[Scipy-tickets] [SciPy] #1289: Wrong / contradictory documentation of stats.mannwhitneyu

SciPy Trac scipy-tickets@scipy....
Mon Sep 27 09:32:56 CDT 2010


#1289: Wrong / contradictory documentation of stats.mannwhitneyu
-----------------------+----------------------------------------------------
 Reporter:  Wrzlprmft  |       Owner:  somebody
     Type:  defect     |      Status:  new     
 Priority:  low        |   Milestone:  0.9.0   
Component:  Other      |     Version:  0.6.0   
 Keywords:             |  
-----------------------+----------------------------------------------------
 {{{
     help(mannwhitneyu)
 }}}
 states:
 “Use only when the n in each condition is < 20”
 while
 [http://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.mannwhitneyu.html
 the online documentation] says:
 “Use only when the number of observation in each sample is > 20”

 If I am not mistaken, the second restriction (“<20”) is closer to the
 truth. However it applies only to the second return value (prob). The
 first return value (u) can be correctly and easily calculated for all
 inputs and SciPy appears to do so. For small inputs (<20) it can only be
 evaluated using a table of critical values (or calculating those critical
 values), while for larger one can use an approximative normal distribution
 to calculate the significance level (prob).

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1289>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list