re: #429 Hal_10000

See my comment above. I fit their data as a function of time. There is a small slope (about 1.2 extra shooting victims per year) but it is also statistically consistent with a flat trend. Also tried randomly sampling the data and the trend goes away.

Also note the P<0.05 test is not going to be very effective here because 1) small sample size; 2) the number of sub-samples is large so the likelihood of finding a P<0.05 by chance is large; 3) you still have the bias that the Mother Jones’ is drawn from media coverage, not an objective source like FBI stats; 4) random sampling will show you why the analysis fails. If you remove one or two incidents form the sample, the conclusions change dramatically.

I smell BS. Many of those tests, if they show significance, they do so even while taking into account the small sample size. Kinda basic stats.