 You don't need to be an 'investor' to invest in Singletrack: 6 days left: 95% of target - Find out more
How do you interpret odd ratios? Bigger number = higher chances of there being a relationship between the two things?
Say you had an odds ratio of 72.08 with 95%CI of 8.88-584.77 with a p-value of <0.001.
What's that in plain english?
Odds ratio is the chance of something happening in the study group vs. control, ie: odds ratio of 2 means the event is twice as likely in the study group as control, and 0.5 is the reverse. Doesn't reflect the chance of there being a relationship.
[i]Edit - what I've described appears to be the Relative Risk. OR is the ratio of an event happening to an event not happening for a given population[/i]
p value reflects the chance that the statistical difference between study and control groups is due to chance. By convention p<0.05 is regarded as good enough.
That 95% CI is enormous.
What's the context behind this?
Andy
Risk of symptomatic knee osteoarthritis as conferred by the presence of coexisting depressive disorders.
Odds ratio - what he said -odds of it in one goup compared to another group
CI - No idea
P value is effectively the odds of it being due to chance with the lower the number the less likely it is chance and 0.05 [ or 0.01being the point of "proof"
So what they're saying is that if you have depression you're 78 times more likely to have symptomatic osteoarthritis of the knee than if you dont have depression? And that they're 95% confident that the odds are somewhere between 8.88 and 584.77, and they're virtually certain it's not down to chance.
That cant be correct can it?
This page explains it well
[url] http://stats.org/stories/2008/odds_ratios_april4_2008.html [/url]
Essentially an odds ratio is the probability of something happening vs the probability of it not happening
Relative risk is the risk of one thing happening vs another thing
Either way whatever study you're reading sounds wrong! CI that wide either mean tiny sample size (unlikely with that p value), or a badly designed study
What practical application is the answer going to have for you?
Step away from the keyboard. Maybe go out for a beer.
Actually having re-read that link I think I'm wrong. I described Odds, not Odds Ratio