log(2, no_of_jobs + 1) will give 0 for 0 jobs, 1 for 1 job, 1.58 for 2 etc. etc.
On 11 Feb 2011, at 14:04, Johan De Meersman wrote:
> How about the square root of the number of jobs, or some other root if you
> want another coefficient? That doesn't have the limiting behaviour a
> logarithmic fu
How about the square root of the number of jobs, or some other root if you
want another coefficient? That doesn't have the limiting behaviour a
logarithmic function offers, though.
On Fri, Feb 11, 2011 at 2:08 PM, Richard Reina wrote:
> Hi Travis,
>
> This is very helpful thank you. Howev
Hi Travis,
This is very helpful thank you. However, is there a way to make it not be
less than a 1. As it's written below someone with one job gets a zero and
someone with no jobs gets a NULL. It would be great if someone with 1 job
got a 1 and someone with zero jobs got a 0.
Thanks again,
R
Maybe some sort of logarithmic expression?
select no_of_jobs, 10 * log(10, no_of_jobs) as job_weight
from data;
Of course, you'd have to tweak your coefficients to match the weighting
system you want to use.
-Travis
-Original Message-
From: Richard Reina [mailto:gatorre...@gmail.com]
S
