Training future statisticians to teach Statistics

Full text

Training Future Statisticians to Teach Statistics
Batanero, Carmen
Faculty of Education, Campus de Cartuja
Granada, 18071, Spain
E-mail: batanero@ugr.es
Díaz, Carmen
Faculty of Psychology, Campus de Cartuja
Granada, 18071, Spain
E-mail: mcdiaz@ugr.es
Statistics education is an important focus of interest for the International Statistical Institute (ISI) since
its foundation, as it is visible in the Education Committee started in 1948 and in the creation of the
International Association for Statistical Education in 1991. The IASE influence is clear in the increasing
presence of statistics at all the educational levels, and in the numerous publications, conferences and research
related to statistics education, including doctoral dissertations which are being carried out in departments of
statistics, mathematics education, psychology or education. It is, however, unusual that future statisticians
take a statistics education course as part of their undergraduate studies. In this paper we suggest that some
didactical training is an important part in the education of future statisticians and reflect about the
components and suitable teaching strategies in the initial didactical training of statisticians.
Interest of statistics education competence for statisticians
The many different possible areas of work for future statisticians are visible in the papers presented at
the ISI Sessions, as well as observing the widespread use of statistics in almost all human activities. A very
simple classification of professional statisticians is as follows (although the same person might play several
of these roles simultaneously or at different stages of his/her professional life):
1. Statistics teachers/lecturers at school, professional training or university (undergraduate or post graduate
training).
2. Theoretical statisticians who take part in research teams and develop new statistical theories, methods and
procedures.
3. Applied statisticians working as consultants for professionals who have specialized in other fields and to
whom they provide data analysis and sampling or experimental design services.
4. Producers of statistical data or reports for official agencies, financial or business institutions to which they
provide information, guide in decision making and facilitate the planning of their functioning.
5. Statistics educators who analyze the teaching and learning of statistics, try to understand its rules and design
didactical devices to help improving the functioning of didactical systems.
The interest of didactical knowledge is clear for teachers/lecturers and statistics educators. Moreover,
those statisticians who create new methods or theoretical models should be able to communicate them at
different complexity levels to other statisticians or scientists: e.g. in research papers, divulgation papers, training
courses directed to applied statisticians or textbooks. Didactical abilities will contribute to this communication
being clear and producing no misunderstanding in the potential users of statistics, since experience shows that
the use and interpretation of statistics are not always adequate (Morrison & Henkel, 1970; Harlow, Mulaik &
Steiger, 1997; Batanero & Díaz, 2006).
The didactical problems and possibilities of statistical consultancy have been debated in the Tokyo 2000
IASE Round Table Conference on Training Researchers in the Use of Statistics and in many Invited Paper
Sessions at different ICOTS or ISI Conferences. Statistical consultancy provides good training opportunities for
both the statistician and the client. Not only do statisticians need to understand enough about their colleagues’
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

disciplines to be effective consultants, but they also can help researchers to learn more statistics and acquire
enough appreciation of statistical concepts so that the collaboration may be more productive.
On the one hand, the client professional should be able to clearly communicate his/her problem to the
statistician and provide him /her with a basic understanding of the topic, the aim of the study and the concepts
involved in the same (Jolliffe, 2001). The statistician, in turn, can help clarify the questions that the researcher
wishes to consider, and both the researcher and the statistician will learn from each other as they move towards a
common understanding. Statisticians need also to be trained to communicate statistical information and results
from statistical analyses in a language accessible to the researcher and meaningful for the potential audience of
the research report. They have a unique opportunity to help their clients acquire statistical literacy using their
own data and the problems in which they are interested (Belli, 2001). All of this justifies that the training of
statistical consultants include some didactical elements.
Finally, the shifting of the economy from tangible products to intangible service-based activities has
increased the difficulty of the work of statisticians producing public statistics and poses several challenges for
statistical education, including the training of official statisticians, as new statistical concepts and methodology
are being created, but also educating the users of these official statistics (Cheung, 1998). On one hand the
statistical information produced by these offices has a special role in modern societies in enabling people getting
information and reacting to social, political and economic phenomena. Therefore, statisticians need to
communicate and diffuse statistics, not only as a technique for dealing with quantitative data, but also as a
culture, in terms of capacity to comprehend the logical abstraction, which makes the quantitative study of
collective phenomena possible (Murray & Gal, 2002). Statistical offices need moreover collaboration of citizens
to collect their data since no statistical method, even if reliable and efficient, can produce sound results from
invalid data. This is why “to maintain high response rates, and therefore provide data collections that are
widely regarded as providing high quality outputs, a statistical office needs to be trusted by the public, and in
particular by its respondents” (McDonald, 2001, p. 121). All of these have increased the interest of statistical
offices towards education, and even in some cases lead to collaborative projects such as Census at School
(http://www.censusatschool.ntu.ac.uk/).
Since didactical problems and situations seem to be pervasive in the work of statisticians, whatever
this work is, it is worthwhile to offer future statisticians some didactical courses in their initial training. In
this paper we discuss what the components of didactical knowledge, beyond the knowledge of statistics and
probability itself, could be, and analyse some activities used in the training of statisticians at the University
of Granada. We hope these ideas may encourage other colleagues to organise courses of statistics education
directed to statisticians at undergraduate or post graduate levels.
What is Didactical Knowledge and How Can it be Taught?
A wide statistical knowledge, even when essential, is not enough for teachers/lecturers/statisticians to
be able to teach statistics. A review of the literature on teachers’ professional knowledge (e.g. Shulman,
1986, Cooney, 1999, Llinares & Kraisner, 2006) show that teachers draw on three main interrelated
knowledge bases that evolve with practice: knowledge of content, knowledge of teaching processes and
knowledge of their students. Research focused on teacher's training is producing a great deal of information
about this 'didactical knowledge', which includes the following complementary aspects:
• Epistemological reflection on the meaning of concepts to be taught (e.g. reflection on the different
meanings of probability). A main point in preparing teachers is the epistemological reflection, which can
help them to understand the role of concepts within statistics and other areas, its importance in students'
learning and the students' conceptual difficulties in problem solving. Statistics and probability are young
areas and the formal development or probability was linked to a large number of paradoxes, which show
the disparity between intuition and conceptual development in this field (Székely, 1986). This
comparative difficulty is also shown in the fact that, even when Kolmogorov’s axiomatic foundation was
generally accepted in 1933, professional statisticians still debate about different views of probability and
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

different methodologies of inference (Batanero, & Díaz, 2006). Biehler (1990) also suggested that this
“meta-knowledge” should include a historical, philosophical and cultural perspective on statistics and its
relations to other domains of science.
• Critical capacity to analyse textbooks and curricular documents, and experience in adapting this
knowledge to different teaching levels, and students’ various levels of understanding. This includes
(Steinbring, 1990) organizing and implementing statistics projects, experiencing students’ multiple forms
of work and understanding; experiments, simulations and graphical representations not just as
methodological teaching aids, but rather as essential means of knowing and understanding statistics. It is
important to remark that general principles that are valid for geometry, algebra or other areas of
mathematics cannot always be applied (Batanero, Godino & Roa, 2004). For example, in arithmetic or
geometry an elementary operation can be reversed and this reversibility can be represented with concrete
materials. This is very important for young children, who still are very linked to concrete situations in
their mathematical thinking. For example, when joining a group of two apples with another group of
three apples, a child always obtain the same result (5 apples); if separating the second set from the total
he/she always returns to the original set; no matter how many times this operation is repeated. These
experiences are very important to help children progressively abstract the mathematical structure behind
them. In the case of random experiment we obtain different results each time the experiment is carried
out and the experiment cannot be reversed (we can not get the first result again when repeating the
experiment).
• Capacity to develop and analyse assessment tests and instruments and interpret students’ responses to the
same; prediction of students' learning difficulties, errors, obstacles and strategies in problem solving
(e.g., students strategies in comparing two probabilities; students' confusion between the two terms in a
conditional probability).
• Experience with good examples of teaching situations, didactic tools and materials (e.g., challenging and
interesting problems; Galton board, simulation, calculators, etc.) and knowing the role of them in
instruction. For example, even when simulation or experimentation with random generators, such as dice
and coins, have a very important function in stabilizing children’s intuition and in materializing
probabilistic problems, these experiences do not provide the key to how and why the problems are
solved. Teachers should realize that it is only with the help of combinatorial schemes or tools like tree
diagrams that children start to understand the solution of probabilistic problems, due to the
complementary nature of classical and frequentist approaches to probability. Moreover, the teaching of
stochastics should provide a pedagogical space, where processes are given more value than facts, ideas
are preferred to techniques, and a great diversity of problems involving other areas are proposed to help
students develop positive attitudes towards this topic (Espasadin, 2006).
Examples of Teaching Situations Oriented to Teachers’ Didactical Training. The Experience at the
University of Granada
It is important to find suitable and effective ways to teach this "didactical knowledge" to teachers –in
this case the future statisticians. Since students build their knowledge in an active way by solving problems
and interacting with their classmates, we should use this same approach in training the teachers, especially if
we want them later use a constructivist and social approach in their teaching (Even and Lappan 1994; Ball &
Bass, 2000, Jaworski 2001). An important view is that we should give teachers more responsibility in their
own training and help them to develop creative and critical thinking (Shulman 1986). That is why we should
create suitable conditions for teachers to reflect on their previous beliefs about teaching and discuss these
ideas with other colleagues. Below we describe some examples of possible didactical activities to train
teachers and statisticians in the didactic knowledge related to statistics. These activities are complementary
from various viewpoints and can be used to provoke teachers' and statisticians’ reflections about the meaning
of elementary stochastic notions, students' difficulties and obstacles, didactical methodology and materials.
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

1. Solving paradoxical problem situations and reflecting on its content. The history of statistics is full
of examples of apparently simple problems that challenged the minds of brilliant mathematicians (see for
example, Székely, 1986). In solving these problems, the future statisticians might reflect on the complex
meaning of stochastic notions, share and predict some learning difficulties in their future students and learn
some principles of teaching and assessment. For example, we ask the future statisticians to find the best
strategy in the following game that has been designed to teach probability at secondary level (this situation is
analysed in detail in Batanero, Godino & Roa, 2004).
Game: We take three counters of the same shape and size. One is blue on both sides, the second is red
on both sides and the third is blue on one side and red on the other. We put the three counters into a
box, and shake the box, before selecting a counter at random. After selecting the counter we show one
of the sides. The aim of the game is to guess the colour of the hidden side. We repeat the process,
putting the counter again in the box after each new extraction. We make predictions about the hidden
side colour and win a point each time our prediction is right.
The lecturer organises the teaching time in several stages: a) playing the game; b) time to individually
look for the best strategy; c) playing again to check the teachers’ conjecture against results in the experiment;
d) general debate where students defend their different preferred strategies and try to give a mathematical
proof about why one of them is best; e) didactical reflection on the game, the statistical content behind it,
what was learned in proving the best solution; stages in a didactical situation and predictable difficulties of
students in the activity.
2. Analysing assessment items or tasks and some students responses to the same. The aim is to reflect on
the complex meaning of stochastic notions, show the utility of the task in teaching and assessment and
predict some learning difficulties. For example we ask the future statisticians to imagine they want to try a
treatment they suspect may alter performance on a certain task. After comparing the means of a control and
an experimental groups (say 20 subjects in each sample) they get (t = 2.7, d.f. = 18, p = 0.01) in a simple
independent means t-test. We ask future statisticians to discuss what is the meaning of this significant result
and decide which of the following sentences (if any) are true (Krauss & Wassner, 2002):
• They have absolutely disproved the null hypothesis (the hypothesis that there is no difference between
the population means).
• They have found the probability of the null hypothesis being true.
• They have absolutely proved your experimental hypothesis (that there is a difference between the
population means).
• They can deduce the probability of the experimental hypothesis being true.
• If they decide to reject the null hypothesis, they know the probability that they are making the wrong
decision.
• They have a reliable experimental finding in the sense that if, hypothetically, the experiment were
repeated a great number of times, they would obtain a significant result on 99% of occasions.
In fact none of the previous statements is true, although many people, including students, scientists and
even methodology instructors have erroneously believed some of this statements or other equivalent
statements to be correct in Krauss & Wassner (2002) and other previous research. After analysing the
sentence above as well as the correct definition of a p-value, we can introduce the future statisticians to some
of the elements in what has been called: “the statistical tests controversy” and ask them read some summaries
of the same (e.g. Batanero & Diaz, 2006). Future teachers might analyse the mathematical concepts involved
in a statistical test (e.g., statistics and parameters, null and alternative hypotheses, significance level and p-
value, sampling distribution, power, etc.) and build a concept map showing the relationships between them.
They later can try to predict and classify possible errors related to each of these concepts and compare with
errors described in the literature, a summary of which would be provided by the lecturer. A deeper level of
analysis would involve discussing the psychological and philosophical explanations for the persistence of
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

these errors, including analysing the different approaches to statistical tests by Fisher and Neyman-Pearson,
and the difference between reasoning by contradiction and the logical reasoning behind a statistical test.
3. Analysing statistical data or graphs in published reports, media or research journals. Future
statisticians have a special interest in any activity related to the analysis of interpretation of data, so that we
can also use these type of activities to provoke didactical reflection. For example, we can ask them to
comment on the appropriateness of the graph presented in Figure 1, which was published in a major Spanish
journal. The graph served to illustrate the possible reduction of deathly accidents as a consequence of
applying the new traffic regulations. According this new law higher fines (€ 301 to € 1,500) might be
imposed for committing the most serious traffic infractions that the Traffic Spanish law provides, such as
driving in a negligent manner, speeding, driving without lights or parking in dangerous places. A driver who
is fined three times for committing any of these infractions within a period of 2 years, may also have his
driver's license definitively revoked. (Incidentally, Spain has the highest road mortality rate in Europe.)
Independently of the accuracy or lack or accuracy of data and its interpretation, the graph is in itself a
compendium of frequent errors in a graphical display of data that students might discover. Other similar data
and incorrect interpretation of statistics in press can be downloaded from the web server “Malaprensa” (bad
press) at http://www.malaprensa.com. More sophisticated analyses include criticising use of statistics in
research reports published in scientific journals, such as using statistical procedures without taking into
account their assumptions, too small sample sizes, incorrect multiple comparisons or incorrect interpretation
of statistical analyses.
Figure 1. Example of misleading graph
4. Analysing textbooks, statistical software or didactic resources. For example, we might ask future
statisticians to compare curricular guidelines for the same educational level in different Spanish autonomies
(geographical regions with self capacity in education) and reflect on their adequacy. We might ask them to
suggest possible ways to introduce the concepts to specific students (e.g. secondary students who still do not
have a deep mathematical knowledge). Another activity is analysing the different sections included in the
journal Teaching Statistics, oriented to teachers of 9-18 year-olds students; this analysis may provide an idea
of the different types of knowledge required in the professional work of statistics teachers: history of
statistics, didactical research, resources, problems, assessment, data sets, computer corner, etc. They might
also visit some of the many didactical resources for teaching statistics on the Internet, for example those
listed in the links section at the IASE web page (http://www.stat.auckland.ac.nz/~iase/), at the ISI Statistical
Literacy Project web page (http://www.stat.auckland.ac.nz/~iase/islp/home) or at other statistics education
web servers. Unfortunately most of these resources are written in English language (so that they are not
useful for teachers who have to teach in a different language), although little by little the situation is
changing and an increasing number of Internet resources in Spanish and other languages are being made
available.
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

Other complementary activities are as follows: A) Reading and discussing short papers by prestigious
statisticians where didactical problems are set; for example the paper by David Moore (1997), summarise the
main points made in the paper by the author and decide with which of these points they agree and why. Other
different students might read papers that present complementary positions about some of these ideas (e.g.
Rossman & Short, 1995; Albert, 1995). B) The lecturer summarises a theoretical model described in
statistics education research and a collective activity is organised to apply the model in the analysis of
students’ statistical work. For example, the lecturer summarises Wild & Pfannkuch’s (1999) model for
statistical thinking and students in pairs try to get examples of fundamental modes of statistical thinking
(recognising the need for data, transnumeration, perception of variation, reasoning with statistical models
and integrating statistics in the context) in some examples of statistical projects carried out by secondary
school students. C) Some students interested in getting a better score in the course carry out a short
bibliographical survey and a synthesis of a theme which interests them and prepare a short historical,
epistemological or didactical essay. D) Finally, a compulsory activity for all the participants is preparing a
didactic unit to introduce a statistical topic to a specific type of student, where future teachers and
statisticians select the statistical topic and type of student (e.g. introducing correlation to psychologists). In
this didactic unit, the future teachers and statisticians have to apply the knowledge acquired along the course
and include the objectives, contents, activities, teaching methodology, didactic resources and software
needed and assessment criteria. Students can consult examples of didactic units in Spanish and foreign
statistics textbooks and curricular materials available at the Department library.
These and other similar activities have been experimented along the past 10 years at different courses
in Statistics Education at the University of Granada, Spain and in some Latin American countries. One of
these courses has been included since 1996 as an optional topic within the Major in Statistics Sciences. The
course has a length of 60 teaching hours and the majority of students are in the 4º or 5º year of studies, and,
therefore, have a good training in statistics. Consequently this course is focused only in the didactical content
and the types of activities described above. Since the content is very sparsely widespread in journals and
books, a summary of the same, which also include possible activities for the course was developed in a basic
textbook (Batanero, 2001) divided into 5 chapters:
1. Introduction: Statistics Education, historical perspective, statistics education within statistics,
psychology and mathematics education: associations, journals, conferences and sources of information.
2. Epistemological foundations: Statistics. Current tendencies. Different conceptions of randomness and
probability. Fundamental stochastic ideas. Exploratory data analysis. Association and causality.
Inference and induction. Different schools of inference. Principles of multivariate analysis.
3. Research on statistical reasoning and learning difficulties: Cognitive development: Piaget and
Fischbein. Heuristics and biases in stochastic reasoning. Didactical research: errors, difficulties,
misconceptions related to randomness. probability, graphing, averages, dispersion, association,
distributions and inference.
4. Curriculum and instruction: Goals in the teaching of statistics. Stochastic Phenomenology. Curriculum
and Instruction. Curricular components. Educational theories and teaching approaches. Assessment.
Teaching resources. Computers and calculators. Recources in Internet.
5. Teaching statistics through project work: Principles in working with projects. Examples of statistical
project to introduce statistics along secondary education.
Final Remarks
In this paper we have described our experience in training future statisticians in statistics education at
the University of Granada. It is clear that the time available (60 teaching hours) is not enough to get even a
basic mastering of didactic knowledge by future statisticians and moreover the structure of the course does
not include teaching practices, which are the most important part of professional development. Even with
these limitations we think this type of course can provide future statisticians with some didactical preparation
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

that will be useful for their future work at the time it increases their interest and sensibility towards the good
practice of statistics.
Another remark is the scarcity of teaching materials devoted to the didactical preparation of statistics
teachers as compared with materials related to teaching mathematics or teaching other topics. The significant
research efforts focusing on mathematics teacher education and professional development in the past decade
have not been reflected in statistics education. This is evident in conferences (e.g., the ICMI Study 15),
journals (e.g., Journal of Mathematics Teacher Education), surveys, and books that hardly take into account
the particular case of statistics. This omission needs to be addressed by promoting research specifically
focussed on the education and professional development of teachers to teach statistics (Shaughnessy, in
press). This need was recognised by both the International Association for Statistical Education and the
International Commission on Mathematical Instruction who are jointly organising a Study around this
specific topic with two components: a Study Conference and the production of a book
(http://www.ugr.es/~icmi/iase_study/). We hope this initiative will lead to a better preparation of teachers,
which in turn serves to improve statistics education at all educational levels.
ACKNOWLEDGEMENT: This work was supported by the projects SEJ2004-00789, MEC, Madrid, and
FQM 126, Junta de Andalucia, Spain
References
Albert, J. (1995). Teaching inference about proportion. Using Bayes and discrete models. Journal of
Statistics Education, 3(3). On line: http://www.amstat. org/publications/jse/v3n3/albert.html.
Ball, D. & Bass, H. (2000). Interweaving content and pedagogy in teaching and learning to teach: Knowing
and using mathematics. In J. Boaler (Ed.), Multiple perspectives on the teaching and learning of
mathematics (pp. 83-104). Westport, CT: JAI/Ablex.
Batanero, C. (2001). Didáctica de la estadística (Statistics education). Granada: Author. On line:
http://www.amstat.org/publications/jse/v12n1/batanero_link1.html.
Batanero, C. & Díaz, C. (2006). Methodological and didactical controversies around statistical inference.
Actes du 36iémes Journées de la Societé Française de Statistique. CD ROM. Paris: Societé Française
de Statistique.
Batanero, C., Godino, J. & Roa, R. (2004). Training teachers to teach probability. Journal of Statistical
Education, 12(1). On line: http://www.amstat.org/publications/jse/v12n1/batanero.html.
Belli, G. (2001). The teaching/ learning process in university statistical consulting labs in the United States.
In C. Batanero (Ed.), Training Researchers in the Use of Statistics (pp. 325-338). Granada, Spain:
International Association for Statistical Education.
Biehler, R. (1990). Changing conceptions of statistics: A problem area for teacher education. In A. Hawkins
(Ed.), Training Teachers to Teach Statistics. Proceedings of the International Statistical Institute
Round Table Conference, (pp. 20-38). Voorburg: International Statistical Institute.
Cheung, P. (2002). Developments in official statistics and challenges for statistical education. In B. Phillips
(Ed.), Proceedings of ICOTS-6. Cape Town: International Association for Statistical Education.
Cooney, T. J. (1999). Conceptualizing teachers’ way of knowing. In D. Tirosh (Ed.), Forms of mathematical
knowledge; Learning and teaching with understanding (pp. 67-84). Dordrecht: Kluwer
Even, R., & Lappan, G. (1994). Constructing meaningful understanding of mathematical content. In D. B.
Aichele and A. F. Coxford, (Eds.), Professional Development for Teachers of Mathematics (pp. 128-
143). Reston, VA: National Council of Teachers of Mathematics.
Espasadin, C. (2006). Stochastics and the professional knowledge of teachers. In A. Rossman & B. Chance
(Eds.), Proceedings of the Seventh International Conference on Teaching Statistics. Salvador (Bahia),
Brazil: International Association for Statistical Education.
Harlow, L. L., Mulaik, S. A., & Steiger, J. H. (1997). What if there were no significance tests? Mahwah, NJ:
Erlbaum.
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz

Jaworski, B. (2001). Developing mathematics teaching: teachers, teacher educators and researchers as co-
learners. In L. Lin & T. J. Cooney (Eds.), Making sense of mathematics teacher education (pp. 295-
320). Dordrecht: Kluwer.
Jolliffe, F. (2001). Learning from experience. In C. Batanero (Ed.), Training Researchers in the Use of
Statistics (pp. 355-370). Granada, Spain: International Association for Statistical Education.
Krauss, S., & Wassner, C. (2002). How significance tests should be presented to avoid the typical
misinterpretations. In B. Phillips (Ed.). Proceedings of ICOTS 6. Cape Town: International Association
for Statistical Education.
LLinares S., & Kraiwer K. (2006) Mathematics (student) teachers and teacher educators as learners. In A.
Gutierrez, & P. Boero (Eds) Handbook of Research on the Psychology of Mathematics Education (pp.
429 – 459). Rotherdam / Taipei: Sense Publichers.
Moore, D. S. (1997). Bayes for beginners? Some pedagogical questions. In S. Panchapakesan (Ed.),
Advances in statistical decision theory (pp. 3-17). New York: Birkhäuser.
Murray, S., & Gal, I. (2002). Preparing for diversity in statistics literacy: Institutional and educational
implications. In B. Phillips (Ed.), Proceedings of ICOTS 6. Cape Town. International Association for
Statistical Education.
Mc Donald, S. (2001). Practical and educational problems in sharing official micro data with researchers. In
C. Batanero (Ed.), Training Researchers in the Use of Statistics (pp. 119-128). Granada, Spain:
International Association for Statistical Education.
Morrison, D. E., y Henkel, R. E. (Eds.) (1970). The significance tests controversy. A reader. Chicago:
Aldine.
Rossman, A., & Short, T. (1995). Conditional probability and education reform: Are they compatible?
Journal of Statistics Education, 3 (2). On line:
http://www.amstat.org/publications/jse/v3n2/rossman.html
Shaughnessy, J. M. (In press). Research on statistics learning and reasoning. In F. Lester (Ed.), Second
Handbook of Research on Mathematics Teaching and Learning. Greenwich, CT: Information Age
Publishing, Inc., & NCTM.
Shulman, L. (1986). Paradigm and research programs in the study of teaching: A contemporary perspective.
In M. C. Witrock (Ed.), Handbook of research on teaching (pp. 3-36). New York: Macmillan.
Steinbring, H. (1990). The nature of stochastic knowledge and the traditional mathematics curriculum -
Some experience with in-service training and developing materials. In A. Hawkins (Ed.), Training
Teachers to Teach Statistics. Proceedings of the International Statistical Institute Round Table
Conference (pp. 2-19). Voorburg: International Statistical Institute.
Székely, G. J. (1986). Paradoxes in probability theory and mathematical statistics. Dordrecht, The
Netherlands: Reidel.
Wild, C., & Pfannkuch, M. (1999). Statistical thinking in empirical enquiry. International Statistical Review,
67(3), 221-248.
ABSTRACT
In this presentation we first suggest that some didactical training is an important part in the education of
future statisticians and reflect about what other components would be needed and what are suitable teaching
strategies in the initial and on-going didactical training of statisticians We finally present and analyse
examples of activities used in courses of statistics education directed to statisticians. These reflections take
into account the experience at the University of Granada, where an optional Statistics Education course was
included ten years ago in the Major in Statistical Sciences and Techniques, as well as the experience in other
courses given to in service statistics lecturers in several Latin American countries.
International Statistical Institute, 56th Session, 2007: Carmen Batanero, Carmen Diaz