As we watch the confirmation do NOT be misled by percentage sampling per se. I’m not sure if it is ignorance of those stating such or just playing to the general population ignorance? Anyone that wants to review O.C. Curves for attributes can determine that probabilities of NOT detecting a defect (important word perhaps or sentence) based upon the hypergeometric, binomial or Poisson with the understanding or definition of “samples” and/or “population” if hypergeometric. Percentage sampling typically under serves with small samples and is overkill with large samples. Ignore the percent reference and find your battle ground if desired on a population of data from which sampling would take place with the assumption that the population of sought information is not the same as the population sampled. In other words…ignore percent being constantly referenced…there is no battle ground there. The battle is not percent, but if the missing documents truly provide pertinent data of which I understand would never be admissible by Bush. This thought is why demographic representative sampling is sought for polls and such. The interest should be if there is just cause to release this relative to the nomination and if it possible that his position at this time could ever reveal anything of importance…THAT is the battle, not the “percent sampling”…
Not trying to get into a debate about this confirmation, but just providing insight relative to percentage sampling and where a battle could really be…not wanting to get into any more than this. Leaving it strictly on percentage sampling that is referenced (7%?). I’m still tired of the space race…
To those that understand attribute sampling it is not meant as an insult and to those that may think percentage sampling if a "homogeneous" population (and that is the key) is needed..that is not the problem people should gravitate towards. Being a bit gun shy I'm reluctant to offer additional understanding "IF" that should occur...but what the hell?
Not trying to get into a debate about this confirmation, but just providing insight relative to percentage sampling and where a battle could really be…not wanting to get into any more than this. Leaving it strictly on percentage sampling that is referenced (7%?). I’m still tired of the space race…
To those that understand attribute sampling it is not meant as an insult and to those that may think percentage sampling if a "homogeneous" population (and that is the key) is needed..that is not the problem people should gravitate towards. Being a bit gun shy I'm reluctant to offer additional understanding "IF" that should occur...but what the hell? 
