Search the phrase “every 28 hours” on Google, Twitter, Instagram or Facebook and thousands of messages like these will come up. This is the phrase that has gone viral after police-involved killings of black males in Ferguson, Mo., and Staten Island, N.Y.: “Every 28 hours, an unarmed black person is killed by police.” It is now being used to describe the deaths of two New York City police officers, to draw comparisons between the deaths of police and deaths of unarmed African Americans killed by police.
PolitiFact rated this claim false in August when a CNN commentator Marc Lamont Hill said it on air. Hill corrected his statement after the PolitiFact article.
Yet the claim is still being perpetuated widely. It’s written on signs at protests, shared through #every28hours or written on social media memes. Where did this statement come from, and is it accurate?
The figure “every 28 hours” comes from an April 2013 report titled“Operation Ghetto Storm” by the Malcolm X Grassroots Movement. The report analyzed officer-involved killings of African-American victims in 2012. The report “is a window offering a cold, hard, and fact-based view into the thinking and practice of a government and a society that will spare no cost to control the lives of Black people,” according to the preface.
It’s not hard to debunk the claims using basic findings and methodologies from the report. (Twitter user @FeministaJones did it in a series of tweetsusing Storify.)
The report looked at the deaths of 313 African-American men, women and children in 2012, who were killed by police officers, security guards or “vigilantes.” Vigilantes are defined “self-appointed enforcers of the law” who had immunity under self-defense legal protections such as the stand-your-ground law or the castle doctrine.
If you divide the number of hours in a year (8,760) by 313 deaths, it does come out to one death per 28 hours. But the rest of the claim is problematic.
Among those killed in 2012, 136 people (44 percent) had no weapon on them when they died. That negates the claim that people who were killed “every 28 hours” were unarmed. Since the study looked at police officers, security guards, police officers moonlighting as security guards, and “vigilantes” that do not fit under any of the other categories, the claim that an African American is killed “every 28 hours” by a police officer is also false.
The “every 28 hours” figure is being quoted by thousands of people, including those who accurately cite the figure based on the report’s findings.
The report’s author, Arlene Eisen, wrote in an article in September that the use of “every 28 hours” as a standalone figure oversimplifies the point made in her report, which was to “prevent future extrajudicial killings of Black people by those paid or sanctioned (security guards and vigilantes) by the national security state, it is important to know that these killings are a result of the perpetual war on Black people.”
There is no accurate or comprehensive figure for the rate of homicides by police in the United States. The Killed By Police Facebook page posts news reports of officer-involved homicides starting May 1, 2013. The FBI’s database is limited, and uses self-reported and conservative estimates. It’s even more difficult to accurately study the racial breakdowns of victims who were killed by police. Even so, the data yield an interesting statistic: A white police officer killed a black person nearly two times a week during a seven-year period ending in 2012, according to a USA Today review of the most recent accounts of justifiable homicide reported to the FBI.
In the absence of reliable official reports, it might make some sense that protesters turned to Eisen’s research of news articles, police department announcements, obituaries, interviews with crime beat reporters, social media postings and other sources.
Yet the methodology of this report is curious.
It takes into account deaths that do not seem to fit into the report’s description of “thinking and practice of a government and a society that will spare no cost to control the lives of Black people.” For example, a 24-year-old woman died in July 2012 at her birthday party, where she hugged an off-duty police officer from behind. He was carrying a loaded gun, which allegedly went off and killed her. A 59-year-old woman died after she was hit by a police officer driving his patrol car responding to an emergency call. A 31-year-old man was fatally shot by police when he ignored commands to stop eating a homeless man’s face.
When The Fact Checker inquired how Eisen arrived at the cases she listed, she said she took cities where she knew African Americans lived in, and searched on Google for “officer involved killing in (city name).” She also corroborated those numbers with cases she knew of in other cities.
Eisen agreed she included outliers that do not necessarily fit the theory that there is a “war against black people.” But even without those outliers, Eisen said the study confirms the point that there is a systematic problem of law enforcement and “other extrajudicial killers” using excessive force that is justifiable by law: “For every case you would want to take out of there (as an outlier), I would say if I had the resources, or the police were required to report, I could replace that case with three or four more,” she said.
Eisen acknowledged the report’s figures are not accurate. She said the real figure may be even higher — more like every 24 hours, a black person is killed by a police, security guard or a vigilante.
“I don’t think the burden should be on me to prove that it’s an accurate list. I’m saying it’s not accurate. It’s a way undercount. If you want a full count, you should demand it from the Department of Justice,” she said.
The Pinocchio Test
There are many variations of the report’s findings out on the Internet. This fact check focused on the claim that “every 28 hours, an unarmed black person is killed by police” — which is not true, based on the findings in the Malcolm X Grassroots Movement report. The victims studied in the report were not all unarmed, and they were not all killed by police.
Yet this alarmist statement continues to be perpetuated on social media and in protests, and it is inaccurate, especially when distilled to a hashtag. Other variations of the statement may be more reflective of the report’s findings, but the report itself does not tell the full picture, either. Perhaps Eisen is correct that her numbers reflect an undercount, but she has little evidence to back that up; it’s just a hunch at this point. Protesters would do better to cite the FBI statistic, which is powerful enough by itself.
Remember: Just because it’s on the Internet doesn’t mean it’s true. And think before you retweet.