What happens when the statisticians get it wrong?

Statistics can be good tools and good sources for planning/decision-making, identifying trends, and applying for grants for municipalities, if they are derived from studies and surveys that are conducted with accuracy and balance. They are also helpful for businesses in learning about a community’s labour market, housing, and industry information.
When a reputable source like Statistics Canada gets the data wrong, though, the results can have fairly negative impacts for local business and municipalities. If, for instance, an important demographic is left out of the study, then those statistics can only be taken with a grain of salt, and leave wide open the argument of validity.
Such is the case with the recent Statistics Canada results which came out earlier this month and indicated a declining population for Bow Island and showed a higher vacancy rate than what locals know to be true. In fact, the census showed Bow Island’s population to be down by more than two percent from 2011 figures. At the same time, both the County of Forty Mile and the Village of Foremost showed increases in population numbers.
While the numbers may be accurate for those who responded to the census, a demographic in the community did not participate and that in itself threw off the numbers.
Polls, especially those being conducted by organizations who have political leanings should also not be taken too seriously. A prime example is during the recent U.S. election when a major news network was announcing poll results from surveys conducted by the Democratic National Convention (they rarely cited where they obtained their poll results). They were hoping the viewing public would assume the poll results came from an unbiased source when the exact opposite was true.
Of course, Statistics Canada did not set out to provide skewed census results for Bow Island, but the public assumes that their results are an accurate representation of the communities they studied.
Now, the Town of Bow Island will be conducting its own census, and has the legal right to do so. The Town will be hiring somebody to go out into the community and ensure every segment of the community is included. They are doing the right thing, of course. If it means more money for per capita grant funding, then what choice do they have?
The question also arises about how many other communities are facing similar outcomes from the recent federal census? What other information is incorrect because not enough data was collected?
By now, every municipality has likely pored over the census results and have been able to determine whether the Stats Canada information is correct for their municipality or not. If not, they are likely going to be footing the bill to hire somebody to do their own census.