Directional discrepancies make 2010 census an April fool

The U.S. Census Bureau went to great lengths to tell us to stand up and be counted on April 1, 2010. So why did they sent out postcards dated March 22 encouraging us to disregard earlier instructions and return the census survey early?

The U.S. Department of Commerce went to great lengths to spread the word about the importance of filling out the Census precisely on April 1, 2010. They even tell us on the form to wait to fill it out until the first day of April.

I’m sure I’m not the only one who felt guilty about populating the form in advance and setting it aside until the end of the month, along with the bills due at that time. It’s not like I can jinx myself, right?

The Census Bureau now, however, has committed the data-gathering sin that would send a true statistician into a frenzy: They told us to turn in the forms early!

Yes, you heard correctly. The Census Bureau sent out reminder postcards dated March 22, urging people to turn in their census forms before the end of the month.

The automated telephone help line assures us in a calm computer voice that if we’re sure we’ll be living at our current address on April 1, we can turn in the form early.

The original form mailed to folks clearly states that the census should reflect our status as of April 1, 2010. Not before, not after. It is a snapshot, a statistically valid sample of one day in the life of the United States.

If Americans fill in the form and mail it early, the data become statistically less valid. For one thing, a significant number of people will die – most unpredictably – between the last week of March and the beginning of April.

If the Census Bureau now has the power to suspend mortality for a week, or for even a day, I applaud them heartily. Maybe we should do this every year if the government demands that death take a holiday (even if it is confidentially).

However, during the last week of March a few among us had premature babies, while some unexpectedly ended up living somewhere else, perhaps through eviction, sentencing, sickness or dissolution of a domestic partnership.

I suspect that most of us mere mortals could neither predict the date of our natural demise any more than we could guarantee that our address and family size would remain unchanged for one magic week. So perhaps the census really is what it has appeared to be all along: a massive April fool’s joke.