A Bout The System

I

ivethelloween

New member
Barbarossa said:
The most impressive thing about those statistics is that they add up to 100.1% :O
Click to expand...

It is because they have to round it off. The actual statistic is likely a long number. For example, take 20.1 percent. In fact it is likely 20.0934859, so they round up to the highest number ergo 20.1. When you do this multiple times you end up with a tad more than 100.0% as a pure figure, thereby giving you 100.1%.

In a nutshell, it comes with the territory when one is doing statistical analysis.
 
Barbarossa said:
:happy: :coffee:

I probably would have rounded them to whole percents then, i.e. 59%, 8%, 20% and 13%.

Ergo attaining a total of 100% exactly :dabs:
Click to expand...

The problem with that approach is that the gap between 20 and 21 percent encompasses far more people -- when you are doing demographic-type studies, thus, you might be over or under reporting your study results by literally 1000's (or 100,000's) depending upon your population base. For example, a country like China - which has a population of 1 billion (give or take), if you fudge 1% you are talking about 10,000,000 people.

Thus, you conclusions may end up being somewhat challenged on principles of reliability and validity because you are not narrowing your results well enough. I'd just as soon be content to see 20.5 -- as this is more near the acutal number, than underreporting it as 20%, or over-reporting it as 21%.

What you suggest is compelling, but when you weigh out the nitty-gritties of statistics, close to precision is always the aim to shoot for.

Interesting discussion BTW
 
kallieb said:
:huh:

It is because they have to round it off. The actual statistic is likely a long number. For example, take 20.1 percent. In fact it is likely 20.0934859, so they round up to the highest number ergo 20.1. When you do this multiple times you end up with a tad more than 100.0% as a pure figure, thereby giving you 100.1%.

In a nutshell, it comes with the territory when one is doing statistical analysis.
Click to expand...

Now is that wise?..
 
If there are 4 options and you only have to deal with 0.1% then your adjustment is only 0.025% on average for each of them. Whilst that may represent a reasonably large number of people, depending on the base, it is statistically insignificant, which is more important. Certainly better than reporting 100.1%.

The proper way to do it would be to both round up and round down, resulting in a figure of 100%.
 
Back
Top