Activity: Police Killings by Race – Faulty Statistics in the Media

Learning Objectives: Students will be able to discuss and debate the conclusions that can be drawn from the statistics presented on a popular TV show; students will also be able to calculate and interpret quantitative evidence in support or against a political argument.

AACU Quantitative Literacy learning outcomes addressed in activity: Interpretation, Representation, Calculation, Application/Analysis, and Communication.

Background: A series of protests began in Ferguson, Missouri after the fatal shooting of Michael Brown, an 18-year old African American male, by Darren Wilson, 28, a white Ferguson police officer. The shooting of Michael Brown occurred on August 9, 2014 and a series of protests occurred in Ferguson 2 weeks after his murder, as well as a week after a grand jury decided not to indict Wilson. Other similar incidents of unarmed black males occurred that year which also received national attention and reactions of protests, including the murder of Trayvon Martin, the chokehold death of Eric Garner. On The O’Reilly Factor, Bill O’Reilly, a conservative pundit, argued that police killings of Blacks is actually decreasing and that Whites are killed more frequently than Blacks by the police. The statistics he used, however, are faulty and do not take into account the proportion of killings in relation to the Black vs. White population in the U.S.

Activity [See accompanying PowerPoint presentation]:

  1. Provide background of the Ferguson protests from December 2014 and the Black Lives Matter activist movement.
  2. Show video of O’Reilly Factor show from December 2014 [If pressed for time, start at 2:20 and end at 3:37]. Bill O’Reilly discusses Ferguson protests and police killings of Blacks vs. Whites. Link to video:
  3. Have students answer the following questions (also in the Powerpoint):

police_killing1a. Let’s assume that the world we live in is Black and White – calculate the percentage of Blacks vs. Whites killed (out of the total # of police killings).

b. Now, also assuming Blacks are 13% of the population in 2012, and Whites are 63% of the population in 2012 – compare the percentages of police killings for Blacks and Whites in 2012 to the percentage of Blacks and Whites in the population (HINT: compare the percentages of Blacks killed by police, to the percentage of Blacks in the population – and do the same for Whites).

c. What makes the statistical statements on Bill O’Reilly’s show faulty? What information did he leave out, that substantially changes the argument he is trying to make that Whites are killed more by police than Blacks?

d. How could the statistics presented on Bill O’Reilly’s show mislead millions of viewers?

e. It turns out that in 2012, the number of Whites killed by police included White Hispanics (which was about 97). That changes the number of non-Hispanic Whites killed by police in 2012 to ~ 227. How does the validity of O’Reilly’s argument change now?

f. In 2014 and 2015, Mapping Police Violence counted 297 people killed by police in the U.S. who were unarmed. Of those people, 117 were African-American, 167 were not, and 13 were not identifiable by race. African-Americans were 14% of the U.S. population in 2013. What proportion of unarmed people killed by police (with an identified race) were African-American in 2013? How does this compare to the proportion of African-Americans in the U.S.?

g. From, we can see how many people were killed by police by month (Jan 2013 – Aug 2015) in total and by race. Calculate the proportion of Blacks vs. other races that were killed by police using 2-3 points on the graph and compare those proportions to the proportions of those races in the U.S. population.

h. [Optional] Write a letter to The O’Reilly Factor or a conservative blog that has posted Reilly’s faulty statistics concerning Blacks vs. Whites killed by police explaining why the statistics they posted are flawed and therefore misleading to the public and how the statistics can be corrected.

Suggested Solutions/Discussion:

 a. Blacks killed = 123/449 = 27.4%; Whites killed = 326/449 = 72.6%

b. 4/13 = 2.1; 72.6/63 = 1.15 ; There is a higher proportion of Blacks killed than Whites.

c. The data presented on the Bill O’Reilly show does not include the percentage of Blacks vs. Whites in the U.S. population. Simply reporting frequencies of killings does not allow an appropriate comparison, since Whites are the majority in this country. In reality, a higher proportion of Blacks are killed by police compared to Whites.

Optional information instructor can relay to students: The data on police killings is limited due to that some police agencies do not have to report justifiable homicides to the FBI; additionally FBI’s data only counts “felons” for which there is no common definition. CDC data also does not get the full picture of police killings of people because officer involvement is not always recorded in the cause of death. (Source: Mother Jones). Though the Death in Custody Reporting Act could eventually change this.

d. Many people do not have the quantitative literacy to critically analyze what O’Reilly presented to them. It’s important to look at proportions of the population rather than frequencies. Watching the statistics presented on the Bill O’Reilly show and not having quantitative literacy skills could influence many people to believe that there is no racism in policing and that there is no validity to the Black Lives Matter movement.

e. This changes the proportion of Whites killed by police to 50%, which is lower than the proportion of Whites in the U.S. population (63%). This gives further evidence that a higher proportion of Blacks are being killed by police compared to Whites. 50/63 = .79; 27.4/13 = 2.1 – The proportion of Blacks killed by police is 2.1 times the proportion of Blacks in the U.S. population. The proportion of Whites killed by police is 0.79 times the proportion of Whites in the U.S. population.

f. 117/(117+167) = 117/284 = 41.2% vs 14% – a much higher proportion of Blacks are being killed by police compared to other racial groups.

g. Answers may vary

h. Letters will vary – it is highly suggested to create a rubric for how the students will be graded if you choose to assign this to them.

Download Content: 

Powerpoint: Faulty Statistics in the Media Activity

Word Document: Faulty Statistics in the Media


(Images from Fox News)