Loose maths sinks ships

Slashdot reports on a survey of a gazillion project managers on why their projects fail. The results are interesting, but the authors also present an all-powerful mathematical formula that lets you slice and dice your project into a single number that supposedly indicates risk. So far, so statistical.

They (or the ACM, publishing the article) helpfully present a “worked example” of the formula, which uses subjective ratings in six areas, each multiplied by a weighting, then all added together to give you your distance from doom. Great. But they (or the ACM) get their maths wrong.

In case the graphic is fixed, here’s a transcription of the worked example right now:

Fit between blah blah blah: 5 x 3.0 = 15.2

Level of blah blah: 6 x 1.9 = 11.6

Use of blah: 1 x 1.7 = 1.7

Similarity to blah: 3 x 1.5 = 4.5

Project simblahcity: 7 x 1.1 = 7.4

Stablahlity of blah: 9 x 0.8 = 7.3

Overall blah: 48

2/6. See me. (And their numbers actually add up to 47.7 not 48, but they’re probably “rounded”.)

I make the correct answer 47.5, due to my superpower ability to both multiply and add. But hey, it’s all fuzzy semi-meaningless stats since it starts from a subjective score. Simply bias your biases to reduce your risk!


Leave a comment

Filed under Random

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s