Double Blind

I believe a voluntary long form census would provide statistically useless information simply because of selection bias – it would make the results easy to manipulate by statisticians as they perform “corrections” to this data.

However, I also do not like the argument of the other side that claims that a mandatory census form would result in reliable information. For example, nothing prevents people from just entering in bogus information in an attempt to retain their confidentiality. I am sure people that don’t like intrusive government questions will just give what they believe to be the most politically correct answer, or an answer that is most to be expected of them when they mail in, as opposed to getting truly reliable answers.

In today’s age, I don’t think anybody believes that the government can keep information secure, so expecting a government agency to keep your census return secure for 90 years is not a reasonable expectation. I am not sure how confidential the census data truly is – if they treated the information just like the CRA treats a tax return then at least there is an audit trail for access to that information. Still, you hear occasionally a CRA employee getting fired or charged for improper access of information, so it makes you wonder how many of these cases were settled internally without it being elevated to the public domain.

As such, while the information is desirable to have, the ability to get that information reliably will continue to degrade over the next few decades.

This entry was posted on Wednesday, July 14th, 2010 and is filed under Commentary. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.