crear su propio texto. En segundo lugar, en términos descriptivos, los resultados indican que los estudiantes con alto nivel de conocimientos del tema a diferencia de los de bajo nivel utilizaron una mayor cantidad de procesos de regulación de problemas a nivel superficial del texto, especialmente en el periodo inicial y medio del proceso de composición. Además, se observa que los estudiantes con mayor nivel de conocimiento emplean con ligera diferencia, una mayor cantidad de procesos de regulación de significado al inicio y al final de la producción textual, mientras que ambos grupos utilizan una cantidad prácticamente similar durante el periodo medio. En tercer lugar, los resultados del análisis de correlación indican que no existen relaciones significativas entre el nivel de conocimiento del tema y el uso de procesos de regulación de la escritura, así como tampoco entre el nivel de conocimiento del tema y la calidad del texto. En cuarto lugar, los resultados indican que existen relaciones significativas entre los procesos de regulación de problemas superficiales en el texto y la calidad del contenido del texto producido, en los estudiantes de alto y bajo nivel de conocimiento previo del tema. Además, se encontraron relaciones significativas entre los procesos de regulación del significado y la calidad de la organización del texto, únicamente en los estudiantes de bajo nivel de conocimiento previo del tema. Finalmente, el trabajo ofrece discusiones de los resultados, señala las implicaciones educativas y menciona los aspectos pendientes para la investigación futura en el campo. / The purpose of this dissertation was to analyze the relationship between the level of prior topic knowledge, the regulatory processes of writing and text quality in a sample of university students. The study involved 24 students from second year of Education in Early Childhood Education, who developed an expository text comparison and contrast. Data collection included computer screen registers, self-reports and documents (notes and text produced). The results of the study are fourfold. First, the results of self-reports show that the two main problems of writing that both groups of students mentioned more often face during the writing of the text, were how to develop comparative synthesis and how organize ideas for creating your own text. Second, in descriptive terms, the results indicate that students with high level of topic knowledge in contrast to the low level used a greater number of regulatory processes at the surface problems of the text, especially in the initial period through the process of composition. It is further noted that students with higher level of topic knowledge used with slight difference, a greater number of regulatory processes of meaning at the beginning and end of textual production, while both groups used a similar amount during the middle period.Third, the results of correlation analysis indicate no significant relationship between the level of topic knowledge and use of regulatory processes of writing, nor between the level of topic knowledge and text quality. Fourth, the results indicate that significant relationships between the processes of regulation of surface problems in the text and the quality of the content of the text produced by students with high and low prior topic knowledge. In addition, significant relationships were found between the regulatory processes of meaning and quality of the organization of the text, only in students with low level of prior topic knowledge. Finally, the paper provides discussions of the results, says the educational implications and mentioned the outstanding issues for future research in the field.

L'objectiu general de la tesi va ser analitzar les relacions entre el nivell de coneixements previs del tema (alt i baix), els processos d'autoregulació (planificació, monitoratge i ús d'estratègies) i els nivells de comprensió de múltiples textos expositius (rendiment en una tasca inferencial, comprensió superficial i transferència de coneixements). 40 estudiants de la Llicenciatura en Ciències de l'Educació amb especialitat en Químic-Biològiques van participar en la investigació. L'obtenció de dades va incloure el registre de protocols de pensament en veu alta, així com diferents mesures de comprensió.En primer lloc, els resultats de l'estudi mostren que no existeix una correlació significativa entre el nivell de coneixement previ i els diferents nivells de comprensió. No obstant això, l'anàlisi del grandària de l'efecte assenyala una tendència per part dels estudiants de baix coneixement previ a obtenir un millor rendiment en la tasca inferencial. Per la seva banda els estudiants amb alt coneixement previ només van rendir bé en la mesura de comprensió superficial. En segon lloc l'anàlisi de la grandària de l'efecte indica que el nivell alt de coneixement previ té un impacte rellevant en un major ús d'estratègies d'autoregulació. En tercer lloc es va trobar una relació positiva entre els processos de planificació i monitoratge per al grup d'estudiants de baix coneixement previ, mentre que en el cas dels estudiants d'alt coneixement no es van trobar relacions significatives. En quart lloc, els resultats indiquen una relació negativa entre el procés de planificació i el rendiment en la tasca inferencial per al grup de baixos coneixements, mentre que per als estudiants d'alts coneixements es va trobar una relació significativa entre els processos de planificació i el rendiment en la tasca inferencial. Finalment, les anàlisis de regressió mostren que ni el nivell de coneixement previ ni els processos d'autoregulació prediuen la variància en els nivells de comprensió. Aquests resultats són discutits en termes de la teoria i investigació de la comprensió tant d'un únic i múltiples textos, així com de les teories d'autoregulació de l'aprenentatge. / El objetivo general de la tesis fue analizar las relaciones entre el nivel de conocimientos previos del tema (alto y bajo), los procesos de autorregulación (planeación, monitoreo y uso de estrategias) y los niveles de comprensión de múltiples textos expositivos (rendimiento en una tarea inferencial, comprensión superficial y transferencia de conocimientos). 40 estudiantes de la Licenciatura en Ciencias de la Educación con especialidad en Químico-Biológicas participaron en la investigación. La obtención de datos incluyó el registro de protocolos de pensamiento en voz alta, así como diferentes medidas de comprensión.En primer lugar, los resultados del estudio muestran que no existe una correlación significativa entre el nivel de conocimiento previo y los diferentes niveles de comprensión. Sin embargo, el análisis del tamaño del efecto señala una tendencia por parte de los estudiantes de bajo conocimiento previo a obtener un mejor rendimiento en la tarea inferencial. Por su parte los estudiantes con alto conocimiento previo solamente rindieron bien en la medida de comprensión superficial. En segundo lugar el análisis del tamaño del efecto indica que el nivel alto de conocimiento previo tiene un impacto relevante en un mayor uso de estrategias de autorregulación. En tercer lugar se encontró una relación positiva entre los procesos de planeación y monitoreo para el grupo de estudiantes de bajo conocimiento previo, mientras que en el caso de los estudiantes de alto conocimiento no se encontraron relaciones significativas. En cuarto lugar, los resultados indican una relación negativa entre el proceso de planeación y el rendimiento en la tarea inferencial para el grupo de bajos conocimientos, mientras que para los estudiantes de altos conocimientos se encontró una relación significativa entre los procesos de planeación y el rendimiento en la tarea inferencial. Finalmente, los análisis de regresión muestran que ni el nivel de conocimiento previo ni los procesos de autorregulación predicen la varianza en los niveles de comprensión. Estos resultados son discutidos en términos de la teoría e investigación de la comprensión tanto de un único y múltiples textos, así como de las teorías de autorregulación del aprendizaje. / The overall objective of the dissertation was to analyze the relationship between the level of background knowledge (high and low), self-regulatory processes (planning, monitoring and use of strategies) and multiple levels of understanding expository texts (performance in a inferential task, surface understanding and knowledge transfer). 40 students from the Bachelor of Science in Education with specialization in Chemical-Biological participated in the investigation. Data collection included recording think-aloud protocols, and various measures of comprehension.First, the results show that there is no significant correlation between the level of background knowledge and different levels of comprehension. However, the effect size analysis indicates a tendency for students with low prior knowledge to better performance on the inferential task. For their part, students with high prior knowledge only performed well in the superficial understanding measure. Secondly, the effect size analysis indicates that the high level of prior knowledge has a significant impact on the greater use of self-regulation strategies. Third, we found a positive relationship between planning and monitoring processes for the group of students with low prior knowledge, whereas in the case of students with high knowledge there were no significant relationships. Fourth, the results indicate a negative relationship between the process of planning and inferential task performance for the low knowledge group, while for higher knowledge students found a significant relationship between planning processes and performance the inferential task. Finally, regression analysis showed that neither the level of prior knowledge nor self-regulatory processes predict the variance in levels of comprehension. These results are discussed in terms of theory and research on both single and multiple texts comprehension and theories of self-regulated learning.