# Fwd: high p values

6 messages
Open this post in threaded view
|

## Fwd: high p values

 Hi This is my function: wilcox.test(A,B, data = data, paired = FALSE) It gives me high p value, though the median of A column is 6900 and B column is 3500. Why it gives p value high if there is a difference in the median? Regards         [[alternative HTML version deleted]] ______________________________________________ [hidden email] mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide http://www.R-project.org/posting-guide.htmland provide commented, minimal, self-contained, reproducible code.
Open this post in threaded view
|

## Re: Fwd: high p values

 We've had this conversation. A) This is off-topic for R-Help. Your question is about the statistical test, not about the R coding. B) A difference in sample statistics, whether or not it "looks" large, is not sufficient for statistical significance. ﻿On 3/19/19, 12:48 PM, "R-help on behalf of javed khan" <[hidden email] on behalf of [hidden email]> wrote:     Hi         This is my function:         wilcox.test(A,B, data = data, paired = FALSE)         It gives me high p value, though the median of A column is 6900 and B     column is 3500.         Why it gives p value high if there is a difference in the median?         Regards         [[alternative HTML version deleted]]         ______________________________________________     [hidden email] mailing list -- To UNSUBSCRIBE and more, see     https://stat.ethz.ch/mailman/listinfo/r-help    PLEASE do read the posting guide http://www.R-project.org/posting-guide.html    and provide commented, minimal, self-contained, reproducible code.     ______________________________________________ [hidden email] mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide http://www.R-project.org/posting-guide.htmland provide commented, minimal, self-contained, reproducible code.
Open this post in threaded view
|

## Re: Fwd: high p values

 In reply to this post by khan123 > This is my function: > > wilcox.test(A,B, data = data, paired = FALSE) > > It gives me high p value, though the median of A column is 6900 and B > column is 3500. > > Why it gives p value high if there is a difference in the median? Perhaps becuase a) because you are testing the wrong data or b) there isn't a significant difference a) You are probably not using the data you think you are. Check ?wilcox.test; the 'data' argument is specific to the formula method. That needs a formula as the first argument, not a numeric vector. What you've done is apply the default, and 'data' has been ignored. So A and B are whatever was lying around in your current environment, not what is in 'data'.  ('data' is a terrible name for a data frame, by the way, as 'data' is an R function). After that: - How many data points do you have in each group? - How much do the two groups overlap? If the answers are 'not many' or 'lots' (in that order), and especially if both apply, you can't expect a significant test result. S Ellison ******************************************************************* This email and any attachments are confidential. Any use...{{dropped:8}} ______________________________________________ [hidden email] mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide http://www.R-project.org/posting-guide.htmland provide commented, minimal, self-contained, reproducible code.
Open this post in threaded view
|

## Re: Fwd: high p values

 In reply to this post by khan123 Any reasonable test of whether two samples differ should be scale and location invariant.  E.g., if you measure temperature it should not matter if you units are degrees Fahrenheit or micro-Kelvins.  Thus saying the medians are 3500 and 6200 is equivalent to saying they are 100.035 and 100.062: it does not tell use how different the samples are.  You need to consider how much overlap there is. Bill Dunlap TIBCO Software wdunlap tibco.com On Tue, Mar 19, 2019 at 9:48 AM javed khan <[hidden email]> wrote: > Hi > > This is my function: > > wilcox.test(A,B, data = data, paired = FALSE) > > It gives me high p value, though the median of A column is 6900 and B > column is 3500. > > Why it gives p value high if there is a difference in the median? > > Regards > >         [[alternative HTML version deleted]] > > ______________________________________________ > [hidden email] mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help> PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html> and provide commented, minimal, self-contained, reproducible code. >         [[alternative HTML version deleted]] ______________________________________________ [hidden email] mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide http://www.R-project.org/posting-guide.htmland provide commented, minimal, self-contained, reproducible code.