The payroll survey samples more than 600,000 worksites; the household survey, some 60,000 households.

在职人员调查抽样了60多万个工作场所，家庭调查抽样了大约6万个家庭。

So what you do is you survey a random sample of that population and then you can calculate statistics on that sample that hopefully can estimate the population as a whole.

所以你要做的是调查该人口的随机样本，然后你可以计算该样本的统计数据，希望可以估计整个人口。