Invited Talks

Topic 1 : Sustaining strengths and building capacity in statistics education

This topic emphasizes the main theme of the conference: “Sustainability in statistics education”. Statistics education gains increasing importance in a world beset with oceans of data and ever-increasing need for diverse statistical skills and knowledge from professionals, workers, and citizens alike. Yet, many challenges exist in sustaining or improving the quality of education in statistics and probability in different contexts, both formal and informal. This Topic seeks proposals for conceptual, empirical, and case-based papers and whole sessions that relate to “sustaining strengths” and “building capacities”, ideas with many meanings for the various stakeholders involved in statistics education, e.g., students/learners, teachers, schools and academic institutions, employers, publishers, public agencies, and providers of official statistics. Proposed papers and sessions in this topic can address, among other things, understanding obstacles or challenges that face different types of learners or teachers (e.g., mathematics or science teachers, new PhDs or graduate students, etc.), and methods or systems for achieving measurable and sustainable results in statistics education in various contexts. Papers may address issues of access, retention and improved knowledge in statistics and probability (e.g., of students, of professionals) beyond entry-level points. Papers may also examine or critically evaluate social processes or organizational factors that affect the place of statistics and of statistics education in curriculum frameworks or in academic systems, and how improvements in this regard can be developed, supported and sustained over time.

Session 1A: Building the capacity of mathematics and science teachers to teach statistics

1A1: Curriculum expectations for teaching science and statistics

Jane Watson   University of Tasmania, Australia

This paper focuses on the curriculum links between Statistics and Science that teachers need to understand and apply in order to be effective teachers of the two fields of study. Meaningful statistics does not exist without context and Science is the content focus of Session 1A. Although curriculum documents differ from country to country, this presentation will use extracts from The Australian Curriculum: Mathematics and the US Common Core State Standards for Mathematics and link the statistical ideas with the relevant parts of The Australian Curriculum: Science and the US Next Generation Science Standards for States, by States. Teachers of Mathematics need to be aware of the potential of Science to provide meaningful contexts within which to set statistical investigations. Similarly teachers of Science, who are developing methods for implementing investigations and experiments in their classrooms, need to be aware of the close ties to statistical decision-making.


1A2: The Statistical Education of Teachers (SET): an American Statistical Association policy document

Christine Franklin   University of Georgia, United States

The American Statistical Association (ASA) has as one of its strategic priorities K-12 teacher preparation in Statistics. The release and widespread adoption of the Common Core State Standards for Mathematics (CCSSM) in the United States have dramatically increased the expectations for teaching statistics especially in grades 6 through 12. Other countries are also including more Statistics at the K-12 level. The Conference Board of the Mathematical Sciences (CBMS) identified the statistical preparation of teachers as an area of concern in their recent publication of the Mathematical Education of Teachers 2 (MET2) document. This paper will briefly outline the SET document, a companion to the MET2 document and ASA’s response to expanding the MET2 recommendations for answering the question: what preparation and support do K-12 teachers need to successfully support students’ learning of Statistics?


1A3: Building high school pre-service teachers’ knowledge to teach correlation and regression

Carmen Batanero   University of Granada, Spain
Maria M Gea   University of Granada, Spain
Carmen Díaz   University of Huelva, Spain
Gustavo R Cañadas   University of Granada, Spain

In this paper we describe a workshop which was aimed to develop the knowledge needed to teach correlation and regression in high school teachers. The workshop was based on a formative cycle model used in our previous research and directed to simultaneously increase the teachers’ statistical and pedagogical knowledge. To develop the teachers’ knowledge of correlation and regression we proposed them to complete a statistical project based on real data taken from the UNESCO web server, followed by a collective discussion of their solutions to the tasks proposed in the project. A didactical analysis of the project helped them to increase their pedagogical content knowledge. Some results in a sample of prospective Spanish teachers will be briefly reported.


Session 1B: Building the capacity of new PhDs and graduate students to teach statistics (panel discussion)

Kari Lock Morgan   Duke University, United States
Zachariah Mbasu   Maseno University, Kenya
Stephanie Budgett   University of Auckland, New Zealand
Patricia Buchanan   Pennsylvania State University, United States


Session 1C: Statistics education outreach across the globe

1C1: Outreach efforts to enhance statistical education and statistical literacy in Hungary

Péter Kovács   University of Szeged, Hungary

In order to improve statistical literacy, the institutions in the tertiary education have three target groups: students, academic staff, and external actors (citizens, secondary school students, etc.). I intend to focus on the Hungarian activities and open events for developing statistical literacy and statistical education within the framework of the International Statistical Literacy Project (ISLP), as well as the possibilities of a combined development of statistical and financial literacy for secondary school students. I emphasize how important it is for students to develop positive attitudes toward statistics and to develop the motivation of the teachers. I also review my own efforts to enhance statistic education at the tertiary level: rethink the content and the outcomes of the courses, the teaching methods, the IT tools, knowledge of students.


1C2: OZCOTS: Bringing statistical educators and statisticians together

Brian Phillips   Swinburne University of Technology, Australia

The first Australian Conference on Teaching Statistics, OZCOTS, was run in 1998 and involved papers given by the Australians at ICOTS5. Its success in bringing together Australians involved in teaching statistics resulted in regular OZCOTS gatherings from 1999 to 2002. In 2006 Helen MacGillivray was awarded one of the first Australian Learning and Teaching Council’s Senior Fellowships, and as part of her programme, OZCOTS was revived, this time running as a two-day satellite to the 2008 Australian Statistical Conference (ASC) and involved a wider audience. OZCOTS 2008 was modelled on the IASE’s satellite conferences to ISI Conferences. Further OZCOTS meetings were held during the 2010 and 2012 ASC’s. In 2012 the publisher Springer requested to publish a volume based on OZCOTS meetings, this will be available in late 2014. This talk will discuss some of the experiences of these OZCOTS events.


1C3: Japanese Inter-university Network for Statistical Education and new trials for development of students’ data analysis skills

Kazunori Yamaguchi   Rikkyo University, Japan
Michiko Watanabe   Keio University, Japan

In knowledge-based society, it is shared understanding throughout the world that “Statistical Thinking” and “Competency of Statistical Analysis” is a substantial skill for detecting and solving new issues. “Japanese Inter-university Network for Statistical Education (JINSE)” has been newly organized in 2012. The first aim of JINSE is to develop Standard Curriculum and Teaching methodology for fostering human resources capable of coping with new issues, and eventually, to establish Quality Assurance system for statistical education by introducing Evaluation Committee consisting of members from academic statistical societies and other educational/economic organizations. In this talk, we introduce activities of this project and new trials for developing students’ data analysis skills in Rikkyo University. One of our trials is to introduce action learning methods and combine leadership developing program.


Session 1D: Building the capacity to teach and understand statistics in emerging economies

1D1: On the commencement of a culture of “statistics acceptance” in a higher education institution relatively new at research

Saleha Habibullah   Kinnaird College for Women, Pakistan

Notwithstanding the fact that Statistics has an important role to play in research, there exists an element of “fear and intimidation” among non-statisticians when they encounter the need to apply statistical methodology to research-related data. This paper presents an account of the ways in which the process of moving toward the development of a culture of “Statistics Acceptance” has been commenced at Kinnaird College For Women, Lahore, Pakistan — a higher education institution acclaimed for imparting quality education to the young women which is relatively new at research. The initiatives described in this paper can be regarded as some of the earliest steps in the direction of research-enhancement through statistical education. The real challenge is to move forward in such a way that an increased awareness, acceptance and acquisition of “the statistical skill-set” can not only be sustained but also enhanced over time.


1D2: Building capacity for developing statistical literacy in a developing country

Temesgen Zewotir   University of KwaZulu-Natal, South Africa
Delia North   University of KwaZulu-Natal, South Africa
Michael Murray   University of KwaZulu-Natal, South Africa
Iddo Gal   University of Haifa, Israel

Developing countries face many challenges when aiming to develop statistical literacy. In the design of interventions and workshops, countries have to consider teachers' limited background in statistics and in mathematics education, as well as their attitudes towards statistics, perceptions of the differences between statistics and mathematics, and more. The paper has two parts. First it will review selected factors associated with teachers' background and attitudes towards statistics that may be unique to a developing country and that may affect their adoption of statistics instruction. Second, it will illustrate some of the points in the conceptual analysis by examining results from brief surveys of two large groups of teachers who attended training programs and workshops jointly hosted by the national statistics office and a university in South Africa, with a focus on teachers’ belief in their ability to teach statistics topics and on their attitudes towards acquiring statistics skills.


1D3: Statistics education in Ethiopia: successes, challenges and opportunities

Ann A O’Connell   Ohio State University, United States
Kassa Michael   Addis Ababa University, Ethiopia

In Ethiopia, efforts to increase enrolment and access to education have placed significant pressures on the training needs of teacher education programs, particularly for programs in Science and Mathematics Education (SMED). SMED is a driving factor in current priorities set forth by the Ethiopian Ministry of Education. Addis Ababa University has initiated several programs and directives focused on preparing well-trained masters and PhD students in SMED to support the expansion of science and mathematics education. In our paper, we take a case-study perspective and review the current Ethiopian curriculum in statistics and probability, and describe ongoing teaching, research and outreach efforts with potential to improve education in statistics, particularly within teacher training programs. Finally, we highlight recent efforts to improve statistical skills and quantitative research for graduate programs in teacher education.


1D4: Sustaining teachers’ capacity for teaching statistical inference through reflective practice

Enriqueta Reston   University of San Carlos, Philippines
Liza Lorena Jala   University of Cebu, Philippines

In this paper, we present reflective practice as an approach to sustain teachers’ ability to teach statistical inference in more meaningful, relevant contexts using real data and technology integration for optimal student learning outcomes. From an initial workshop with college statistics teachers on statistical inference and reflective practice, we followed up selected teachers and built multiple case studies to examine how these teachers’ understanding of statistical inference evolved and influenced their classroom practice. Using Taggart’s (2005) Reflective Thinking Model, we explored how teacher’s capacity for reflective practice improved as they moved up from the technical to the contextual and finally, to the dialectical level. Teachers reflect on their own teaching experiences to improve instructional decisions and classroom practice in these levels using techniques such as self-reports, peer observation and assessment, journal writing and focus group discussion to support the integration of reflective practice into their teaching.


Session 1E: Building capacity in Statistics majors

1E1: Skills needed for modern day statisticians

Andrej Blejec   National Institute of Biology, Slovenia

Currently we face tremendous changes in availability of information, including Big data, which are not only large in numbers of observations or variables or both, but also unstructured. New internet-based tools for analysis and visualization enable “data scientists” to quickly produce appealing data summaries. What skills are needed for modern day statisticians in this fast changing environment, where data are often short lived and results quickly become obsolete? We must emphasize visualization, with dynamic and interactive graphics. Data need to be structured and often collected from a plethora of sources, requiring additional programming skills as well as proficiency in data analysis. But no technical or theoretical skills can make statistics alive without context. As always, statisticians need to communicate with professionals and general public. To be competitive with all others who want to analyze data, we have to be flexible, responsive and productive, using statistical know-how as the competitive advantage.


1E2: Challenges and issues in developing real-world curriculum for data scientists in Japan

Takuya Kudo   Accenture Japan Ltd, Japan
Michiko Watanabe   Keio University, Japan
Tomiaki Morikawa   Keio University, Japan
Manabu Iwasaki   Seikei University, Japan
Yoshihiro Hayashi   Accenture Japan, Ltd., Tokyo, Japan
Tomoyuki Furutani   Keio University, Japan

Like many countries, Japan has challenges in producing sufficient and appropriately-trained statisticians. In response to recent Big Data demands in society, new courses related to data science are being developed in collaboration with IT companies, and an Institute of Professional Data Scientists established, collaborating with universities to develop a competency framework for professional data scientists. This paper considers a number of challenges, including reducing the domestic shortage of skilled analytics professionals, preparing such graduates for complicated real-world business problems, and developing advanced approaches to collecting and analyzing data that can drive value for the country’s businesses and public sector organizations. In discussing the challenges and opportunities, the paper refers to an example of close collaboration between the Keio University Research Institute and Accenture Japan, Ltd, in establishing the “Data Business Creation Laboratory,” an initiative to help organizations in Japan benefit from data-driven decision-making using analytical methods and skills.


1E3: Building capability in statistics majors: drawing strength from a diverse region

Alice Richardson   University of Canberra, Australia

Although structures, systems, problems and outlooks may differ between, and even within, countries, there are common challenges and opportunities which are benefitted by discussion of similarities and contrasts between countries. A unique opportunity was presented in November 2013 when academics from Vietnam, Australia, and USA came together with Vietnamese employers and other international statistics trainers from France, Philippines, Korea and Japan. Sponsored by the UN Population Fund, the group shared experiences and developed a way forward for statistics training in Vietnamese universities. Motivated by this conference, this paper uses reference to three diverse systems, namely Australia, Vietnam and the United States, to identify issues, directions and structural constraints. Possibilities are explored for using both policy and structure to build capacity in statistics majors, in terms of both quantity and quality. Commonalities in lessons learnt and ongoing challenges can inform capacity building in statistics majors worldwide.


Session 1F: The importance of attitudes in statistics education: sustaining learning processes and outcomes

1F1: Student attitudes toward statistics from a randomization-based curriculum

Todd Swanson   Hope College, United States
Jill VanderStoep   Hope College, United States
Nathan Tintle   Dordt College, United States

Recently, a large national (US) sample was used to evaluate attitudes toward statistics among undergraduate students. The majority of the courses in the sample used a similar (AP Statistics) curriculum. Recent interest in the use of randomization-based methods in introductory statistics raises substantial questions about the conceptual effectiveness and attitudes of students in such courses. To begin to better understand the advantages and disadvantages of a randomization-based curriculum, we evaluated student attitudes in courses using randomization curricula and compared these to students using a traditional curriculum. Overall, there were only small, statistically and practically insignificant, differences in student attitudes between the two samples. While randomization approaches remain new, our analysis suggests that these curricula may not be harming nor improving students’ attitudes toward statistics relative to traditional courses in undergraduate courses.


1F2: How do attitudes change from one stats course to the next?

Anne Michele Millar   Mount Saint Vincent University, Canada
Bethany White   Mount Saint Vincent University, Canada

Are attitudes sustained across the term break? We usually measure attitudes at the beginning and end of a course. However, at the end of the course students are suffering from end of term stress and exam anxiety, and these may be reflected in their attitude scores. We would hope that positive attitudes are sustained, while negative attitudes improve during the break. Sustaining positive attitudes leads to sustained student engagement. We consider data for two introductory statistics courses taught over two semesters at a mid-size primarily undergraduate university.


1F3: A fallacy in student attitude research: the impact of the first class

Michael Posner   Villanova University, United States

Student attitudes have been shown to be an important measure in short-term learning and long-term retention of material. However, instruments like the Survey of Attitudes Toward Statistics (SATS) fail to show an increase in student attitudes over time. Most researchers give students the test after the first week of classes, not recognizing the impacts of the first class or first few classes on students’ attitudes and first impression. This study quantifies the impact of the first class on student attitudes, using a randomized trial design where half of the students in each class take the SATS before and half after the first class, on almost 500 students who studied statistics in different areas or contexts. Preliminary results are included here. Additional results will be discussed during the presentation.


1F4: Comparing attitudes toward statistics among students enrolled in project-based and hybrid statistics courses

Caroline Ramirez   University of the Pacific in California, United States
Marjorie Bond   Monmouth College, United States

Considerable research has been devoted to how students learn statistics, such as through data analysis projects, simulations, online learning courses, and traditional lecture. This study analyzes the differences in attitudes towards statistics among introductory statistics students experiencing two different learning environments: a project based statistics course versus a mix of traditional and online learning course (hybrid course). The Survey of Attitudes Toward Statistics (SATS-36) is used to compare the attitudes of college students from a small liberal arts college enrolled in an introductory statistics course during the Fall 2012 and Spring 2013 semesters. This study suggests that an examination of these environments is needed to better understand students’ attitudes.