Language Log on Odds Ratios
There was an interesting post on the presentation of odds ratios on the Language Log Blog, the other day. They give some examples of odds ratios being deceptive, confusing and misunderstood. It's been said plenty of times, by plenty of people (including me) but it's interesting that linguists are saying it too.
posted by J at 4:33 AM
2 Comments:
Isn't the problem that journalists generally don't understand either risk ratios or odds ratios? I prefer ORs to RRs for most purposes, but that's because the stuff I'm interested in tends to not to have stable base rates. On the other hand most medics prefer RRs because they incorporate base rate information (one hopes) from relatively stable base rates. As a rule of thumb RRs are more useful for a specific practical application and ORs for science.
That said ... any journalist looking at medical research ought to understand the difference between RR and OR - especially for non-rare events.
A similar issue arises with r (base rate sensitive) and d (base rate insensitive.
Yes, I guess you're right. Some medical journals are now insisting on RRs, I suppose because they are read by doctors.
I reviewed a paper the other day (for a health psychology journal where the authors thought that ORs were RRs. Which was a little depressing.
Post a Comment
<< Home