Date:         Sat, 7 Sep 1996 17:42:40 +0000
Reply-To:     pg <cristian@TAMBAQUI.ESALQ.USP.BR>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         pg <cristian@TAMBAQUI.ESALQ.USP.BR>
Subject:      Dependent samples
Comments: To: david izraylovich <aldair@WORLD.STD.COM>

Hi, all SAS - L's!
 
I am working with an incomplete block design. When I examined
the standardized residuals, I have found a strong dependency
between pairs of observations. Perhaps the researcher who
performed the experiment don't washed the equipment correctly
and therefore this dependency appears.
 
There is an  "statistic" in mind who says all day to me:
 It is all garbage!!!! Put all this pages, diskettes, etc. in the
paper-basket and have a good beer!
 
But there is a "Hare Krishna" in my mind too, remembering
me that the researcher spended two years dealing with
all this stuff and surely she will be in very problems if I decide to
choice the good-beer alternative...
 
Does anybody know a method to "capture" this dependency
and deal with it? References are well accepted.
Sugestions for the choice too.
Cristian A. Carranza            | Fortuna plango vulnera, stillantibus ocellis
Cristian@tambaqui.esalq.usp.br  | quod sua michi munera subtrahit rebellis
                                | Verum est, quod legitur, fronte capillata,
                                | sed plerumque sequitur, ocassio calvata.