Papers/Conferences
DESCRIBING DATA WELL IN R-INSTAT
Maxwell Fundi1, Lily Clements2, David Stern3, Roger Stern2,3 François Renaud1 and Alex Sananka1
1African Maths Initiative
2Statistics for Sustainable Development
3University of Reading
maxwell@africanmathsinitiative.net
ABSTRACT
In 21st century, there is an increasing need to have skills to derive meaning from the growing data around us. In Africa, too much of statistical teaching is theoretical. This leaves students with a lack of data handling skills, and often unprepared to find meaning in data. The African Data Initiative (ADI) aims to change this. A first step has been to develop R-Instat, an open-source, free software based on the increasingly used statistics software R. This paper explains some of the decisions behind R-Instat’s approach to encouraging descriptive analysis. It also proposes how this could support the teaching of good descriptive statistics.
Click here to view and download this paper.
STILL COMING DOWN FROM THE MOUNTAINS
Roger Stern1,2, Ric Coe2,3 and David Stern1 1University of Reading 2Statistics for Sustainable Development 3ICRAF
r.d.stern@reading.ac.uk
ABSTRACT
Statistics has changed in many ways since the 1960s, when many African countries became independent. Among these changes are an increasing emphasis on data and a set of unifying principles that can simplify the teaching of statistical modelling. These changes have yet to impact training in statistics in many countries. Access to technology is needed if these changes are to be incorporated into statistics teaching, and this is now feasible in many African universities
Click here to view and download this paper.
EMBEDDING DATA MANIPULATION IN STATISTICS EDUCATION
James Musyoka1, John Lunalo2, Cathy Garlick4, Steven Ndung’u2, David Stern4, Danny Parsons5 and Roger Stern3,4
2Maseno University, Department of Statistics and Actuarial Science, Kenya
2African Maths Initiative, Kenya
3Statistics for Sustainable Development, UK
4Statistical Services Centre, University of Reading, UK
5Mathematical Institute, University of Oxford, UK
johnlunalo95@gmail.com
ABSTRACT
University courses in statistics in many African countries are dominated by data analysis. This is just one component of the subject and students therefore lack knowledge on many important practical laboratory work components. Here we consider how data collection and data entry can be included in training courses. The next important stage of preparing the data, so it is ready for analysis is also considered. These stages, before data analysis, may use a combination of a spreadsheet, a statistics package and special software. Examples of each are considered.
Click here to view and download this paper.
STATISTICS IN MATHS CAMPS
Zachariah Mbasu, Thomas Mawora and David Stern
African Maths Initiative
Maseno University
Statistical Services Centre, University of Reading
zmbasu@gmail.com, tmawora@maseno.ac.ke, d.a.stern@reading.ac.uk
ABSTRACT
It is generally agreed that statistics is an important discipline to be introduced at school level. However, only small components of the subject and a narrow scope are introduced at primary and secondary school level curriculum in Kenya. This paper discusses the emerging prominence of statistics sessions at Math Camps in Africa. It shows how maths camp student participants have developed the knowledge and skills to support further learning of important statistical concepts. This has involved hands on sessions where students interact with real data sets.
Click here to view and download this paper.
MAKING MULTILEVEL DATA IDEAS MORE ACCESSIBLE
Danny Parsons1, David Stern2 and Roger Stern2,3
1Mathematical Institute, University of Oxford, UK
2University of Reading, UK
3Statistics for Sustainable Development, UK
danny@aims.ac.za
ABSTRACT
Each year increasing amounts of data are being produced and there are growing trends towards data becoming more accessible, particularly online. Here we present a range of examples where data are conveniently arranged in multiple linked rectangles or data frames. They are often omitted from all but advanced statistics courses. However, they are common in practice, hence their omission leaves graduates poorly prepared for real world problems. The obvious example is a survey that is at multiple levels. Other examples include multiple time series with spatial data, where the spatial information is in a separate data frame; and data sets in a single rectangle (data frame) but where the analyses are on summary data. The statistical software, R-Instat, resulting from the African Data Initiative is designed to make it easy to handle such data.
Click here to view and download this paper.
OPEN EDUCATIONAL RESOURCES FOR STATISTICS TEACHING
James Musyoka1, Roger Stern2 and David Stern3
1Maseno University, Department of Statistics and Actuarial Science, Kenya
2Statistics for Sustainable Development, UK
3Statistical Services Centre
jkmusyoka@maseno.ac.ke
ABSTRACT
A series of papers at this conference have built on the African Data Initiative (ADI) which was started to improve statistical literacy and understanding. One immediate deliverables of this initiative has been new statistical software which is free, easy-to-use, open source and which encourages good statistical practice. Here we show how this software together with other open educational resources can be used to improve statistics teaching. This is demonstrated using an undergraduate course which was offered to about 300 students at Maseno University, Kenya. The resources include those from an e-learning course, called e-SMS (Statistics Made Simple) and an electronic statistics book called Computer Assisted Statistics Textbook (CAST). The course also made extensive use of Moodle to enable a “blended” approach to be undertaken.
Click here to view and download this paper.
SEEDING THE AFRICAN DATA INITIATIVE
DAVID STERN
University of Reading
d.a.stern@reading.ac.uk
ABSTRACT
The African Data Initiative (ADI) is a highly collaborative project that aims to transform statistics education and how people use and understand data, both in Africa and beyond. The first major activity of ADI has been the development of R-Instat, a front-end to R, tailored to African needs and developed largely in Africa. This paper describes the background, initial activities and the principles of ADI. The principles provide structure to guide and communicate thinking behind ADI decision making, for both existing and future activities. The ADI collaboration exists primarily through a common desire to contribute towards Africa’s data revolution alongside a collective principal based approach.
Click here to view and download this paper.
THE GROWING ROLE OF COMPUTERS FOR TEACHING STATISTICS IN KENYA.
Parin Kurji1, Brigid McDermott1, David Stern2 and Roger Stern3
1Department of Plant Sciences, University of Nairobi, Kenya
2Department of Mathematics and Statistics, Maseno University, Kenya
3Statistical Services Centre, University of Reading, United Kingdom
parinkurji@gmail.com
ABSTRACT
Until recently, the teaching of statistics in East Africa has been a traditional chalk-and-talk affair. In the last few years computers have become more widely accessible. At the same time many statistical resources of the highest quality are freely available for Africa, including ComputerAssisted Statistics Textbooks (CAST), an electronic textbook, GenStat Discovery Edition, (a statistics package), and training resources such as the Southern African Development Community (SADC) Training Pack DVD prepared by Statistics Services Centre (SSC) , Reading University. This means that change is not only possible but is within reach of lecturers all over Africa. Experiences at two Kenyan universities are described. Initiatives for undergraduates and postgraduates in both service teaching and specialist teaching of statistics are discussed
Click here to view and download this paper.
TRAINING OF LECTURERS AT MASENO UNIVERSITY, KENYA
James Musyoka, Joyce Otieno and David Stern Department of Mathematics and Applied Statistics, Maseno University, Kenya
jamusyoka@hotmail.com
ABSTRACT
Lecturers in Kenyan universities usually start their work without any prior training on how to teach students. Hence the teaching tends to be traditional and new training techniques are rarely implemented. For effective teaching, lecturers need to develop skills on preparation and presentation of new course materials, using modern methods. One of many changes to the statistics teaching at Maseno University was to teach an MSc. course in a way that also provided staff training. The course combined an international e-learning course with lecture notes which had previously been taught by a visiting lecturer. Two junior staff were given the day to day responsibility of teaching the new MSc course, with a senior lecturer observing, advising and taking overall responsibility. This paper describes the course and the learning experience
Click here to view and download this paper.
INCREMENTAL MODERNISATION OF STATISTICS TEACHING AND CURRICULUM AT MASENO UNIVERSITY, KENYA
David Stern, Omolo N Ongati, John Ogonji Agure and Betty Ogange
Department of Mathematics and Applied Statistics, Maseno University, Kenya volloholic@hotmail.com
ABSTRACT
Modernisation of statistics teaching is a continual problem the world over. The advances in statistical methods and tools along with the growing demand of applied practitioners creates a dual need of people with the theoretical knowledge to take the subject further and those with the practical knowledge and skills for the many current problems requiring statistical support. The universities in Kenya are largely still teaching theory as was done 40 years ago. Change is possible and initiatives like the RUFORUM M.Sc. in Research Methods show that with sufficient resources a modern curriculum can be created in Kenya in a short space of time. Maseno is a young university, less than 20 years old. With very few resources the department of mathematics and applied statistics has been taking a more gradual approach to modernise their teaching. This paper describes what has been achieved, our current work and what is planned.
Click here to view and download this paper.
THE USE OF COMPUTER-BASED TESTS TO CONSOLIDATE STATISTICAL CONCEPTS IN KENYA
David Stern1, Doug Stirling2, Ian Dale3 and Roger Stern3
1Department of Mathematics and Statistics, University of Maseno, Kenya
2Institute of Fundamental Sciences, Massey University, New Zealand
3Statistical Services Centre, University of Reading, United Kingdom
volloholic@hotmail.com
ABSTRACT
A customised African release of CAST (Computer Assisted Statistics Textbooks) has provided much-needed access to a series of modern statistics textbooks for many students in Africa. The recent addition of a collection of interactive CAST exercises has proved valuable in strengthening the learning of statistical concepts by students. A new CAST testing system has been developed to present the exercises as formal test questions. Students can practice beforehand with similar randomised exercises and can get immediate feedback as soon as the test is finished. This paper describes a pilot study in Kenya to evaluate the added benefit of using the CAST tests and is its first evaluation. Use of these tests is linked to an on-line course called “Statistics Made Simple”. The effectiveness of the tests is discussed.
Click here to view and download this paper.
E-LEARNING OF STATISTICS IN AFRICA
Ian Dale1, Cary Clark1, Roger Stern1, Sandro Leidi1 and David Stern2
1Statistical Services Centre, University of Reading, United Kingdom
2Department of Mathematics and Statistics, Maseno University, Kenya
i.c.dale@reading.ac.uk
ABSTRACT
Some general conclusions are drawn from the experience of converting two face-to-face statistics courses into facilitated, part-time e-learning courses, and running them for several cohorts of African students. The ‘Statistics in Applied Climatology’ course is for National Meteorological Services staff who need to strengthen their statistical skills for the analysis of climatic data. Part of SIAC was turned into an e-learning course in 2005, and since then it has been completed by over 200 people from 29 African countries. The ‘Statistics Made Simple’ course was initially developed for UK science students starting a research degree, who have not yet successfully learned key statistical concepts. A number of MSc Statistics students in Kenya also participated: they appreciated both the learning approach and the course content. The success of the courses can be attributed to the enthusiasm and determination of the participants, the high-touch facilitation, and the good quality of the materials
Click here to view and download this paper.
BUILDING STRENGTH FROM COMPROMISE; A CASE STUDY OF FIVE YEAR COLLABORATION BETWEEN THE STATISTICAL SERVICES CENTRE OF THE UNIVERSITY OF READING, UK AND MASENO UNIVERSITY, KENYA
James Musyoka1, David Stern1,2 and Roger Stern3
1Maseno University, Kenya
2African Institute of Mathematical Sciences – Next Einstein Initiative, Maseno University, Kenya
3Statistical Services Center, University of Reading, UK
jkmusyoka@maseno.ac.ke
ABSTRACT
Statistics teaching and practice at Maseno University has benefited immensely from its collaboration with the Statistical Services Centre (SSC) at the University of Reading. The SSC, a self-sustaining entity providing statistical consultancy to a wide range of clients in Africa, has also found a trusted pair of hands to help with its work in the developing world. The success of this collaboration is due to the long-standing working relationship between staff at these two institutions which this paper describes from both points’ of view. This collaboration is not without challenges and this paper also discusses the compromises made by both parties to make this collaboration work.
Click here to view and download this paper.
A MOBILE WEB FOR ENHANCING STATISTICS AND MATHEMATICS EDUCATION
Jamie Lentin¹, Anna H. Jonsdottir², David Stern³, Victoria Mokua³ and Gunnar Stefansson² ¹Shuttle Thread, Manchester, England
²University of Iceland; Science Institute, Reykjavik, Iceland
³Maseno University, Maseno, Kenya
gstefans@gmail.com
ABSTRACT
A freely available educational application (a mobile website) will be presented. This provides access to educational material and drilling on selected topics within mathematics and statistics with an emphasis on tablets and mobile phones. The application adapts to the student’s performance, selecting from easy to difficult questions, or older material etc. These adaptations are based on statistical models and analyses of data from testing precursors of the system within several courses, from calculus and introductory statistics through multiple linear regression. The application can be used in both on-line and off-line modes. Results presented include analyses of how the internal algorithms relate to dropout rates, passing a course and general incremental improvement in knowledge during a semester.
Click here to view and download this paper.
MEASURING THE EFFECTIVENESS OF USING COMPUTER ASSISTED STATISTICS TEXTBOOKS IN KENYA
Bernard Manyalla, Mbasu Zachariah, David Stern and Roger Stern
Department of Statistics, Maseno University, Kenya
zmbasu@yahoo.com
ABSTRACT
There is currently a big push to integrate technology into Kenyan education at all levels. Statistics is a subject that can benefit immensely from this improved access to technology. This study attempts to quantify the effect Computer Assisted Statistics Textbooks (CAST) has on student interest and performance. In Kenya CAST has now been used across all academic levels from schools to postgraduate, the implementations in a Diploma program is such that it allows for a quantitative analysis of performance against standardised grading, while the implementation in schools led to good data on student motivation. These implementations will be the focus of this paper.
Click here to view and download this paper.
REFLECTIONS ON USING TECHNOLOGY TO TEACH STATISTICS IN KENYA
David Stern
African Institute of Mathematical Science – Next Einstein Initiative Maseno University, Kenya
dstern@nexteinstein.org
ABSTRACT
Over the last ten years or more there have been numerous successful integrations of technology in statistics teaching in Kenya. Between them they show overwhelming evidence that technology can significantly improve student learning, but none have really impacted the “status quo”. This paper examines some past initiatives to identify commonalities that have contributed to their successes while also investigating why they have not been widely adopted. Many of the challenges will sound familiar to educators all over the world, for example the lack of academic recognition for good teaching, heavy workloads and institutional resistance to change. Questions are posed relating to how resources, support structures, and incentives or reward schemes might create an educational environment within which good initiatives can “go viral”
Click here to view and download this paper.
MSC TRAINING IN RESEARCH METHODS SUPPORT
D. A. Stern, R. Coe, R. D. Stern, and B. M. McDermott
Maseno University, Kisumu, Kenya
ICRAF, Nairobi, Kenya
Statistical Services Centre (SSC), University of Reading, UK
volloholic@hotmail.com
INTRODUCTION
Computers have been being brought into statistics education since the 1960’s (Sterling & Pollark, 1966). It is now accepted that there is a need to have a less theoretical approach to statistics education. In Kenya the need to change the way statistics is taught is recognised and accepted (Odhiambo, 2002). Internationally the GAISE (Guidelines for Assessment and Instruction in Statistics Education) reports (Aliaga, et al., 2005) (Franklin, et al., 2005) outline the problems as well as proposing solutions at the school and early college level. Applied statistics is a key component of research but ‘classical’ experiments and surveys have now become a relatively small part of the research portfolio in many organizations. This is illustrated by some of the international agricultural research institutes that have replaced their biometry or statistics units with wider service, research support units. The need for a modernised approach to mathematics education, now that computers have become ubiquitous, is being recognised and parallels the discussions about statistics education. This change in the mathematical subject as a whole broadens the training and reduces the emphasis on calculation. This implies that the changes needed are not making statistics training less mathematical; rather they are leading the way to bring mathematics and statistics training up to date. One the key advocates for changing the Maths curriculum, Wolfram states “I am not even sure if we should brand this subject as math, but what I am sure of is that this is the mainstream subject of the future.” (Wolfram, 2010) (Wolfram, The Practical Approach to Maths Education, 2010). The broadened role of statistics in research is already being rebranded and called ‘research methods support‘.
Click here to view and download this paper.
DEVELOPING STATISTICS EDUCATION IN KENYA THROUGH TECHNOLOGICAL INNOVATIONS AT ALL ACADEMIC LEVELS.
David Stern
Maseno University, Kenya
Volloholic@hotmail.com
INTRODUCTION
The value of using technology in education is no longer in question and has been show to work even in underprivileged areas (Mitra, 2010) (Mitra & Dangwal, 2010). For statistics in particular the guidelines and resources already exist and what needs to be done is generally accepted. However the changes have not yet happened in a sustainable way particularly in countries like Kenya where the teaching is still fundamentally based on the syllabus and approach from the 1960’s. The problem in Kenya, and in other countries in the region, is not a lack of initiatives but the lack of an educational culture which embraces and encourages these changes. The solution proposed is a change in postgraduate training designed also to nurture innovation in schools and universities.
Click here to view and download this paper.
The Seasonal Forecast by Stephen Kiplimo Kogo.
Seasonal forecasting is one of the major tasks performed by National Meteorological Services in different countries. They are able to do this with the help of historical records of rainfall, temperature or humidity among others and the sea surface temperatures. Climate Predictability Tool (CPT) software is used to conduct the forecasting upon receiving appropriate data inputs. Different statistical methods are employed in CPT which include canonical correlation analysis and principal component regression. These methods facilitate the creation of model(s) to be used in forecasting and proper understanding on how they do them is key in evaluating the usefulness of the resulting forecast. Usually the forecast is of the 3-month rainfall totals, but here we also consider forecasting the number of rain days and the occurrence of cold (rain-bearing) clouds
Click here to view and download this paper.