🚀 go-pugleaf

Dennis Roberts said:
:
: My question is ...  is there such a creature?  And, if there is, what does
: it do?  Yes, this may show my ignorance but ...  it sure would not be the

The Welch Aspin test (or t-test) is  a variation of the student's t-test which
assumes the variances are not equal:

                          X1 - X2 -(u1 - u2)
          t =             __________________

                          _    2        2 _  1/2
                         |    S1       S2  |
                         |    ___  +   __  |
                         |    n1       n2  |
                         |_               _|

 This is what is usually referred to as the "Separate Variance t-test"
 as is done in SPSS.  The degrees of freedom are not the usual ones

                       2       2    2
           df=       (S1/n1 + S2/n2)
               ___________________________________________
                 2    2                 2     2
              [(S1/n1) /(N1 - 1)] +  [(S2 /n2) /(n2 - 1)]

: FIRST time.  Any help/reference would be very much appreciated.  You may
: want to send a PREPLY so as not to clog up the network with answers to my
: specific question ... what is the Welsh/Aspin test and what does it do?
:

All the other assumptions of the t-test are still in effect, only the
heteroskedasticity assumption is relaxed.  See Scnedecor and Cochran or Steele
and Torrie for treatment of this.


--


Stephen P. Baker                                phone:  (508) 856-2625
Lecturer in Biostatistics                               (508) 856-3131 fax
Department of Academic Computing                        (413) 253-3923 home
University of Massachusetts Medical School    e-mail: sbaker@umassmed.ummed.edu
55 Lake Avenue North                                  -.- -.. .---- .--. ..-.
Worcester, MA 01655