How statistical process control charts mislead

The statistical process control chart is designed to help differentiate between what are usually called “common” causes of variation and “special” causes. But this is very misleading. Typically, six-sigma consultants and various authors advise you to ignore data between control limits because it is “random.” They say to only make process adjustments when the process [...]

Questionable definitions cause indicators to mislead

The previous post about counting  flu deaths points out a typical reason an indicator is misleading: definitions. What is a flu death?  Is it a death from which the flu is positively diagnosed through a laboratory analysis of the virus found in the victim’s body? Or is it a death from some complication or secondary [...]

Who is misleading on flu death count?

The US Centers for Disease Control and Prevention (CDC) reports that:  during the 1990s, flu-related deaths ranged from an estimated 17,000 during the mildest season to 52,000 during the most severe season (36,000 average). Studies going back to 1976 have found that flu-related deaths ranged from a low of 4,700 to a high of 56,600 [...]

Why software development productivity cannot be measured

The cost of developing software can be significant for any business. So it would make sense to try and measure the productivity of software development efforts. Trouble is, there is no way to even define productivity in this case. There is no definition of the output of software development: is it a line of code? [...]