The NYT provides a nifty animated graphic visualizing the large sampling error associated with the monthly jobs report: http://www.nytimes.com/2014/05/02/upshot/how-not-to-be-misled-by-the-jobs-report.html (second screen). What’s nice about it is that you can watch the various scenarios’ random draws and see how easily you fall into pattern recognition mode, despite knowing it’s a simulation.
Not being seduced by good-looking point estimates is one of those things you’d hope everyone would learn as part of basic statistical literacy, but seems to be very rare and I think animations like this are (like xkcd cartoons) the best hope of fixing that particular cognitive bias.
The NYT provides a nifty animated graphic visualizing the large sampling error associated with the monthly jobs report: http://www.nytimes.com/2014/05/02/upshot/how-not-to-be-misled-by-the-jobs-report.html (second screen). What’s nice about it is that you can watch the various scenarios’ random draws and see how easily you fall into pattern recognition mode, despite knowing it’s a simulation.
Not being seduced by good-looking point estimates is one of those things you’d hope everyone would learn as part of basic statistical literacy, but seems to be very rare and I think animations like this are (like xkcd cartoons) the best hope of fixing that particular cognitive bias.