This is a session of Topic 1: Data and context in statistics education: towards an evidence-based society
(Tuesday 13th, 11:00-12:30)
Lies, damn lies, statistics: lessons from past and present for the future
Organizer
- Irena Ograjenšek (Slovenia)
Abstract
Back in 1954 in his introduction to the now classic book entitled “How to Lie with Statistics” Darrell Huff pointed out that in a fact-minded culture (“an evidence-based society” in our terminology) in which “statistical methods and statistical terms are necessary in reporting the mass data of social and economic trends, business conditions, ‘opinion’ polls, the census”, the secret language of statistics, while very appealing, is primarily used “to sensationalize, inflate, confuse, and oversimplify”.More than six decades after the book was published for the first time, we are still forced to acknowledge the fact that a “well-wrapped statistics … misleads, yet it cannot be pinned on you”. More than ever we’re hoping to educate new generations of researchers, politicians, journalists … who will present statistics to the general public with honesty and understanding. We also haven’t given up hope to raise a general public who will know what these statistics and accompanying words mean.
In the framework of this session we explore an abundant number of lessons in “statisticulation” from the past and the present hoping that in the future the resulting semantic and visual nonsense can be at least partially, if not entirely, avoided.
Papers
| Paper | Title | Presenter(s) / Author(s) |
| 1G1 | The “compleat” applied statistician | Donald L Bentley (United States) |
| 1G2 | Unintentional lies in the media: don’t blame journalists for what we don’t teach | Jessica Utts (United States) |
| 1G3 | Election poll “lies” | Zenel Batagelj (Slovenia) |