IUBio

YOUR HELP REQUIRED : Perception of Numbers

Jonathan Foweraker fowerake at math.utah.edu
Thu May 23 18:01:07 EST 1996


peter.murray at s054.aone.net.au (PRM) wrote:
>
>Experience, example, and common sense suggest that carefully engineered visual 
>tools (graphs etc.) can far more effectively convey insight into a numerical 
>organism (i.e. business) that can numbers.  Are pictures, and therefore 
>graphs, an intrinsically more native and effective way to communicate concepts 
>than numbers.
>

It is also true that carefully engineered visual tools can far more effectively
mislead, whereas the actual numbers will give a far more realistic viewpoint.

For example:

Increases can be enlarged/reduced by using log/exponential scales.

By the appropriate choice of axes range effects can be magnified/minimised:
e.g. think of a graph showing a rise (say in concentration) followed by a fall;
A clearly defined 'pulse'? If the rise is from 102 to 107, and then back to
101 then no - although if the y-axis only ranges from 100 to 108 then the graph
might indicate (to the unobservant) that there is a significant rise.

On the other hand there are "Lies, damn lies and statistics".

However in general I agree that visual techniques are more appropriate at
getting the picture across than just tables of numbers. The best example I
can think of (no reference - sorry) is a picture showing Napolean's march
on Mosocow. The picture manages to convey the route, the number of men -and
so the number of casualties and also the date. It probably has more information
on it (I remember some colour coding) - and also shows where different
divisions/battalions split off to achieve other objectives.

Cheers,


Jonathan.




More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net