james sweeney joins us, always 14 things to speak of, as a chief economist, how do you formulate 1.1 ion jobs on friday, how do you derive that statistic? james: it is difficult, but it is causing us to put more attention on the nonseasonally adjusted numbers. some of these numbers have been a little more stable than the seasonally adjusted numbers. i don't know how the factors in, that include 2008 are expected to work, so we think you lost a little bit of the jobs numbers from some of these a few months ago and you will get them back this month. tom: we observed a standard error of the april report on may 7. do we have the same standard exercise this time around? james: i think it was bigger a few months ago, i think the data fog is starting to break, but it is not broken. in the initial return to work phase, it was hard to forecast these numbers. it is not quite that hard. the most important variable is not payrolls with that big standard deviation, but what is the underlying growth that we will see over the next 12-18 months. given that in the individual month is hard to forecast,