Proceedings

  • Students learn by doing: unfortunately many only do as little as is required of them to meet formal assessment requirements. We examine how a radical change in assessment strategy was chosen as a method of improving the learning experience for a group of students engaged in a final year undergraduate course in Time Series. In 2006, following attempts to change student attitudes through the use of different teaching techniques, we decided instead to focus on altering our assessment strategies. This paper will show that assessment appears to be one of the most effective tools to improve learning. While we remain committed to investigating and implementing improved methods for course delivery, our recent experience indicates the need to complement those methods with innovative assessment to improve the whole learning experience.

  • In this paper we present part of a study carried out to identify Lebanese teachers' representations of probability as well as their teaching practices. We compare grades and judgments that teachers attribute to fictitious students. Our result show inconsistencies in grading on the same teacher: discrepancy between the quantitative judgment (grade) and the qualitative judgment that he attributes to the same answer. The comparison between the quantitative judgments (grades) and the qualitative judgments reveals a great diversity among teachers: a convergence in grading can hide very different qualitative judgments. This comparison contributed largely to the study of certain representations and practices of teachers, in particular those concerning the concept of independence of events.

  • The aim of this study is to assess and investigate how many combinations can be manipulated by preschoolers in a probabilistic task. The task was computer- based and included 3 trials with alterations among the colour- combinations and proportions of the sample space. With the use of intuitive thinking or more concrete reasoning children at the age of 5-6, showed that they posses the notion of most/least likely when there are 2 and 3 combinations (not 4).

  • The Bologna educational reform taking place across Europe has reached Spain and Portugal. It places the student as the center of the entire process of teaching and learning, recognizing two main types of learning: presential and non-presential. With this in mind, the assessment methods need to be redefined and adapted to this new reality, where tutoring and learning must be integrated in the assessment process. This study takes place in two Introductory Statistics courses, one located in Vigo (Spain) and one in Guimarães (Portugal) and the results show an improvement in students' grades and also allow a more solid learning process achieved by the continuous nature of the method. Having the material virtual in Learning Management Systems (LMS) simplifies the work of the teacher and encourages the students to develop new study habits necessary for their success within the new reality of the Bologna reforms.

  • Teaching statistics to future statisticians should take into account both statistical knowledge and personal skills needed in the statisticians' professional life. Based on a questionnaire designed to assess competencies actually needed by statisticians in the workplace, this paper aims to detect if the competence framework helps to better understand the transition from university to work. With this objective, three metric profiles are proposed to locate knowledge, "relational skills" and actually performed activities, and graduate scores. The graduates in Statistical Sciences from "Sapienza" University of Rome from March 2000 to March 2001 were acquired as emblematic cases in Spring 2004. There were 146 respondents out of the total of 296 graduates in the time period. The Rasch analysis is applied to analyze the data and to build on the profiles. In particular the analysis was performed applying the Rasch family response models to polytomous items. The item parameters and the latent trait value for each respondent are estimated by the joint maximum likelihood method. The analysis results could be useful in order to design curricula for university degrees in Statistics that would make the university-workplace transition process easier.

  • This paper describes a classroom experience using a data gathering and analysis tool to scaffold a learning process that involves students in analysis of their own understanding about data. An AP-Statistics class uses data about their own sleep patterns to investigate measures of variability. The class applies various measures of variability to the sleep data set and comments on their efficacy using a survey created in Fathom. The students comment on the measures in groups and construct a common understanding of which measure is the best and why. Students in a statistics class should find themselves routinely engaged in data analysis. We conjecture that encounters with their own assessment data increases their appreciation of data analysis at the same time that it helps them identify weak areas of understanding. The students are not only learning about data variability but also about statistical process, data gathering and analysis.

  • With the growing focus on assessment and accountability, programs at many universities are now expected to define learning goals and objectives at the program level (as opposed to the course level) and to devise strategies for assessing whether these goals and objectives are being met. This paper will look at the role of assessment at the program level and how the resulting information about student learning can be used to make informed decisions about curriculum.

  • The main role of assessment is to support learning, and any view of assessment implies a corresponding view of learning. Research on students' conceptions of statistics, learning in statistics and assessment, suggests that there is a clear variation from narrow to broad views. Another dimension is students' perceptions of their future professional roles and how that impacts on their present studies. In order to support the learning process, assessment should be structured in such a way as to make apparent to students the full range of variation in conceptions and to encourage them towards the broadest and most inclusive ideas. Further, it is important that the approach to assessment has coherence with the overall pedagogical approach.

  • How do innovative pedagogical techniques improve learning and mastery of introductory statistics? This research study examines proficiency grading and assignment resubmission and compares them to traditional statistical teaching methodology in two introductory statistics classes. The control class received traditional numeric grades, while the experimental class received grades on a three-tiered proficiency ranking and the opportunity to resubmit assignments to increase their proficiency score. Students in the control class scored higher on a common final exam (although not statistically significant), and believed the material was better taught, while students in the experimental class claimed to have learned more and were more satisfied with the grading in the course. Future research will expand data gathering and improve the research design.

  • Many subjects in the school curriculum engage with contexts where multiple factors interact. Historically however, data have rarely been used at school level in such contexts because of the difficulties inherent in understanding multiple variable relationships. Stronger links across traditional subjects has proved an elusive aspiration for curriculum developers. We are currently engaged in a pilot project with the Northern Ireland Curriculum Authority (CCEA) to use some innovative interfaces with multivariate summary data as a focus for multiple perspectives on various contexts. Innovations in curriculum design offer opportunities for innovation in assessment. Often statistics assessment focuses primarily on accurate performance of routine calculations or graphical construction. Here the use of data is primarily to enhance understanding. This paper will explore mechanisms for embedding assessment of key statistical concepts within cross-curricular activities.

Pages

register