Getting a precise handle on the changing economic and social fabric of small towns is being seriously muddied up by the Census Bureau's new approach to questioning the American population. This is certainly not the result of any desire to sideline the roughly one-quarter of America's population living in small towns and rural areas.

But sidelined we are. We pay the same taxes on a given level of income as people who live in big cities, but that doesn't buy us the same timeliness and quality of the numbers that we need to navigate troubled economic waters.

Our sidelining is the result of serious, well-intentioned thinking at the Census Bureau about how to improve national and state baseline stats about the economy, housing, families, education, population growth and movement, ethnicity, and immigration. This involved moving away from a detailed questionnaire that was sent to one out of every six people when the census was taken every ten years. That was 47 million people for the 2000 Census, making it the Mother of All Surveys.

With such a humungous sample size, the private sector and economic development agencies could reliably slice and dice Census data right down to local big city neighborhoods, or to quite small towns. The disadvantage of this humungous sample was that it was only renewed every ten years, and took well over a year to process that mountain of information into usable form. So companies and government agencies would have to work with a giant snapshot taken in 2000 until at least 2011, or more likely 2012.

To counter this time gap, the thinking at the Census Bureau ran towards creating a continuous survey of a considerably smaller, but still substantial, number of people. The bureau would ask the same detailed questions, and then some, as the previous decennial questionnaire. During every workday of the 52 weeks of the year, a relatively small number of people would be queried. This would still add up to, impressively, roughly three million people. The Bureau could then renew fundamental economic and social information every year, rather than every ten years.

Where the detailed Census questionnaire was mostly administered by a sizable army of temporary employees, the new survey would be conducted by a smaller but permanent army of trained professionals. The resultant increased quality of the collected numbers would partially counter the reliability problems that arise when fewer people are queried.

However, just as all politics is ultimately local, so is economic development strategy. The way that smaller annual sample was defined and organized knocked out much of the previous system's ability to reliably hone a focus on small areas — specific neighborhoods of big cities as well as small towns and rural areas. Big cities, however, have universities and sizable planning departments that can do much to fill the gap. We don't.

The bureau's problematic approach kicked off in January 2005. It uses sampling that creates annually renewed results for places with at least 65,000 residents. For places with 20,000 to 64,999 residents, they must accumulate three years' worth of questionnaire responses before the same detailed information exists. That meant lumping together and averaging the results for 156 weeks between early 2005 and late 2007.

For towns with less than 20,000, it will take five years before an adequate number of respondents are built up. Sometime after Labor Day 2010, the Bureau will publish small town stats that average all the information collected between early 2005 and late 2009. In 2011, the Bureau will average the results for 2006-10, and on it will go.

That running five year average is obviously split right in two by prosperous growth that proved to be a bubble, and the worst times since the Great Depression. If six months is a long time in politics and a market economy, 60 months is a century or two. There are certain kinds of economic problems where multi-year averages are used to "smooth out" data to trace structural trends. For the most part, figuring out small town economic development strategies is not one of them.

The meaning of the Census Bureau's numbers for small towns that will be published in the autumn of next year can be qualified technically as somewhere between nada, zilch, and bupkes.

In all fairness, the people making the decisions at the bureau probably thought that they were offering a considerable compensation to small towns in terms of losing the previous rock-solid benchmark stats. Small places would get their new detailed economic and social profile at least one year, probably two, before they would have under the old once-a-decade system. After that, the info would be annually renewed every year.

That is a flawed decision, with an obvious technical fix: increase the sample size for small and mid-sized places so we can have the same annually renewed stats as larger places. This would certainly cost more, but not as much as the losses that result from small towns having to shoot in the dark.

Most important changes create both winners and losers. Possibly the greatest good for the greatest number of people was invoked to justify changes that disadvantage people who live in small places.

A complementary scenario is that this technocratic reasoning prevailed because small towns aren't well-organized to make enough noise in Washington to get our interests recognized. Imagine the storm that the farm lobby could stir up if parallel moves were made on the Census of Agriculture.

On two recent occasions, representatives from the Census Bureau attempted to take a purportedly moral high ground in requesting the publisher of this newspaper to run free advertisements for this large, well-funded government agency — in a time when all newspapers, especially those serving small towns, are struggling with an economic crisis that shakes local businesses' resources to place ads.

I think this is the same mentality that reworked the Census in a direction which is contrary to the capacity of small towns to think and act our way out of the current crisis. That includes every town that is covered by this newspaper and the Shawangunk Mountain Guide. The ball is in our court to inform the Senators and Congresspeople who represent our towns that we deserve equal access to timely, accurate stats.