•  
  •  
 

Abstract

This paper is about errors in statistical data. My remarks are in two parts. The first part deals with the proposition that the statistical data we consume, analyse and produce contain more error from more sources than we sometimes recognise. The second asks: How can we better deal with these errors?

The topic of error in statistics can become highly technical, but for the most part this is not a technical presentation. It deals with common sense and experience; much of it is anecdotal.