Should you rely on statistics as they are presented if they come from a reliable source? Discuss.
Pages: 1
- Index »
- Community »
- Debate and Serious Talk »
- Did you know 95.7% of statistics are made up on the spot?
Poll
Are statistics generally reliable?
| Of course, numb3rs don't lie! | 44% | 44% - 21 | ||||
| Numbers don't lie, statisticians can-think for yourself | 55% | 55% - 26 | ||||
| Total: 47 | ||||||
you mean like the statistic that 95.7% of statistics are made up on the spot?
Thats what he was trying to prove, he is saying that and most people would of belived him except for you and now me. I think we should delete these 2 posts.THA wrote:
you mean like the statistic that 95.7% of statistics are made up on the spot?
Indeed, yet it fails in that I wouldn't consider him a reliable source. Now, if he could prove a reputable news agency makes up statistics..............
The trick with statistics is to look not just at what is presented, but what isn't presented.
The trick with statistics is to look not just at what is presented, but what isn't presented.
Last edited by Bubbalo (2006-07-05 15:31:44)
well statistically this thread will be closed soon
and i didn't make that up.
and i didn't make that up.
That's a 32.54265567356566465435431312583229832538294324% rise since the last time i heard that stat.THA wrote:
you mean like the statistic that 95.7% of statistics are made up on the spot?
One man's reliable source is another man's propaganda spitter.
Unfortunately for statistics, they operate mostly to influence the weak-minded to go with (what they present to be) the flow.
Unfortunately for statistics, they operate mostly to influence the weak-minded to go with (what they present to be) the flow.
Last edited by unnamednewbie13 (2006-07-05 16:55:32)
Statistics are easy to lie with, my maths teacher has fun doing it all the time. He is a prick.
Whoa... Can't believe these forums are still kicking.
Hmm, last time I heard it was that 84.1% of statistics are made up, odd.
As with any other information, I assess a statistic's reliability on where it came from. Reputable, verifiable sources usually have the nod from me.
The paradox is only a conflict between reality and your feeling what reality ought to be.
~ Richard Feynman
~ Richard Feynman
Statistics are easy to lie with, but they themselves do not lie.Flecco wrote:
Statistics are easy to lie with, my maths teacher has fun doing it all the time. He is a prick.
Statistics are not inherently reliable. On the other hand, they are not inherently unreliable either.
If those sentences seem self-evident -- and not contradictory -- you are likely at lower than average risk for susceptibility to statistical manipulation. And therefore in the minority.
It's always a good thing to know more about the methodology employed. Flaws therein may not be deliberate, but will affect outcomes regardless. If the methodology is not available, no claims can be made as to reliability.
This is a nice little article on uncovering flaws in statistics.
If those sentences seem self-evident -- and not contradictory -- you are likely at lower than average risk for susceptibility to statistical manipulation. And therefore in the minority.
It's always a good thing to know more about the methodology employed. Flaws therein may not be deliberate, but will affect outcomes regardless. If the methodology is not available, no claims can be made as to reliability.
This is a nice little article on uncovering flaws in statistics.
that's the point ...THA wrote:
you mean like the statistic that 95.7% of statistics are made up on the spot?
I think you have to look at the methodology of the study. I mean, first off, there are three accepted yet totally different ways to compute an average. Second, you have to look at the size of the population and how it was comprised. That alone can totally determine your outcome. I have yet to see a statistical ethics course, yet it seems to me to be the most fundamental element of it ... yet never addressed.Bubbalo wrote:
Indeed, yet it fails in that I wouldn't consider him a reliable source. Now, if he could prove a reputable news agency makes up statistics..............
The trick with statistics is to look not just at what is presented, but what isn't presented.
Last edited by puckmercury (2006-07-05 20:47:10)
Everyone should remember that when someone presents a statistic, it is usually to support a point, and in that case it has been selected from all available data. The data is real, the statistic is real, they are not lies, and even if you assume that it has been collected following all accepted procedures of sampling etc it can still be misleading.
As an example of what i am saying i would like to say i am the best BF2 player in my group, as my score per min is higher than all the others. If you go by any other BF2 statistic i am not the best player though. And this raw statistic dose not take into account how i got my points.
A maths teacher friend of mine has a particular thing about peoples incorrect use of the word 'statistic', i will see if i can find out from him what that is again and post it up.
As an example of what i am saying i would like to say i am the best BF2 player in my group, as my score per min is higher than all the others. If you go by any other BF2 statistic i am not the best player though. And this raw statistic dose not take into account how i got my points.
A maths teacher friend of mine has a particular thing about peoples incorrect use of the word 'statistic', i will see if i can find out from him what that is again and post it up.
Stats are generally more reliable than polls, as a pollster will usually stop as soon as they get the minimum amount of info that supports a predetermined result, as opposed to using a scientific method in which you generally try to disprove a theory. That is not to say that stats cannot be misrepresented, for instance John Kerry like to throw out the fact that "40 million people lost their insurance in the US last year" during his presidential run. Which was true, but what he didn't say is he was not removing people from the list once they got new jobs and got insurance again, you know when you change companies and there is that 90 day probationary period before you are allowed to receive benefits from the new company. Kinda like subtracting all the withdrawals from your bank account without adding the deposits.
Pages: 1
- Index »
- Community »
- Debate and Serious Talk »
- Did you know 95.7% of statistics are made up on the spot?