Professor Voltolini, It isn't necessarily true that a before and after experiment is always a "good" idea. Carry-over and period effects can be a real problem.
You seem to be describing geographical clusters. Lots of work in the last few years has been on looking at clusters of correlated observations. It's most simple form is the paired design with only 2 observations per cluster. Warren May University of Mississippi Medical Center [EMAIL PROTECTED] (Voltolini) wrote in message news:<00b201c1e051$ced62cc0$a54bfea9@p0h1e7>... > Dear friends, I understand that doing a before and after experiment is a go > od situation to use a paired design because you can measure the same indivi > duals but.... I saw some examples considering the measuremente of the same > area as a paired situation but, the individuals are not controled, so I can > measure different individuals. So, the pair is not a couple of individuals > but the two areas. Is it correct ? If you are measuring the individuals in > side the two areas, it seems to me that I need to pair the individuals and > not the areas ! I also read another example pairing the data because the tw > o samples have genetic relatedness ! > > So... what is the correct logic of pairing ? > > > Thanks...... > > > V. > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Prof. J. C. VOLTOLINI > Grupo de Estudos em Ecologia de Mamiferos (ECOMAM) > Universidade de Taubate, Departamento de Biologia > Praca Marcelino Monteiro 63, Bom Conselho. > Taubate, SP. CEP 12030-010. BRASIL. > Tel: 0XX12 - 2254165 (Lab. Zool.) ou 2254277 (Depto. Biol.) > E-Mail: [EMAIL PROTECTED] > http://www.mundobio.rg3.net/ > http://www.sobresites.com/ecologia/institui.htm > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > > -- . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
