Government spends more money on survey to give us an inferior product

In 2010 I wrote a blog about the The Senseless Census

Now actually I always felt that the government should rethink the census. Although the info we get from the National Household Survey (NHS) is useful, for marketers and business much of the data comes to us too late. There has to be a way in this modern age of technology to speed up the process.  But what I would have expected the government to do when making changes to the census is consult with the users. The data generated by the long-form census questionnaire provide decision-makers in the public and private sectors with a deep and rich set of facts about Canadians, facts that are reliable at the local, regional and national levels.

The concern at the time is that the response rate will likely be substantially lower and the resulting data less robust if the survey became voluntary, given that hard-to reach segments of the population will not likely be included among respondents. The experience of survey researchers and social scientists is that those in lower-income groups, ethnic minorities, and the wealthiest citizens would be least likely to answer questions voluntarily.

This would lead to skewed data and doubts about the accuracy of information that is relied upon by public policy and business decision-makers. Without robust census data, it would be exceedingly difficult for governments to respond effectively to shifting patterns of need in the populace or to introduce changes that provide the greatest value for money. One particularly problematic outcome of the elimination of the mandatory long-form questionnaire would be the eradication of the only reliable, national source of information on aboriginal educational achievement.

Census long-form questionnaire constitute crucial input for the sample designs of other national surveys. The long-form data are also combined with other survey data to compute and extrapolate rates for key social and economic indicators. For example, local health authorities can use their own survey data combined with census data to calculate rates of health service utilization and many other vital statistics.

Also concern about biased results on important dimensions were a concern such as income, education, housing status, and many others. Researchers across the country, working on projects in all areas of public policy and business decision-making, would not have data with which to correct for these biases. Finally there was discussion that it would be impossible for researchers to compare numbers from census to census, and analyze trends.

imageproxy

So here we are in 2013 and what do we have after the survey has been done?

The Globe and Mail reports that there continues to be serious doubts about the value of the NHS when compared to the mandatory long-form census. There is a general consensus that the NHS results are of some use at the national and provincial levels, but of no use at the local level. The harshest critics of the new voluntary system call its data “worthless,” because it does not produce a random, non-biased result, and because it is no longer mandatory – an obligation that had successfully gotten more and more aboriginal communities to take part.

The data from the 2011 NHS is certainly not worthless, but even the chief statistician admits it has not produced as valuable results as the mandatory long-form census would have. He also admits that it cost $22-million more than expected because more copies had to be printed and field workers had to chase down non-respondents.

So let me see, we spent an additional 22 million dollars for an inferior survey because the government was concerned about privacy. Now we all know that governments are big proponents of privacy (if you believe that I have got some cheap swamp land to sell you in Florida). It was not long ago our government introduced an Internet spying bill until there were so many complaints they had to withdraw it.

Canada needs a proper, randomized, long-form census and the data it can produce. Yes we need to modernise and find ways to improve our methodology of doing national surveys and maybe some questions needed to be removed because of their intrusiveness and concerns for privacy, but clearly the decisions made a few years ago are now coming back to haunt us.

Leave a Reply

Your email address will not be published. Required fields are marked *

17 − 5 =