November 16, 2016

Voters’ perceptions of crime continue to conflict with reality

Despite double-digit percentage decreases in U.S. violent and property crime rates since 2008, most voters say crime has gotten worse during that span, according to a new Pew Research Center survey. The disconnect is nothing new, though: Americans’ perceptions of crime are often at odds with the data.

Leading up to Election Day, a majority (57%) of those who had voted or planned to vote said crime has gotten worse in this country since 2008. Almost eight-in-ten voters who supported President-elect Donald Trump (78%) said this, as did 37% of backers of Democrat Hillary Clinton. Just 5% of pro-Trump voters and a quarter of Clinton supporters said crime has gotten better since 2008, according to the survey of 3,788 adults conducted Oct. 25-Nov. 8.

Official government crime statistics paint a strikingly different picture. Between 2008 and 2015 (the most recent year for which data are available), U.S. violent crime and property crime rates fell 19% and 23%, respectively, according to the FBI’s Uniform Crime Reporting Program, which tallies serious crimes reported to police in more than 18,000 jurisdictions around the nation.

Another Justice Department agency, the Bureau of Justice Statistics, produces its own annual crime report, based on a survey of more than 90,000 households that counts crimes that aren’t reported to police in addition to those that are. BJS data show that violent crime and property crime rates fell 26% and 22%, respectively, between 2008 and 2015 (again, the most recent year available).

So what explains the gap between perceptions of crime and the data?

For one thing, official government crime statistics lag behind the times. The FBI and BJS didn’t publish their crime reports for 2015 until fall of this year, meaning they don’t capture recent changes in crime.

Chicago and other large U.S. cities have had well-documented problems with violent crime in 2016 that may have contributed to public perceptions, and a preliminary analysis published by the Brennan Center for Justice in September projects that by year’s end the violent crime rate will have risen nearly 6% from 2015 levels in the nation’s 30 largest cities (including a 13% rise in the murder rate). But even if those trends materialize, the Brennan report cautions that the violent crime rate “remains near the bottom of the nation’s 30-year downward trend.”

Campaign season also may have amplified some voters’ perceptions of rising crime. Trump, in particular, made crime a central focus of his successful campaign for the White House. Citing the recent increases in violent crime in some big cities, he warned at the Republican National Convention that “decades of progress made in bringing down crime are now being reversed.”

But perhaps the best context for understanding the conflict between voters’ perceptions of crime and the data is that voters are usually more likely to say crime is up than down, regardless of what official statistics show.

Since 1989, Gallup has asked respondents whether they think there is more or less crime in the U.S., compared with the year before. In 21 of the 22 years Gallup asked this question, a larger share of respondents said there was more crime. Only in 2001 did roughly equal shares of respondents say there was more crime (41%) versus less (43%). (Gallup has also asked Americans since 1972 whether there was more or less crime in their area; in all but six years, a substantially larger share said crime was up.)

These polling trends stand in sharp contrast to the long-term crime trends reported by the FBI and BJS. Both agencies have documented big decreases in violent and property crime rates since the early 1990s, when U.S. crime rates reached their peak. The BJS data, for instance, show that violent and property crime levels in 2015 were 77% and 69% below their 1993 levels, respectively.

Topics: Criminal Justice, Social Values, Political Issue Priorities, Violence and Society, Political Polarization, 2016 Election

  1. Photo of John Gramlich

    is a writer/editor at Pew Research Center.


  1. Anonymous9 months ago

    The problem with grouping all violent crime together is you lose insight into RANDOM crime. The ONLY violent stat we can really call random crime is robbery. Assault and battery is filled with domestic violence and other known party crimes. Murder isn’t even an accurate depiction because most murder is gang violence.

    If you JUST look at robbery, you will see the trend of random victim violence in a city. And robbery *is* up in all major cities.

    Here’s an oversimplified example of my point. If City A had 10 robberies and 50 domestic violence reports in 2015, and then 20 robberies and 20 DV’s, violent crime appears lower. But IS IT as it pertains to your risk in public? No your risk is now double.

  2. Anonymous9 months ago

    Perceptions bred by the MSM’s “It bleeds ,it leads” mentality and the need to fill the 24 hour news cycle.

  3. Anonymous9 months ago

    Crime is down, but crimes targeting police are up at least 67%. There is a “war on police.” If the police aren’t safe, no one feels safe.

    Another problem is the perception that black/illegal alien criminals disproportionally target white people. Afraid to be called racist, mayors won’t let the police do their jobs aggressively.

    1. Anonymous9 months ago

      “Another problem is the perception that black/illegal alien criminals disproportionally target white people.”

      This is another outstanding example of how “voters’ perceptions of crime continue to conflict with reality” (assuming you vote). That is an objectively false statement; the majority of people commit crimes against individuals of their own race/ethnicity.

  4. Douglas Kelly9 months ago

    News programming in every medium continually overstates crimes, particularly those of a heinous nature and those with any racial or religious elements as well as those in large cities. “If it bleeds, it leads,” has been the byword of news reporting for as long as I can remember. This does nothing but scare the public into believing there more crime than there really is. This creates racial tensions, anti-Muslim attitudes, anti-immigrant attitudes, and more, where none should exist. Therefore, the public sees a need for more and stricter law enforcement, harsher penalties. It creates divisiveness between cultural, religious and national origins.
    In my opinion from observations, much of the incorrect perceptions of crime originate in our governing groups for political gain.

  5. Anonymous9 months ago

    You have conflicting information. On the one hand you ask if crime is up on the survey BUT the reports you cite are “violent crimes and property crimes”. These are very different. Perhaps overlapping the prison populations during that time frame will temper your reporting people are out of touch with reality.

  6. Anonymous9 months ago

    People live in the real world, so one can’t say they’re at odds w reality. The “double-digit percentage decreases” are numbers that are reported. Does everything get reported even if there are no charges? I think the report should focus on how the govt numbers are at odds with reality. If you want to find out what is going on in the real world, the ppl living in it know the truth.

  7. Jack Taylor9 months ago

    People believe this because homocide rates in cities have been climbing:…