The Perception Gap SOC 3333 · Crime Lab · Midwestern State University
This page is about the distance between what Americans believe about crime and what the data show. Chapter 1 says criminology is the scientific study of crime, which only matters if science and common sense disagree. They do, and you are about to measure the disagreement on yourself. Lock in your guesses before anything is revealed. Your numbers are not recorded anywhere, so keep track of them. You will need them for your Forum 1 post.
First, a fact. Then, your guess.
Here is the fact: in 2025 the United States recorded what the FBI called the largest one-year drop in violent crime and murder since it began keeping national records in 1937. Murder fell an estimated 18.1 percent in a single year. That October, while the drop was underway, Gallup asked Americans the question it has asked since 1989: is there more crime in the U.S. than there was a year ago, or less?
What percent of Americans said there was MORE crime in the U.S. than a year ago, in October 2025?
49 percent. In the best national crime year ever measured, about half the country said crime was rising. And that was the good news: 2025 was the first reading below 50 percent since 2001. In 2023, with murder already falling at a record pace, 77 percent said crime was up.
Same survey, one more question. What percent said there was more crime in their OWN local area?
30 percent. People consistently rate the nation as more dangerous than the place they actually live. Hold on to that pattern. If most people think their own area is fine, where is the impression of a dangerous nation coming from? That question is the seed of your forum post.
Sources: Gallup, "Crime in U.S. Seen as Less Serious for Second Straight Year," survey of Oct 1 to 16, 2025 (49% more crime nationally, 30% locally, 77% in 2023, 64% in 2024). FBI, "FBI Releases Historic Early Look at Annual Crime Data," May 13, 2026 (murder down est. 18.1% in 2025, largest decline since 1937).
Two lines that should move together, and do not
The orange line is belief: the percent telling Gallup there is more crime in the U.S. than a year ago, in every year Gallup asked. The blue line is a hard outcome: the U.S. murder rate per 100,000 people. Murder is the best measured of all crimes, which is why criminologists lean on it. Drag the slider across the years and watch what belief does while murder falls.
The scorecard across 37 years: a majority said crime was rising in 13 of these 17 readings. Over the same span the murder rate fell from a 1991 peak of 9.8 per 100,000 to about 4.1 today, cut by more than half. The belief barely responded to two decades of decline, spiked with cable news and elections, and only began matching reality in 2024 and 2025.
Honesty note: this chart puts two different measures on two different scales in one picture. That is done to compare direction, not size, and it is exactly the kind of design choice Chapter 2 will teach you to interrogate. Gallup did not survey every year, so the lines connect survey years only. The 2025 murder rate is preliminary (FBI early-look estimate applied to the 2024 rate, roughly 4.1). Sources: Gallup crime trend tables (percent "more crime," 1989 to 2025). Murder rates: FBI Uniform Crime Reports compilations (1989 to 2019), FBI/BJS published estimates 2020 to 2023, FBI "Crime in the Nation, 2024" (5.0 per 100,000), FBI early look for 2025.
What the measurements actually say
Three independent measurement systems, one story. This is what "scientific study" means in practice: not one number, but several imperfect counts that agree with each other.
The FBI's early look at 2025 (17,000+ agencies covering 96% of the population)
The Council on Criminal Justice's city sample, year-end 2025
Homicide fell 21 percent across the 35 large cities reporting it, and CCJ projects the 2025 national homicide rate at roughly 4.0 per 100,000, which would be the lowest ever recorded in data reaching back to 1900.
The decline has continued into the following year. Across 586 agencies reporting January through May 2026, which offense dropped the most compared with the same months a year earlier?
Every category fell again the following year. Bar lengths are drawn relative to the largest decline. You can check your own city against these numbers at crimeindex.org, linked in the LMS. A criminological question worth carrying into the forum: the decline is real, national, and historic. Why is it not a bigger story?
Sources: FBI early look at 2025 crime data (released May 13, 2026). Council on Criminal Justice, "Crime Trends in U.S. Cities: Year-End 2025 Update" (40-city sample). The Crime Index (AH Datalytics), January to May 2026 year-to-date, 586 agencies. Retrieved August 2026.
Crime, deviance, both, or neither?
Chapter 1 draws a line between crime (an act that violates criminal law) and deviance (an act that violates social norms). The two overlap, but neither contains the other. Sort each act below. Some have a clean answer and some are genuinely contested, and the contested ones teach the most.
What the sort shows. The legalistic definition gives you a crisp rule, an act is a crime when a criminal statute forbids it, but it stays silent about which acts get statutes and which statutes get enforced. The interactionist definition says crime is whatever a society successfully labels as crime, which explains why your answers changed with jurisdiction, era, and enforcement. Sutherland's three domains are all here: the making of laws (who wrote the wage theft rules?), the breaking of laws (everyone speeds), and reactions to law-breaking (who actually gets arrested?).
The same act, wandering across the line
In 1969, Gallup asked Americans for the first time whether marijuana use should be legal. Twelve percent said yes. The act has not changed since 1969. The chemistry has not changed. What changed is the label, which is the interactionist point in one drug.
Support was 12 percent in 1969. What percent of Americans said marijuana should be legal in Gallup's 2025 survey?
64 percent, and the law followed the norm. Majority support arrived in 2013, right after Colorado and Washington voted to legalize. Today marijuana is legal for adults in roughly half the states and still a crime in Texas. Notice the 2025 number also dipped from the 2023 record, a reminder that moral boundaries move in both directions. Mala prohibita acts, wrong because prohibited, are exactly the acts where you should expect the line to wander. Nobody expects a Gallup rebound for robbery.
Sources: Gallup, "Grassroots Support for Legalizing Marijuana Hits Record 70%" (1969: 12%, 2013: 58%, 2023: 70%) and Gallup 2025 survey as reported by NORML, Nov 2025 (64%, down from 68% in 2024).
Forum 1: your post
Write roughly 250 to 300 words for your initial post, then reply to two classmates. Post in the Forum 1 thread in the LMS, and check that week’s module for the posting and reply deadlines. Build your post around these four moves:
- Before opening this page, did you believe crime in the U.S. was rising, falling, or holding steady? Name where that impression came from: news, social media, family, personal experience, true crime shows, all of the above.
- Report your numbers. What did you guess on the Start here tab for the national question, and how far off were you from the published 49 percent? Nothing here is recorded, so quote your own guesses.
- Explain the gap with the chapter. Pick at least one concept from Chapter 1, define it in your own words, and use it to explain why belief about crime and measured crime can drift so far apart for so long. Good candidates: criminology as scientific study versus common sense, Sutherland's three domains, the consensus and conflict perspectives on where law comes from, or the interactionist definition of crime.
- End with one sentence on what evidence would change your mind about crime trends, and where you would go to find it.
For your replies: extend or challenge a classmate's explanation of the gap using course material. If they blamed the media, ask which audiences and whose interests. If they leaned on consensus, offer the conflict reading, or the reverse. "Great post, I agree!" earns no credit, and you knew that already.