Congressman calls survey "unscientific" because it is random | Hacker News Reader

Congressman calls survey "unscientific" because it is random

nytimes.com

22 pointsazernik11 comments

> “This is a program that intrudes on people’s lives, just like the Environmental Protection Agency or the bank regulators,” said Daniel Webster, a first-term Republican congressman from Florida who sponsored the relevant legislation.

I guess I don't have to take this guy seriously anymore. Perhaps tomorrow he'll suggest disbanding police departments?

I had to double check that I wasn't reading The Onion after I read that quote.

Yet he voted for the NDAA.

In Canada, the prime minister did a similar thing, announcing soon before the census's administration that the long version (administered to a random subset) would no longer be mandatory due to "privacy-related complaints".

The decision prompted the resignation of the Chief Statistician, a man appointed to the position by the prime minister. The people in charge of the decision equivocated repeatedly over how and why it happened: http://en.wikipedia.org/wiki/Canada_2011_Census#Voluntary_lo...

A cynic would say that the Prime Minister advocated for the position in order to take advantage of the resulting lack of data and potential for "truthiness", and only used liberty and privacy arguments as political cover.

We had a random long version census in 2011 in Germany, too. It was mandatory. There was a moderately large political debate about the privacy implications. A very small number of people refused to participate, under a threat of penalty that, afaik, was never acted upon. I wasn't asked.

The last real census (in the western part, anyway) before 2011 was in 1987. It was originally planned for 1983, but had to be postponed and modified aftr it was met with huge, widespread resistance. This culminated in the German supreme (ie. constitution,al) court issueing a landmark ruling on privacy rights, establishing informational self-determination as a fundamental right.

The apparent total lack of understanding from the US congress about the very things they legislate upon strikes me lately.

I'm a Canadian myself and our current government isn't any better (or only slightly) in this regard. It seems to me that the basic step would be to ensure you understand what you're talking about before trying to make up laws.

What the hell is wrong with these people. Randomness in the chosen dataset is an inherent concept of statistics. Any high schooler is familiar with this for christ's sake.

What's intriguing is that his bio on his congress homepage says that he has an engineering background; it kinda makes you wonder what kind of engineering.

http://webster.house.gov/Biography/

Very few people I know are aware of statistical research and how it is to be done to avoid bias, etc.

I would not be surprised that Senator Webster is unaware. Especially since he has spent most of his working life as a politician.

I think we should just explain that it's not totally random. The surveyors very carefully choose specific numbers of people from particular demographics and weight the responses, to make a representative picture of the population.

>>“Knowing what’s happening in our economy is so desperately important to keeping our economy functioning smoothly,” said Maurine Haver, the chief executive and founder of Haver Analytics, a data analysis company. “The reason the Great Recession did not become another Great Depression is because of the more current economic data we have today that we didn’t have in the 1930s.”

Dr. Haver's "data analysis" company benefits enormously from the free availability of ACS data, freeing her & her clients from having to gather these data on their own. She is not exactly an impartial voice in this debate.

Sure, that means you should make sure you verify her claims more closely, but that doesn't automatically mean she is wrong.

Reply on news.ycombinator.com