/www.nature.com/articles/s41586-021-04198-4 (via @emollick)
Surveys are a crucial tool for understanding public opinion and behaviour, and their accuracy depends on maintaining statistical representativeness of their target populations by minimizing biases from all sources. Increasing data size shrinks confidence intervals but magnifies the effect of survey bias: an instance of the Big Data Paradox … We show how a survey of 250,000 respondents can produce an estimate of the population mean that is no more accurate than an estimate from a simple random sample of size 10
It basically confirms my earlier observation in asthma genetics
this result was possible with just 415 individuals instead of 500,000 individuals nowadays
Tom Brewster had an interesting idea: selling hsi own data. Why should anyone else make money with it?
When I decided to sell the secret details of my personal life, I had high hopes I’d get a willing buyer. It didn’t go well.
I had been curious to see if I could make money from my online information – something that data brokers across the world are doing every day; collecting it, combining it with others’ information and flogging it to marketing firms or anyone willing to pay. So I put myself on eBay.
The article is really interesting to read, yea, yea.
It is a rather old topic here at Science Surf –searching for truth in science while here is a fresh new look. How can we trust twitter messages and alike, is there any truth in big data? What is just an internet meme? Or a phenomenon — a meaning what is experienced as given? A pheme? A new EU project explains Continue reading Veracity