Poll: Is 9/10 a good enough statistic?

[sheic99]

No. 9/10 means nothing without knowing the sample size. I remember a study a couple of years ago, where a "researcher" claimed that a homosexual couple was about 3 times more likely to be an abusive relation ship. Seems like a fairly bad thing right? The "researcher" polled 9 people in total.

I never liked statistics. Most often, they're used to make a dubious point look legitimate.

It can be very much so with a small sample size and a biased question. If done correctly, it can be quite beneficial.

 

[FalloutJack]

No. 9/10 means nothing without knowing the sample size. I remember a study a couple of years ago, where a "researcher" claimed that a homosexual couple was about 3 times more likely to be an abusive relation ship. Seems like a fairly bad thing right? The "researcher" polled 9 people in total.

I never liked statistics. Most often, they're used to make a dubious point look legitimate.

It can be very much so with a small sample size and a biased question. If done correctly, it can be quite beneficial.

Well yes, but come on... What are the things you hear 'proven' by analysis the most often? Heh, the bad ones.

 

[SilentCom]

If it is opinion based then I would have to argue that the 9/10 people prefer something over another but something being better is a matter of opinion. It also means that out of the 90% of people who like it better, there is a good chance that some may share similar preferences in games as you. This could mean that you have a higher liklihood of liking the 90% preferred game over the other.

 

[Zantos]

To say 9 out of 10 people prefer something is pretty good if it's a properly conducted survey. Still doesn't mean you can't have your own opinion of it. If you took a survey here you'd probably see some 90% of people that think, for instance, Portal is a fantastic game. Doesn't mean the other 10% aren't allowed to think it wasn't that great.