Hi, I read on a paper the below statement, don't know how that was calculated. Basically, there are 2 continuous variables x1 and x2, as independent variable for predicting cancer recurrence. So this is a survival analysis. Now the author try to check the correlation between x1 and x2. He calculated Spearman rand correlation (~0.22), then he had the following statement: "Only approximately 5% of the variability in the estimates of recurrence using either of these scores was explained by the other". How was that done? with a Cox model including both x1 and x2, can we get that number (5%)? Thanks for any hints John
a question
3 messages · Duncan Murdoch, array chip
On 05/08/2011 2:19 PM, array chip wrote:
Hi, I read on a paper the below statement, don't know how that was calculated. Basically, there are 2 continuous variables x1 and x2, as independent variable for predicting cancer recurrence. So this is a survival analysis. Now the author try to check the correlation between x1 and x2. He calculated Spearman rand correlation (~0.22), then he had the following statement: "Only approximately 5% of the variability in the estimates of recurrence using either of these scores was explained by the other". How was that done? with a Cox model including both x1 and x2, can we get that number (5%)?
I would guess it is r^2 = .22^2 = 0.0484, and I suspect the claim is unfounded. Duncan Murdoch
Nice guess Duncan! Otherwise, I can't think of other ways to get that number. Thanks John ----- Original Message ----- From: Duncan Murdoch <murdoch.duncan at gmail.com> To: array chip <arrayprofile at yahoo.com> Cc: r-help <r-help at r-project.org> Sent: Friday, August 5, 2011 11:21 AM Subject: Re: [R] a question
On 05/08/2011 2:19 PM, array chip wrote:
Hi, I read on a paper the below statement, don't know how that was calculated. Basically, there are 2 continuous variables x1 and x2, as independent variable for predicting cancer recurrence. So this is a survival analysis. Now the author try to check the correlation between x1 and x2. He calculated Spearman rand correlation (~0.22), then he had the following statement: "Only approximately 5% of the variability in the estimates of recurrence using either of these scores was explained by the other". How was that done? with a Cox model including both x1 and x2, can we get that number (5%)?
I would guess it is r^2 = .22^2 = 0.0484, and I suspect the claim is unfounded. Duncan Murdoch