Hi all-staters!
Does anybody have some suggestions and literature references about the
following problem.
In an human behavior study we investigated the entire population of a
prison in Northeast Brasil, to we need calculate confidence intervals
for the parameter estimators also because we investigated the entire
population of the prison.
I think it depends what is defined as the target population:
1) We have to calculate CIs if the target population is an
(hyper)super-population :
like people in all prisons of the world, in all prisons of Brasil, etc.
2) No CIs: if the target population = study population = this prison????
Thanks Bernd
------------------------------------------------------
Dr Bernd Genser
MSc, PhD
Estatístico
Instituto de Saúde Coletiva
Universidade Federal da Bahia
Rua Padre Feijó, 29 - Canela
Salvador (BA) - Brazil - 40.110-170
tel.: ++55 (71) 245-0544/246
cel.: ++55 (71) 88137239
email: [log in to unmask]
------------------------------------------------------
|