3 ms·
Is it really that hard to do a t-test comparing the buys and the sells to find out if the difference is statistically significant at this sample size? I'm sure
by erikpukinskis 13y ago
Is it really that hard to do a t-test comparing the buys and the sells to find out if the difference is statistically significant at this sample size? I'm sure there are better tests, but that one is literally the first test you learn in a statistics course. Are there no journalists at the Post who have taken a basic stats course?
They could literally just punch in the 10 numbers (buys and sells for each of the five events) into http://studentsttest.com/ http://studentsttest.com/ and get a result.
We should really figure out some way to enable more science students to pursue journalism. What's the point of having thousands of brilliant scientists doing research if journalists are just going to mangle it every time? Distribution is just as important as production.