Lying With Statistics 101:
Apr. 15th, 2014 08:30 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
theweaselking
"OK, here are the stats. Make a graph so it looks like gun deaths have fallen since 2005."
*graph guy sweats*
— hrtbps (@hrtbps) April 15, 2014
Page generated Mar. 1st, 2026 01:02 pm
(no subject)
Date: 2014-04-15 12:39 pm (UTC)-- Steve's not sure whether this mitigates or spurs his disdain.
(no subject)
Date: 2014-04-15 12:44 pm (UTC)Nicked from the weaselking
Date: 2014-04-15 02:19 pm (UTC)(no subject)
Date: 2014-04-15 02:59 pm (UTC)(no subject)
Date: 2014-04-15 03:49 pm (UTC)(no subject)
Date: 2014-04-15 05:34 pm (UTC)But there's supposed to be only one copy of the image. It's just set so that INCREASED murders are shown as a DROP on the graph.
(no subject)
Date: 2014-04-15 03:44 pm (UTC)(no subject)
Date: 2014-04-15 05:03 pm (UTC)(no subject)
Date: 2014-04-15 07:13 pm (UTC)(That said, I did go through a mental process of "there is fuckery here. ...aha! The headline talks about gun deaths, but the graph title is specifically about murders. I guess that gun killings committed under SYG law aren't legally murders, and the graph is being used to suggest that because there are fewer killings which meet the legal def of murder, there are fewer gun deaths, even though that's not the case." And then, having an explanation, I stopped looking.)
(It's been a bad kind of day.)
(no subject)
Date: 2014-04-15 10:14 pm (UTC)(no subject)
Date: 2014-04-15 10:27 pm (UTC)(Also worth noting that the graph talks about firearm "murders", and technically a person who successfully invokes Stand Your Ground as a defense isn't a murderer per se, they just committed homicide in "self defense". So if the people who designed this really used murder counts to make the graph, that's also weaselly. They should use deaths by firearm instead for a more honest assessment.)
(no subject)
Date: 2014-04-15 10:58 pm (UTC)(no subject)
Date: 2014-04-16 01:13 am (UTC)(no subject)
Date: 2014-04-16 01:09 am (UTC)(no subject)
Date: 2014-04-16 01:21 am (UTC)(If you'd like to claim that the lying graph ALSO conflates fake-justified "stand your ground"-aquittal homicides with standard murders, feel free. Link sources. As it is, the chart implies that *murders* went down while sourcing statistics that say *murders* went up. While taking numbers from a source that has a vested interest in distinguishing "murders" from fake-justified "stand your ground"-aquittal homicides, and thus presumably made that distinction.)
(no subject)
Date: 2014-04-16 01:04 am (UTC)(no subject)
Date: 2014-04-16 01:09 am (UTC)(no subject)
Date: 2014-04-16 05:07 am (UTC)(Because Reuters managed to not make the attribution for the data source misleading, unlike the title and the graph.)
ETA:
The FDLE specifically notes that justifiable homicides "should not be included in summary offense totals for homicide" (page 7).
The FDLE also notes that when it applies the label of "justified homicide" it is not referring to a determination made by a court of law, but only to an investigating officer's scoring of whether or not the death in question was "the killing of the perpetrator of a serious criminal offense either by a law enforcement officer in the line of duty or by a private citizen, during the commission of a serious criminal offense."
(Just to make sure everyone is on exactly the same page, and no-one is going to try to start using FDLE statistics to talk about firearm-deaths-where-the-shooters-were-legally-aquitted. Because that would be even more disingenuous than implying that all incidents in which SYG was invoked had been either-or situations wherein one person had to be shot or another person would die.)