In many early intensive behavioural intervention (EIBI) programs, such as the St.Amant Autism Programs in Manitoba, the Assessment of Basic Language and Learning Skills-Revised (ABLLS-R) is used as an assessment guide, a curriculum guide, and a skills tracking system. The ABLLS-R lists a total of 544 potential training tasks. In the St.Amant Autism Programs, the Assessment of Basic Learning Abilities-Revised (ABLA-R) is also used to assess a child’s ability to learn to perform 6 tasks, called levels, which are hierarchical in difficulty. Research has demonstrated that training tasks comparable to failed ABLA-R levels are very difficult to teach using standard prompting and reinforcement procedures, and training tasks that are mismatched to a client’s highest-passed ABLA-R level result in more aberrant behaviours than matched tasks. The purpose of this study was to evaluate whether the difficulty of the training tasks taught to children enrolled in the St.Amant Autism Early Learning Program matched the learning abilities of the children. In order to do so, five questions were examined based on the ABLLS-R, the ABLA-R, and archival data from the St.Amant Autism Programs. The results reveal several potentially important findings. First, observers who were knowledgeable about the ABLA-R reliably categorized 99 of the 544 ABLLS-R tasks into individual ABLA-R levels. Second, for a random sample of those 99 ABLLS-R tasks, Autism Consultants from the St.Amant Autism Programs averaged 90.5% agreement that those tasks were taught at the categorized ABLA-R levels. Third, across a sample of 14 children, 81% of their training tasks were mismatched to each child’s highest-passed ABLA-R level. Fourth, for that sample of 14 children and across their 31 maladaptive behaviour assessments, 61% of the assessments had a score representative of maladaptive behaviours. Fifth, for that same sample of children, their rates of acquisition of new training tasks were lower for mismatched tasks than for matched tasks. These findings have important implications for potentially improving the services that children with ASD are receiving from the St.Amant Autism Programs and other EIBI programs. / February 2016

La investigación tiene como objetivo analizar el conocimiento especializado del profesor de matemáticas en la enseñanza-aprendizaje de los problemas aritméticos de enunciado verbal a partir de la resolución de tareas formativas, las cuales permiten mejorar la práctica docente en la enseñanza de problemas aritméticos de enunciado verbal (PAEV). Para el diseño de las tareas, se utilizó el modelo MTSK (Contreras, Climent, Montes & Carrillo, 2017), el cual permite observar el conocimiento especializado de los docentes de matemática. La metodología responde a un paradigma interpretativo y enfoque cualitativo. Por tal motivo se ha seleccionado un estudio de diseño, ya que se desarrolla en el campo de la Ciencia del aprendizaje (Molina, 2011). El trabajo empezó con una ardua investigación sobre el MTSK, la resolución de problemas y las tareas formativas. Esta investigación permitió diseñar un instrumento con indicadores que provienen de cuatro de los seis subdominios del modelo MTSK: Conocimiento de los Temas (KoT), Conocimiento de la Práctica Matemática (KPM), Conocimiento de la Enseñanza de las Matemáticas (KMT) y Conocimiento de las Características del Aprendizaje de las Matemáticas (KFLM). Con los indicadores se diseñaron cuatro tareas formativas, las cuales fueron validadas, al igual que los indicadores, a través de juicios de expertos. Se concluyó que a través del modelo MTSK es posible diseñar tareas formativas para docentes que permiten analizar la enseñanza-aprendizaje de la matemática. / The research work aims to analyze the Mathematics Teacher Specialized Knowledge in the teaching-learning of Arithmetic Problems of Verbal Statement (PAEV) from the resolution of formative tasks, which allow improving the teaching practice in the teaching of Arithmetic Problems of Verbal Enunciation (PAEV). We use the MTSK model to design tasks (Contreras, Climent, Montes & Carrillo, 2017) since it allows observing the specialized knowledge of mathematics teachers. The methodology responds to an interpretive paradigm and a qualitative approach. For this reason, we selected a design study since it is developed in the Education science field (Molina, 2011). The methodology responds to an interpretive paradigm and a qualitative approach. For this reason, we selected a design study since it is developed in the Education science field (Molina, 2011). This research allowed the design of an instrument with indicators that come from four of the six subdomains of the MTSK model: Knowledge of Topics (KoT), Knowledge of Mathematical Practice (KPM), Knowledge of Mathematics Teaching (KMT), and Knowledge of the Learning Characteristics of Mathematics (KFLM). Four training tasks were designed with the indicators, which were validated, like the indicators, through expert judgments. It concluded that through the MTSK model it is possible to design training tasks for teachers that allow analyzed teaching-learning of mathematic. / Tess