Re: ERROR BAR AND ODD RATIO
- From: sanz <zarghu@xxxxxxxxxxxxxx>
- Date: Wed, 5 Nov 2008 21:59:07 -0800 (PST)
On Oct 18, 12:49 pm, Bruce Weaver <bwea...@xxxxxxxxxxxx> wrote:
zar...@xxxxxxxxxxxxxx wrote:
Hello,
I am doing a project and need some help with few concepts . Topic is
"prevalence of Hepatitis B in pregnant women and its risk factors" .I
have calculated the ODD RATIO.. but i need to know what it
tells ... ?? What is an odd ratio ? and the CI that we get in SPSS ...
when we calculate the risk factor ...
You need to do some reading. There are lots of online notes that
explain what an odds ratio is, and how to calculate the OR (and
its 95% CI) for a 2x2 table. E.g.,
http://www.bmj.com/cgi/content/full/320/7247/1468http://slack.ser.man.ac.uk/theory/association_odds.htmlhttp://www.childrensmercy.org/stats/journal/confidence.asp
Second is about ERROR BAR ... what does an error bar show ... I think
Error bars could show any of a number of things. You'll have to
compute the 95% CI (using the methods described at the links given
above) to see if that's what your error bars show.
the two horizontal lines is the Upper and lower confidence interval
and the box inside is the mean ... so if i am having an error bar for
surgery as a risk factor for Hepatitis B ... I have answers YES and
NO ... suppose 20 yes ... 80 No ... what would the lines reprsent .. I
have problems with interpretation of results ... I need some help ....
Why are the gaps unequal between them and what do they tell us ... ??
Please ... help , thanks
--
Bruce Weaver
bwea...@xxxxxxxxxxxxxxxx://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."
Thanks Bruce ... :)
.
- Prev by Date: Re: Quadratic weighted Kappa and the Intraclass Correlation Coefficient
- Next by Date: Re: Quadratic weighted Kappa and the Intraclass Correlation Coefficient
- Previous by thread: Quadratic weighted Kappa and the Intraclass Correlation Coefficient
- Next by thread: Re: First order variation in PDF
- Index(es):
Relevant Pages
|
