Pitt, Hy (1991, ASQC) Pitt Training Associates, Milwaukee, WI
This abstract is an edited version of the author's original.
This paper provides examples of often-seen and often-heard statistical statements and conclusions using language that tends to distort the results suggested by the data, ascribes scope and abilities not inherent in the methods used, misleads the reader, and causes unwarranted actions to be taken. These confusing and incorrect statistical conclusions are unfortunate because statistical methods are experiencing an unprecedented revival and implementation in our attempts to improve quality and productivity.The author lists common statistical statements and explains why they are misleading. The most notorious example is: "The process is out of control." The correct wording is: "The process is out of statistical control." The author also suggests further clarifying the statement by expressing the conclusion in terms of the center line of the control chart, or of some process parameter, such as: "The mean has shifted from the target of 20.5 cm," or "The variability has increased.
"Courses in statistics can be improved if: (1) the distinction between population and sample is more strongly stressed; (2) the use of more precise language is reinforced; (3) language that is more imaginative and stimulating than the usual statistical jargon is encouraged; and (4) students are provided more practice in presenting statistical results.
Chemical and process industries,Problem solving,Process control,Statistics