CMESG Working Group 2010: Data Analysis and Visualization

From MathWiki

CMESG Working Group 2010 on Data Analysis and Visualization

A working group for the 2010 CMESG Conference at Simon Fraser University on May 21 to 25.

New links for our groups should probably start with the name CMESG Working Group 2010: Data Analysis and Visualization -- or maybe just CMESG:DAV for short -- to avoid collisions with other groups. e.g. CMESG Working Group 2010: Data Analysis and Visualization: Draft Report for a draft report. The link looks shorter if one displays alternative text, e.g. [[CMESG Working Group 2010: Data Analysis and Visualization: Draft Report|draft report]] produces draft report.
Table of contents

Links for the working sessions

  • Gapminder
    • TED talk by Hans Rosling (
    • Download [ Gapminder software]
    • How to use Gapminder ( on your own data.

  • USCOTS (


Attending to Data Analysis & Visualizing Data

Linda Gattuso, Georges Monette, Veda Abu-Bakare

Data analysis and visualization of data have become an integral part of the elementary and secondary curricula as well as for pre-service teachers’ courses. Not only must we consider how to make these important and relevant topics meaningful to students but a major hurdle to be overcome is that of ensuring that our future teachers have the confidence and knowledge to attend to this strand of the curriculum.

It is important that students of all ages and teachers develop statistical thinking in the manner of a statistician. How can we guarantee that this will occur?

What activities, experiments, simulations, and resources can we use and develop with students and pre-service teachers? In what ways can the technology that is readily available motivate and deepen understanding? How can we use existing indices and databases such as Statistics Canada’s Consumer Price Index, E-Stat and CANSIM to empower our students and pre-service teachers and help them make sense of our data-centric world?

Further, we must consider the fact that this teaching generally takes place within mathematics or mathematics methods courses (in the case of teaching training). How can we promote the synergy of these two disciplines, that of mathematics and that of data analysis while fostering learning? Or moreover, do these two disciplines originate in two distinctive and irreconcilable ways of thinking?

In our group, we will present activities drawn from research articles and share personal experiences so that we may elicit discussions that can shed light on the questions mentioned above. We encourage active and productive participation that we hope will give participants new resources to support them in preparing teachers and in the actual teaching of data analysis and visualization of data.


  • Burrill, G. and Elliott, P. (2006). Thinking and reasoning with data and chance. NCTM, Reston Virginia.
  • Tufte, E. (1983). The visual display of quantitative information. Graphics Press, Cheshire, Conn.
  • Tufte, E. (1990). Envisioning information. Graphics Press, Cheshire, Conn.

L’analyse des données et leur représentation

Linda Gattuso, Georges Monette, Veda Abu-Bakare

L’analyse et la représentation des données font maintenant partie intégrale des programmes d’enseignements élémentaire et secondaire ainsi que de la formation des futurs enseignants. Non seulement doit-on considérer comment faire en sorte que ces sujets soient importants et pertinents pour les élèves mais il faut aussi s’assurer que les futurs enseignants acquièrent la confiance et les connaissances nécessaires pour réussir cette tâche. L’important est que les élèves de tous âges et les enseignants développent la pensée statistique tout en expérimentant le parcours du statisticien. Comment peut-on s’en assurer?

Quelles activités, expérimentations, simulations et ressources peuvent être utilisées et/ou développées avec les élèves et les enseignants en formation. De quelle façon la technologie déjà existante peut-elle motiver et approfondir la compréhension ? Comment peut-on utiliser les indices et les bases de données comme celles sur le site de Statistique Canada pour outiller nos élèves et futurs enseignants et les aider à donner un sens à notre monde centrée sur les données ?

De plus, il faut tenir compte du fait que cet enseignement est réalisé généralement dans le cadre de cours de mathématiques ou de didactique des mathématiques (dans le cas de la formation à l’enseignement). Ces deux enseignement, celui de la mathématique et celui de l’analyse des données, peuvent-il se compléter et s’enrichir mutuellement et favoriser l’apprentissage. Ou encore ces deuz disciplines proviennent-elles de deux pensées différentes et inconciliables ?

Dans notre groupe, nous proposerons certaines activités tirées d’articles de recherches ou d’expériences personnelles que nous pourrons simuler et ainsi amorcer une discussion tentant d’apporter des réponses aux questions ci haut. Nous favoriserons une participation active et constructive qui nous l’espérons donnera aux participants de nouvelles ressources pour les soutenir dans la formation et l’implémentation de l’enseignement de l’analyse et de la visualisation des données.


  • Burrill, G. and Elliott, P. (2006). Thinking and reasoning with data and chance. NCTM, Reston Virginia.
  • Tufte, E. (1983). The visual display of quantitative information. Graphics Press, Cheshire, Conn.
  • Tufte, E. (1990). Envisioning information. Graphics Press, Cheshire, Conn.


We can flesh out plans here or in wiki pages we link to from here. e.g. Plan Day 1, Plan Day 2, Plan Day 3.

Introduction: A plug for probability and statistics

Why is it important that we value this area? How are data analysis and chance addressed in various curricula?

Tasks: 1) Task from Watson, A. & Mason, J. (2005). Mathematics as a constructive activity: Learner generated examples. Lawrence Erlbaum Associates, New Jersey, page 9. What feature of the data set is captured by each of the statistics?

Mode, Median and Mean Construct a data set of seven numbers for which the mode is 5, the median is 6, and the mean is 7. Alter it to make the mode 10, the median 12, and the mean 14; alter it to make the mode 8, the median 9, and the mean 10. Is it possible to preassign any value to each mode, median, and mean independently and restrict the data set to just five data points? How small a data set can achieve a preassigned mode, median, and mean? How much choice of data set is there then?

Big Ideas:

1) Variation all around us - description of phenomena

2) Randomness, its patterns and its quirks, applications relating to randomness (codes, etc.)

3) Simulations and what can be learned from them

4) Relationships between variables (correlation, causation, etc.)

GM: Fascinating. A very effective illustration of the point that observational data doesn't justify causal conclusions. However, I think that stopping at 'correlation is not causation' is akin to preaching total abstinence. In fact most causal 'judgments' influencing everything from public policy to personal everyday choices need to be based on observational data. Many decisions need to be made long before relevant experimental data is available and a strict insistence on correlation is not causation is not helpful. Recall Fisher's strong defence of tobacco in the late 50s because, after all, all the evidence against tobacco was based on observational data. We would like people to understand the primacy of experimental data over observational data for causal inference but we would also like to help them develop the judgment to assess causal claims based on observational studies. This requires some understanding of confounding factors, mediating factor and the difference between conditional association controlling for confounding factors versus marginal or unconditional association.

5) Effective displays of data

6) Information vs Knowledge

7) Writing in/about Statistics Miller, J. (2006). How to communicate statistical findings: an expository writing approach. Chance 19(4), 43-49.

8) Reading in/about Statistics

VA: Here is an abstract that I have submitted for a JSM Statistical Education Roundtable August 2010, Vancouver. Will know in March if accepted...

Abstract: Readings for Intro Stat courses Many of us who teach Intro Stat courses rely for the most part on the textbook and related materials from the publisher. In this round table, we consider using recent books on probability and statistics that have been written for the lay audience such as The Black Swan: The Impact of the Highly Probable by Nassim Nicholas Taleb and Struck by Lightning: The Curious World of Probabilities by Jeff Rosenthal. What important statistical ideas can we convey by exposing our students to the popular literature? How can this literature be used to give our students an appreciation of the discipline, its achievements and its controversies?


Add here links related to the working group:

Miscellaneous Links

Add here links of more general interest:

On visualization

On statistics education

On the misuse of statistics

Sources of data

  • Copied with permission from Peggy Ng:
  1. Data analysis issues
    1. Numbers Racket?
    2. Simpson’s paradox
      - Black-White Income Gap:
    3. Data collection
      Can you trust a dataset where more than half the values are missing?
    4. Experimental design
      A `Politically Robust' Experimental Design for Public Policy Evaluation:
    5. T-test
      A Theory of Mind investigation into the appreciation of visual jokes in schizophrenia:
    6. Setting a comparative question
      Dress for Less and Less
    7. The Indiana Jones of Economics, Part I:
    8. Who's Watching American Idol?
    9. Just be careful ….
      1. 81% in Poll Say Nation Is Headed on the Wrong Track:
      2. The Ebb and Flow of Movies: Box Office Receipts 1986 – 2007:
      3. Dennis the Denver dentist and Laura the Louisiana lawyer:
      4. Who Wants To Be Rich?
      5. Self-reported delinquency among young people in Toronto:
    10. Just wrong ….
      Data on minority doctorates suppressed:
    11. Just wonder ….
      1. Who gets any sleep these days? Sleep patterns of Canadians:
      2. Canada's bison industry:
      3. Diamonds: An update:
      4. Canada's trade in beer:
      5. Scientists and engineers and urban growth:
      6. A new look at commuting distance:
      7. Delayed transitions of young adults:
      8. Who's calling at dinner time?
    12. Just curious ….
      The arrival of 'merit-blind admissions:
  2. Statistical graphics
    1. Online atlas of the millennium development goals:
    2. OECD Factbook 2008: Economic, Environmental and Social Statistics:
    3. The First Measured Century – Trends Challenge:
    4. Hans Rosling: Debunking third-world myths with the best stats you've ever seen:
    5. Why Are American Presidential Election Campaign Polls So Variable When Votes Are So Predictable?
  3. Personal income tax
    Federal Personal Income Tax: Slicing the Pie:
  4. Public finance in Canada
    1. Government finance: Revenue, expenditure and surplus:
    2. Government spending on social services:
    3. Employment trends in the federal public service:
  5. Canadians
    1. Snapshot of past 100 years in Canada:
    2. Earnings and Incomes of Canadians Over the Past Quarter Century, 2006 Census:
    3. Canada's income redistribution strategy: take from the rich, give to the median:
    4. Income of Canadians:
    5. Family income:
    6. Income of individuals:
    7. Family earnings instability:
    8. Earnings instability:
    9. Payday loans:
  6. Income inequality in Canada
    1. Why Inequality Matters in 1,000 Words or Less:
    2. 2006 Census: Earnings, income and shelter costs:
    3. A Quarter Century of Economic Inequality in Canada:
    4. Trends in income inequality in Canada from an international perspective:
    5. Income inequality and redistribution:
    6. High-income Canadians:
    7. Persistence of low income among working-aged unattached individuals:
    8. Inequality in wealth:
  7. Students
    1. Participation in post-secondary education:
    2. Is postsecondary access more equitable in Canada or the United States?
    3. Why are youth from lower-income families less likely to attend university?
    4. High school dropouts returning to school:
    5. Adult learning in Canada: Characteristics of learners:
    6. Sources of growth in degree holders across urban and rural Canada:
    7. Performance
      1. Performance of Canada's youth in science, reading and mathematics:
      2. Survey of Canadian Attitudes toward Learning:
    8. Costs
      1. How students fund their postsecondary education:
      2. Who gets student loans?
    9. Benefits
      1. Education and earnings:
      2. Impact of compulsory school laws on educational attainment and earnings:
      3. Adult education and its impact on earnings:
      4. Pathways from education to the labour market among Canadian youth:
  8. Health
    1. Access
    2. Access to health care services:
    3. Going to the doctor:
    4. Out-of-pocket spending on prescription drugs:
    5. Life and death
      1. Birth outcomes by neighbourhood income and recent immigration in Toronto:
      2. Income inequality and working-age mortality:
      3. Inequality and health:
      4. Pain and inequality:
      5. Correlates of medication error in hospitals:
  9. WORK
    1. Pay and hours
      1. Payroll employment, earnings and hours:
      2. Understanding regional differences in work hours:
      3. Hours polarization revisited:
      4. General Social Survey: Paid and unpaid work:
      5. How families respond to layoffs:
      6. The death of a spouse and the impact on income:
      7. International mobility: A longitudinal analysis of the effects on individuals' earnings:
      8. Globalisation, Jobs and Wages:,3425,en_2649_201185_38796127_1_1_1_1,00.html
    2. Productivity
      1. GDP per capita and productivity in Canada and the United States:
      2. Labour productivity, hourly compensation and unit labour cost:
      3. Work stress and job performance:
      4. Health Reports: Job satisfaction, stress and depression:
      5. On sick leave:
    3. Women
      1. Wives as primary breadwinners:
      2. Returning to work after childbirth:
      3. Employment growth among lone mothers in Canada and the United States:
      4. Employment and earnings among lone mothers:
      5. Gender differences in quits and absenteeism:
    4. Work-related issues
      1. Diverging trends in unionization:
      2. Balancing career and care:
      3. Other interesting daily posting of public policy and Canadian data on HEALTH, ENVIRONMENT, INNOVATION, RETIREMENT & RELATED ISSUES . .. … please visit
  10. Income inequality in U.S.
    1. Class matters:
    2. New data show income concentration rose again in 2006:
    3. Controversies about the Rise of American Inequality: A Survey:
    4. Inequality and prices:
    5. Portraits of the Assets and Liabilities of Low-Income Families:
    6. Income inequality in the world
      1. Vast majority income: