A Bout The System

[kallieb]

In 2004, a study by Statistics Canada found that 59.1% of Canadians aged 25 to 54 had a post-secondary certificate or university degree, 8.0% had taken some post-secondary education, 20.1% were high school graduates, and 12.9% had less than a high school education.

The most impressive thing about those statistics is that they add up to 100.1%

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]

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

Ergo attaining a total of 100% exactly

[dinsdale]

Didn't statistics start the second world war? Or was that Hitler?

[kallieb]

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

Ergo attaining a total of 100% exactly

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

[thewizeard]

The most impressive thing about those statistics is that they add up to 100.1%

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.

Now is that wise?..

[Defy]

Routine is boring. I always do my best to keep things fresh.

[sear]

1 Norway
2 Iceland
3 Australia
4 Ireland
5 Sweden
6 Canada
7 Japan
8 United States
9 Switzerland
10 Netherlands
11 Finland
12 Luxembourg
13 Belgium
14 Austria
15 Denmark
What do they say about myth vs reality

OK, I'm off my soap box now!!

Australia pwns Canada...

[DooMeD68]

try to be spontaneous - otherwise it can become monotonous !!

[Mr JP Fugley]

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%.

[kallieb]

1 Norway
2 Iceland
3 Australia
4 Ireland
5 Sweden
6 Canada
7 Japan
8 United States
9 Switzerland
10 Netherlands
11 Finland
12 Luxembourg
13 Belgium
14 Austria
15 Denmark
What do they say about myth vs reality

OK, I'm off my soap box now!!

Australia pwns Canada...

see!!! never make assumptions. Who would have though good ol' down-under-land would be in the top 3. With all that, you need a better spokesman than Paul Hogan