Hi Jeff: The data is what the data is. FBI numbers have their issues but it's all we have as to reported crime. There's nothing wrong with making assumptions using big city dashboards as long as people understand it's limitations.
Based on "Unmasking The Discrepancy: The Overlooked Surge in Violent Crime" addressing a 44 percent increase in violent crime per the 2022 National Crime Victimization Survey.........
The Need For A National Survey-The Great Majority Of Crimes Are Not Reported
The National Crime Victimization Survey was promoted 50 years ago as NECESSARY for understanding crime in America because the vast majority of crime is not reported to law enforcement.
The latest 2022 data from the Bureau of Justice Statistics indicates that 42 percent of violent crime is reported and 32 percent of property crimes are reported. So what the FBI (or big city crime dashboards) offer is a small portion of crime in the US.
Approximately 75-80 percent of what we call crime is property-related. If 32 percent of property crimes are reported, then that means that the overwhelming majority of crimes in the US are not brought to the attention of law enforcement.
Only 7 percent of identity thefts were reported to law enforcement per the Bureau of Justice Statistics. Are we to make policy based on the reported 7 percent or the 93 percent that happened and were not reported?
The National Crime Victimization Survey was demanded by those concerned with understanding the totality of crime in America. The national media criticized a system where only a small percentage of crime was analyzed. Criminologists and politicians insisted on accurate counts.
Now, it’s ignored.
Crimes Reported To Law Enforcement Have Big Issues With Reliability and Accuracy
Beyond being a small subset of total crime, there are other issues to consider with reported crimes:
There is a long history of manipulating data (including homicides) downward at the local level for endless reasons. In Washington, D.C., there are disputes over what counts as a homicide.
There is well-documented mistrust among some minority communities and law enforcement leading to a reluctance to report crimes.
The majority of violent crimes involve someone the victim knows (including family members) making crime reporting difficult. Many of these events are seen by the victim as a private matter.
The wait times for police officers to arrive at a crime scene can be considerable, well over an hour is common. We have lost thousands of police officers due to resignation and retirement. If there’s no report because people were tired of waiting for officers to arrive, there’s no crime counted.
Major law enforcement organizations are still having issues transitioning to the FBI’s National Incident-Based Reporting System (although the percentage has greatly improved). Some suggest that it’s having an impact on crime reporting.
There are over 18,000 police agencies in the US. Getting all to define and report crime accurately is a daunting task. For example, there are thousands of law enforcement agencies stating that there were no hate crimes in their jurisdiction for the latest reporting period.
Some states and cities promote “crime-free” housing meaning that occupants could lose their homes if a crime (i.e., domestic violence) is reported.
Note that arrests have plummeted over the last two decades and arrests declined sharply since the beginning of the police use of force protests (2014) and COVID-19 (2020) thus there is evidence that events affect numbers. If there’s no arrest, was the crime counted in official statistics?
Let’s go back to the Bureau Of Justice Statistics report itself and look at the vast increases in groups. There are two categories, overall rates of violence and rates of what the Bureau of Justice Statistics calls “serious violent crime” (excluding simple assaults). Results:
Huge increases in overall rates of violence and serious violence.
There are huge increases in violence and serious violence for crimes against females.
Huge increases in violence and serious violence for crimes for African Americans and Whites.
Rates doubled for serious violent crimes for Hispanics.
Rates doubled for serious violent crimes for those aged 12-17.
Rates doubled for serious violent crimes for those aged 50 and above.
Rates skyrocketed for serious violent crimes for those in the poorest category.
So for me what you are using for data (big city crime dashboards) is appropriate and useful. The homicide data is probably 98 percent accurate and homicides have traditionally been used as an indicator of overall crime.
It's just that people need to comprehend the limitations of that data. For example, Baltimore has very high rates of crime and violence and they seem to be declining for 2023 police reports. But Maryland can be seen as one large city (with some rural counties with small crime numbers) thus crime reductions for the state may increase regardless of Baltimore’s contribution.
And I'm guessing that reductions in reported crime are a regression to the mean and not a matter of interventions after a 50 percent increase in homicides and a 36 percent increase in aggravated assaults per the Major City Chiefs Association.
I would also point out that there are dozens of cities (using Major City Chiefs Association data) recording increases in categories of violence.
There's definitely a point to be made for 2019 being the last year for good actionable data. 2020 was an outlier for sociopolitical reasons, and the new methodology for 21-23 (and low compliance rates for agencies and reporting nationwide) has made apples to apples comparison something of a difficulty.
I don't think that means we can't compare 22 and 23 data and get some real crunchy stuff out of it though. As you stated quite well, 2022 had better data than 21, and if anything 23 had even more reporting compliance, so a significant reduction in murder across 22-23 is an absolutely valid data point.
If you want to get a general sense of crime from a more reliable source, the CDC's mortality database aggregates every cause for every death that occurs in the US. Unlike police agencies, coroners and medical examiners are required by law to report all deaths, so the data is more complete.
The only problem is the data isn't very timely, as year 2022 deaths should finally be coming out this month. But it also has lots of other interesting stuff like drug overdose and car crash deaths, which are also somewhat of a proxy for crime.
"One of the most common responses I hear from people pushing back on the belief that murder fell…"
I often comment on crime articles by citing data - from multiple sources, gathered by different methods - all of which show that US crime is low (except for gun crime, but that's a somewhat separate issue) and has steadily dropped for decades. The pushback from The Fearful is fascinating.
There is a portion of our citizenry who insist on believing, who seemingly NEED to believe, that crime is rampant and that we're all in Danger (Yikes!). Where does that come from? Is it Faux Nooz and/or the manipulative (R) pols who gin up the Fear of Crime to their political benefit? Is it simply myopic Fear of The Other? Seems like some smart peeps should study that, doncha' think?
"all of which show that US crime is low (except for gun crime"
How are you defining low and high here and why do they differ? They're kind of subjective, right? As is how to feel about them. To me, at least, believing gun and non-gun crime are both high seems less strange than believing one is high and the other is low
Nah, you missed my (obtusely worded) point. There's gobs of crime data that are, by definition, objective (data being numbers-based, and all that) and they all show that all categories of US crime are at, or near, their all-time lows - far lower than at their peak in the early 90's . No subjectivity about it. Also, overall US crime is roughly comparable to crime in "more peaceful" Euro nations - except for Gun Crime.
US Gun Crime is far higher than in our peer nations and far higher than it should/could be in the US, due in part to our sheer numbers of guns (more guns than peeps in the US) combined with the unwillingness of (R) politicians to support the sane gun laws that are supported by the great majority of US citizens. Data consistently shows that the more guns in a nation/state/municipality/household, the more gun crime there is (Duh!).
While our gun crime is far higher than it could/should be, the actual odds of any American suffering from gun crime are very low. The highest # of Gun Homicides/year in the US was 21,036 (2022), meaning that only .00063% of our 333 Million citizens were slain by another person with a gun (and most murders are committed w/ a gun). The highest number of shootings was 41,605, thus only .0012% of us were wounded by another w/ a gun. Furthermore, most gun crime occurs in a small number of historically violent zip codes and many victims were acquainted w/ their attackers.
The odds of a random American being shot/killed by a random bad guy w/ a gun are very low.
I agree crime is lower than in the 90s and there's nothing subjective about that. I just wasn't sure how to interpret current levels as high or low - that part could be subjective but with a definite frame of reference it becomes a factual data point.
But on overall crime I'm not sure the comparison to Europe is really all that useful. Data on crime reporting, particularly crimes less serious than a shooting with a victim, is very spotty and varies enourmously by communitiy, police agency and jurisdiction.
It's very, very hard to compare crime in Europe to crime here. Certainly there's more gun crime here, but you can't really look at, say rape or larceny statistics here and there and say which is worse. Everything from the legal definition to the counting (if you steal 3 laptops is that one crime or 3?) to the reporting is different. I suspect, though, given the way higher level of even non-gun homicide here, that we are probably much worse than them in most violent crime categories despite whatever the "official" apples to oranges numbers say.
On gun laws, probably there are some better gun control laws that could decrease shootings, but I doubt there's very many. What do you think would move the needle? If you compare countries in North America and the Caribbean, many of the countries with the strictest gun laws (Mexico, Jamaica) have the highest murder rates.
And within countries the majority of gun crimes are limited to a small set of violent neighborhoods that only a fraction of poor neighborhoods and a tiny fraction of all neighborhood. People don't shoot each other much in relatively poor El Paso, Texas, but they do a lot in Juarez, Mexico across the river, despite the stricter gun laws. They don't shoot each other much in relatively poor West Virginia, either, but similarly rural and poor rural Mississippi leads the nation despite very similar gun laws. How do you explain that if it's not local culture?
Hi Jeff: The data is what the data is. FBI numbers have their issues but it's all we have as to reported crime. There's nothing wrong with making assumptions using big city dashboards as long as people understand it's limitations.
Based on "Unmasking The Discrepancy: The Overlooked Surge in Violent Crime" addressing a 44 percent increase in violent crime per the 2022 National Crime Victimization Survey.........
The Need For A National Survey-The Great Majority Of Crimes Are Not Reported
The National Crime Victimization Survey was promoted 50 years ago as NECESSARY for understanding crime in America because the vast majority of crime is not reported to law enforcement.
The latest 2022 data from the Bureau of Justice Statistics indicates that 42 percent of violent crime is reported and 32 percent of property crimes are reported. So what the FBI (or big city crime dashboards) offer is a small portion of crime in the US.
Approximately 75-80 percent of what we call crime is property-related. If 32 percent of property crimes are reported, then that means that the overwhelming majority of crimes in the US are not brought to the attention of law enforcement.
Only 7 percent of identity thefts were reported to law enforcement per the Bureau of Justice Statistics. Are we to make policy based on the reported 7 percent or the 93 percent that happened and were not reported?
The National Crime Victimization Survey was demanded by those concerned with understanding the totality of crime in America. The national media criticized a system where only a small percentage of crime was analyzed. Criminologists and politicians insisted on accurate counts.
Now, it’s ignored.
Crimes Reported To Law Enforcement Have Big Issues With Reliability and Accuracy
Beyond being a small subset of total crime, there are other issues to consider with reported crimes:
There is a long history of manipulating data (including homicides) downward at the local level for endless reasons. In Washington, D.C., there are disputes over what counts as a homicide.
There is well-documented mistrust among some minority communities and law enforcement leading to a reluctance to report crimes.
The majority of violent crimes involve someone the victim knows (including family members) making crime reporting difficult. Many of these events are seen by the victim as a private matter.
The wait times for police officers to arrive at a crime scene can be considerable, well over an hour is common. We have lost thousands of police officers due to resignation and retirement. If there’s no report because people were tired of waiting for officers to arrive, there’s no crime counted.
Major law enforcement organizations are still having issues transitioning to the FBI’s National Incident-Based Reporting System (although the percentage has greatly improved). Some suggest that it’s having an impact on crime reporting.
There are over 18,000 police agencies in the US. Getting all to define and report crime accurately is a daunting task. For example, there are thousands of law enforcement agencies stating that there were no hate crimes in their jurisdiction for the latest reporting period.
Some states and cities promote “crime-free” housing meaning that occupants could lose their homes if a crime (i.e., domestic violence) is reported.
Note that arrests have plummeted over the last two decades and arrests declined sharply since the beginning of the police use of force protests (2014) and COVID-19 (2020) thus there is evidence that events affect numbers. If there’s no arrest, was the crime counted in official statistics?
Let’s go back to the Bureau Of Justice Statistics report itself and look at the vast increases in groups. There are two categories, overall rates of violence and rates of what the Bureau of Justice Statistics calls “serious violent crime” (excluding simple assaults). Results:
Huge increases in overall rates of violence and serious violence.
There are huge increases in violence and serious violence for crimes against females.
Huge increases in violence and serious violence for crimes for African Americans and Whites.
Rates doubled for serious violent crimes for Hispanics.
Rates doubled for serious violent crimes for those aged 12-17.
Rates doubled for serious violent crimes for those aged 50 and above.
Rates skyrocketed for serious violent crimes for those in the poorest category.
So for me what you are using for data (big city crime dashboards) is appropriate and useful. The homicide data is probably 98 percent accurate and homicides have traditionally been used as an indicator of overall crime.
It's just that people need to comprehend the limitations of that data. For example, Baltimore has very high rates of crime and violence and they seem to be declining for 2023 police reports. But Maryland can be seen as one large city (with some rural counties with small crime numbers) thus crime reductions for the state may increase regardless of Baltimore’s contribution.
And I'm guessing that reductions in reported crime are a regression to the mean and not a matter of interventions after a 50 percent increase in homicides and a 36 percent increase in aggravated assaults per the Major City Chiefs Association.
I would also point out that there are dozens of cities (using Major City Chiefs Association data) recording increases in categories of violence.
Best, Len.
More importantly, why does everyone believe crime has exploded?
There's definitely a point to be made for 2019 being the last year for good actionable data. 2020 was an outlier for sociopolitical reasons, and the new methodology for 21-23 (and low compliance rates for agencies and reporting nationwide) has made apples to apples comparison something of a difficulty.
I don't think that means we can't compare 22 and 23 data and get some real crunchy stuff out of it though. As you stated quite well, 2022 had better data than 21, and if anything 23 had even more reporting compliance, so a significant reduction in murder across 22-23 is an absolutely valid data point.
If you want to get a general sense of crime from a more reliable source, the CDC's mortality database aggregates every cause for every death that occurs in the US. Unlike police agencies, coroners and medical examiners are required by law to report all deaths, so the data is more complete.
The only problem is the data isn't very timely, as year 2022 deaths should finally be coming out this month. But it also has lots of other interesting stuff like drug overdose and car crash deaths, which are also somewhat of a proxy for crime.
"One of the most common responses I hear from people pushing back on the belief that murder fell…"
I often comment on crime articles by citing data - from multiple sources, gathered by different methods - all of which show that US crime is low (except for gun crime, but that's a somewhat separate issue) and has steadily dropped for decades. The pushback from The Fearful is fascinating.
There is a portion of our citizenry who insist on believing, who seemingly NEED to believe, that crime is rampant and that we're all in Danger (Yikes!). Where does that come from? Is it Faux Nooz and/or the manipulative (R) pols who gin up the Fear of Crime to their political benefit? Is it simply myopic Fear of The Other? Seems like some smart peeps should study that, doncha' think?
"all of which show that US crime is low (except for gun crime"
How are you defining low and high here and why do they differ? They're kind of subjective, right? As is how to feel about them. To me, at least, believing gun and non-gun crime are both high seems less strange than believing one is high and the other is low
Nah, you missed my (obtusely worded) point. There's gobs of crime data that are, by definition, objective (data being numbers-based, and all that) and they all show that all categories of US crime are at, or near, their all-time lows - far lower than at their peak in the early 90's . No subjectivity about it. Also, overall US crime is roughly comparable to crime in "more peaceful" Euro nations - except for Gun Crime.
US Gun Crime is far higher than in our peer nations and far higher than it should/could be in the US, due in part to our sheer numbers of guns (more guns than peeps in the US) combined with the unwillingness of (R) politicians to support the sane gun laws that are supported by the great majority of US citizens. Data consistently shows that the more guns in a nation/state/municipality/household, the more gun crime there is (Duh!).
While our gun crime is far higher than it could/should be, the actual odds of any American suffering from gun crime are very low. The highest # of Gun Homicides/year in the US was 21,036 (2022), meaning that only .00063% of our 333 Million citizens were slain by another person with a gun (and most murders are committed w/ a gun). The highest number of shootings was 41,605, thus only .0012% of us were wounded by another w/ a gun. Furthermore, most gun crime occurs in a small number of historically violent zip codes and many victims were acquainted w/ their attackers.
The odds of a random American being shot/killed by a random bad guy w/ a gun are very low.
I agree crime is lower than in the 90s and there's nothing subjective about that. I just wasn't sure how to interpret current levels as high or low - that part could be subjective but with a definite frame of reference it becomes a factual data point.
But on overall crime I'm not sure the comparison to Europe is really all that useful. Data on crime reporting, particularly crimes less serious than a shooting with a victim, is very spotty and varies enourmously by communitiy, police agency and jurisdiction.
It's very, very hard to compare crime in Europe to crime here. Certainly there's more gun crime here, but you can't really look at, say rape or larceny statistics here and there and say which is worse. Everything from the legal definition to the counting (if you steal 3 laptops is that one crime or 3?) to the reporting is different. I suspect, though, given the way higher level of even non-gun homicide here, that we are probably much worse than them in most violent crime categories despite whatever the "official" apples to oranges numbers say.
On gun laws, probably there are some better gun control laws that could decrease shootings, but I doubt there's very many. What do you think would move the needle? If you compare countries in North America and the Caribbean, many of the countries with the strictest gun laws (Mexico, Jamaica) have the highest murder rates.
And within countries the majority of gun crimes are limited to a small set of violent neighborhoods that only a fraction of poor neighborhoods and a tiny fraction of all neighborhood. People don't shoot each other much in relatively poor El Paso, Texas, but they do a lot in Juarez, Mexico across the river, despite the stricter gun laws. They don't shoot each other much in relatively poor West Virginia, either, but similarly rural and poor rural Mississippi leads the nation despite very similar gun laws. How do you explain that if it's not local culture?