Scientific Program > Topic 3 > Session 3E >
Presentation 3E3. Survey Sampling: Learning by Doing. A Twenty Years Graduate Level Teaching Experience.

Presenter
Jean-Hugues Chauchat (France) chauchat@univ-lyon2.fr

 

Presentation Abstract

For the past twenty years, we have been using an original technique to teach statistics and survey sampling methods to postgraduates studying economics and statistics at University Lumiere, Lyon, France. The students must put their knowledge into practice by carrying out a survey sample for a client who they will have found themselves. This may include a marketing study for a shop, a brand or a public service, or measuring the audience ratings of a radio station or local television station...

These applied studies enable the students to get hands-on experience in the world of work. They learn how to understand the client's needs, propose a tailored study, draw up and try out the questionnaire, organize the sample design, collect information, write up, check, correct and class the results, calculate the estimates and confidence intervals, and translate the results into non-specialist's language for the client.

More than 100 different surveys have already been carried out by students on this program over the last 20 years. Furthermore, every six years, during the regional parliamentary elections, the entire group (25 students) carries out an estimate of the results for the public local television station, on the basis of the first ballot papers counted in a sample of 300 polling stations; our results are broadcast live on television 30 minutes after the close of polling. The "Rhone-Alpes" region boasts ten million inhabitants. The ballot is carried out according to the list system, in eight constituencies; there are roughly 5000 polling stations in all.

This process lasts an entire semester and requires the students to put all their statistical and data-processing knowledge into action. This involves data collection and data analysis of the results of the last election in all polling stations, in order to form homogeneous strata; random selection of stations in these strata, with probabilities proportional to the number of registered voters, with a program to calculate estimated confidence intervals for the results of each list in each constituency (in 1998 there was a total of 72 lists of candidates); the drawing-up of tables and graphs of the results to be broadcast live on television. The estimates are updated several times in the course of the evening, as the counting in our sample of polling stations progresses. Roughly 300 first and second year students are called upon to observe the counting process in the sample polling stations, and phone in the results. Thirty PCs connected on an intranet are used to record, calculate and draw up tables and graphs of the estimates.

In terms of teaching statistics, this technique has a very rare advantage: a real complex survey can be carried out and the estimated results can be quickly compared to the 'true population values', as the official results are published between six to eight hours after the close of polling.

 

Manuscript
Download in Adobe Acrobat format (99 Kb).

 

Back to top

ICOTS-6, The Sixth International Conference on Teaching Statistics - International Program Committee (IPC) Website.
Copyright © 2001 by the IASE. All rights reserved. This information is subject to change without notice. This page was last modified on July 2, 2002.
For questions or comments, contact the Webmaster, Dagan Ben-Zvi.