Evaluating the New York Times’ investigative report: “A Drumbeat of Multiple Shootings, but America Isn’t Listening”

31 May , 2016  

New York Times Banner

On May 22nd, the New York Times published an investigative report on mass shootings. It claimed that about three-quarters of mass shootings involved blacks. 86 percent occurred in zip codes “poorer than the nation as a whole.” But their data came from less than reliable sources.

“The Times compiled its list of 358 shootings with four or more casualties from largely crowd-sourced lists managed by the social media network Reddit and Gun Violence Archive”

There is a reason that researchers generally rely on FBI and CDC data rather than media reports. News stories often contain only the first, breaking information about a case. The FBI and CDC followed the investigations for these cases and found whether they involved self-defense or accidents or were gang-related.

Relatively few news stories explicitly state whether gangs are involved, and the Gun Violence Archive appears to only count instances where the news stories explicitly mention gangs. The New York Times writes: “Where motives could be gleaned, roughly half [of mass shootings] involved or suggested crime or gang activity.” But this means that, even by their own count, the true number of these cases is above fifty perfect.

The Times reports that for mass shootings: “Nearly three-fourths of victims and suspected assailants whose race could be identified were black. Some experts suggest that helps explain why the drumbeat of dead and wounded does not inspire more outrage.” But an alternative explanation is that gang fights over drug turf don’t draw the same attention as other types of shootings. Gang members, after all, are relatively unsympathetic individuals.

To study this issue, the Times would need to find out the amount of media coverage given to gang shootings where the victims are black. Then the Times would need to do the same for whites, and compare the two results. To see whether the difference is related to the type of shooting, the Times would need to compare media coverage of gang shootings with that of non-gang shootings. But this is not what the Times does.

Though immensely better than relying on news stories, the federal data are not perfect. The federal data rely on data from the individual police departments, and it is clear that they rarely go back and adjust justifiable homicide data after the conclusion of an investigation. Instead, the data are based on preliminary assessments of whether the homicide was justifiable.

The definition of “mass shootings” used by the New York Times could be clearer.  They aren’t just moving from four or more killed to four or more killed or wounded.  But unlike the traditional measures, they apparently include injuries or deaths involving the criminals who committed the crime towards whether they reach four or more killed or wounded.  You could have three gang members who are wounded in an attack that they started and one victim who is wounded, and this case would be included in the New York Times count.

The New York Times could more clearly define the term “mass shootings.” “Mass public shootings” generally must have four or more fatalities, whereas “mass shootings” need only have four or more killed or wounded. But unlike the traditional measures, the Times apparently counts the injuries or deaths of the criminals. The Times could count a case where three gang members and one victim were wounded.

The FBI gives very specific definitions. “Mass public shootings” only include shootings that occur in “public” places such as “commercial areas (divided into malls, businesses open to pedestrian traffic, and businesses closed to pedestrian traffic), educational environments (divided into schools [pre-kindergarten through 12th grade] and IHEs), open spaces, government properties (divided into military and other government properties), houses of worship, and health care facilities.” Most of these places would strike a potential attacker as being ripe for media attention. Since mass public shootings are unique in the fact that they are usually intended to attract attention, it makes sense to have a separate term for shootings committed in public places. The original FBI definition looked at cases where four or more people were killed in an attack (the Obama administration recently changed this, but we use this definition because it facilitates comparisons to earlier data).

The term also excludes shootings that were part of some other crime or that resulted from gang or drug violence. The reason is clear. They want to look at attacks where the intent is simply to kill as many people as possible, not to commit a crime such as robbery. In 2015, there were four cases that satisfied the FBI’s definition of mass public shootings.

The New York Times justifies their definition of “mass shootings” including non-fatal injuries by arguing that it “captures many victims who some criminologists say are too often ignored: people who might have died given a slightly different trajectory of a bullet, or less-sophisticated medical care.” However, we already have FBI data on firearm murders and firearm aggravated assaults that account for the deaths and injuries that the Times wants to study. No explanation is offered for why relying on news stories provides a more accurate measure.

There are other misleading statements. For example, the Times claims that justifiable homicides are a “relatively small subset of the shootings that kill nearly 11,000 people.” But the number of firearm murders in 2014 was 8,124. Justifiable homicides account for the difference between homicides (10,945) and murders (8,124). A justifiable homicide is one without criminal intent to kill someone. Justifiable homicides may be in defense of oneself or of someone else. There is clearly a difference between using a gun to save a life and using one to commit murder.

Previously the CPRC did an analysis of the data used by the New York Times(see here, for our discussion of a similar story by the Washington Post last year).  We looked at data for the first seven months of the year to give people a good idea of how few of these shootings involved either 2 or more killed or 3 or more killed and how many were gang related.

Screen Shot 2016-05-31 at Tuesday, May 31, 12.48 AM

Some other issues for the Gun Violence Archive are raised here.




Leave a Reply

Your email address will not be published. Required fields are marked *