Shrink's Views

ramblings of an unknown psychiatrist

“Lies, Damned Lies and Statistics”

Posted by Dheeraj Kattula on May 22, 2009

Warning:

If you cannot stand a bit of statistics/maths or dry humor please stop here. If you do not mind that then, please read the disclaimer.

Disclaimer:

The following research is purely fiction. The quotes are original. The author of the post believes there is some ‘truth’ in these quotes.

———————————————————————————————-

Benjamin Disraeli, the British politician said There are three kinds of lies: DisraeliLies, Damned Lies and Statistics.

It is interesting to know how one comes to such a conclusion. First collect data. Without data you cannot conclude any thing. Here you collect all statements which are ‘lies’. Lies- being defined as that which does not correspond to truth. Truth- being defined as that which corresponds to reality. ‘Lies’ are categories. Also define a continuous variable called ‘distance from truth’. So each lie is distant from truth by some degree.

A huge sample of ‘lies’ is collected. And the distance from truth is noted for each lie. The frequency distribution at each distance is seen.

Hey! You do not get a normal curve (with a single peak at the mean) in this as you get for most distributions. This is a trimodal curve. There are three peaks!!!

On analysis it was found that the ones with low distance from truth were uttered mostly by common people in the sample. These are Disraeli’s ‘lies’. The ones whose distance from truth was farther were mainly politicians. Their utterances were the ‘Damned Lies’.

The ones who were farthest from the truth were researchers, scientists and officials whose lies were officially ‘statistics’.

This is not to say that the scientific community has always got it wrong. It is just to remind ourselves that we have to be humble in science. We could be the ones farthest from truth if we ‘seek’ to find our pet conclusions. Let us remember the Nobel Laureate Ronald Coase’s adage “ If we torture data long enough, it will confess.

Advertisements

2 Responses to ““Lies, Damned Lies and Statistics””

  1. Never heard that quote by Coase (“ If we torture data long enough, it will confess.”)… It is both hilarious and true.. Great post and very informative!! 🙂

    PS: Could you rearrange the widgets in your blog? It seems loaded to the left 🙂 (Just a suggestion)

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

 
%d bloggers like this: