Tweet
Statistics are a tricky thing. They can be used to analyze issues and propose solutions. They can also be used to paint a picture the person using them wants to paint. That’s because there are just so many ways to lay out the data. This was a major issue back at the peak of the inflation period. At the peak of the inflation, Biden made a claim that “inflation for the last month was flat.” Most conservatives and Republicans cried, “liar!” and noted that inflation was still at over 8%. Would you believe they were both right? That’s because each was looking at the inflation number they WANTED to look at.
You see, inflation can be measured year-over-year or month-by-month. A true picture of what is happening requires both numbers. To see this, let’s take an extreme case. Imagine that inflation is flat (0%) for an entire year. Prices for things don’t change from June of last year to June of this year. On July 1st of this year, all prices jump 10%. Then prices don’t change again for another year. At the end of July of this year, the year-over-year inflation number (which compares July of this year with July of last year) reflects 10% inflation. So does the month-by-month inflation number (which compares July of this year with June of this year). At the end of August of this year, however, the year-over-year inflation number (which compares August of this year with August of last year) is still 10%, but the month-by-month inflation number (which compares August of this year with July of this year) is 0%. The first number tells us that our prices compared to last year are significantly higher; the second number tells us that prices are no longer rising.
If you only look at the year-over-year number, you will be left with the impression that things are still getting worse and prices are continuing to rise. The year over year number is going to stay at 10% for every month for a year! This is the message that conservatives and Republicans want to put forward, because it paints the current administration in a bad light. If you look at the month-by-month number, you are left with the impression that “inflation is not happening.” This is the impression Democrats want to leave because it says, “Biden’s policies are working.” If you look at both numbers, you can see the true story: “things jumped badly, but the worst is over.”
There is another reason for emphasizing the month-over-month number: perception and fear. You see, the economy is not just about prices and numbers; it is also about perception. The single greatest thing you can do to trigger a recession is convince everyone there is going to be one. If people believe bad times are coming, they begin to batten down the financial hatches. That slows spending, which triggers sellers to lower prices so as to attract more buyers, which can quickly become a vicious cycle. That cycle can lead to a recession. Mr. Trump has been clear that he hopes for a recession this year, because then he can beat Mr. Biden over the head with it. He has little/no concern that a recession would hurt millions of people; he just wants to win. So he (and his supporters) pound on the message “the economy is bad.” They focus on the year-over-year number.
Mr. Biden, on the other hand, is going to pound on a positive economic message. A positive economic message can avert a recession. It can also sway voters to see him in a better light. Mr. Biden wants to achieve both things: a win politically, and a healthy economy. He tends to focus on the month-by-month number.
You see how statistics can be used to paint the message you want to paint?
Now let’s take the crime claims. To listen to most conservatives and Republicans, you would think that crime is a festering problem in large cities and blue states. They cite huge crime numbers in large cities as their proof. But they ignore a simple fact: if you bring more people together, you are likely to get more crime. Imagine the crime rate was the same everywhere, let’s say one crime per 100 people per year. That would mean that a small town of 500 people would see five crimes in a year, but a city of five million people would see 50,000 crimes. Those who want to paint large cities in a bad light would emphasize the 50,000 crimes per year, and ignore the per capita (per person) rate that shows that crime is the same everywhere: for every 100 people, there will be one crime.
What if we look at crimes per capita for a moment? What would that tell us. It actually tells us that crimes in so-called “red states” (in this case, meaning states that went for Trump in 2020) is higher than it is for “blue states” (i.e., states that went for Biden in 2020). And it is higher by a fairly significant amount. Indeed, red states have had a higher murder rate per capita than blue states for every year from 2000 to 2020. That gap has widened from 9% in 2003 to 43% in 2020. Note that this includes the years Trump was in office. Collectively, the per capita red state murder rate exceed the blue state murder rate for those 21 years by 23%. In fact, if the blue state murder rate had been as high as the red state murder rate, there would be over 45,000 more dead people in the blue states. Even if you take the major cities out of the red states, the red state murder rate exceeds the blue state murder rate by 12%.
Why is this the case? Well, many studies have correlated crime to poverty and education, and the rate of poverty is higher in red states than blue states, as is the percentage of people who do not go beyond a high school education and the percentage of people who do not complete high school. Red states also generally have more lax gun laws, and more guns per capita.
All of this is captured in this article dated from last January (https://www.thirdway.org/report/the-...murder-problem)
So what is missing from the article? There is one number I would love to see explored: shootings per capita instead of “murders per capita.” You see, “murder” only happens when someone dies. Some recent work has shown that “deaths per shooting” is lower in urban and suburban areas because hospitals and ambulance services are nearer and victims can receive treatment more quickly. How would that affect the data? I think I’m safe assuming that there are more shootings than there are deaths. I don’t know how the data per capita would change the red state vs. blue state analysis.
I am not going to guess. All I would be doing is assuming an outcome I want to see. I prefer to just look at the data and see what it tells us. That is, I believe, what we ALL should be doing.
You see, inflation can be measured year-over-year or month-by-month. A true picture of what is happening requires both numbers. To see this, let’s take an extreme case. Imagine that inflation is flat (0%) for an entire year. Prices for things don’t change from June of last year to June of this year. On July 1st of this year, all prices jump 10%. Then prices don’t change again for another year. At the end of July of this year, the year-over-year inflation number (which compares July of this year with July of last year) reflects 10% inflation. So does the month-by-month inflation number (which compares July of this year with June of this year). At the end of August of this year, however, the year-over-year inflation number (which compares August of this year with August of last year) is still 10%, but the month-by-month inflation number (which compares August of this year with July of this year) is 0%. The first number tells us that our prices compared to last year are significantly higher; the second number tells us that prices are no longer rising.
If you only look at the year-over-year number, you will be left with the impression that things are still getting worse and prices are continuing to rise. The year over year number is going to stay at 10% for every month for a year! This is the message that conservatives and Republicans want to put forward, because it paints the current administration in a bad light. If you look at the month-by-month number, you are left with the impression that “inflation is not happening.” This is the impression Democrats want to leave because it says, “Biden’s policies are working.” If you look at both numbers, you can see the true story: “things jumped badly, but the worst is over.”
There is another reason for emphasizing the month-over-month number: perception and fear. You see, the economy is not just about prices and numbers; it is also about perception. The single greatest thing you can do to trigger a recession is convince everyone there is going to be one. If people believe bad times are coming, they begin to batten down the financial hatches. That slows spending, which triggers sellers to lower prices so as to attract more buyers, which can quickly become a vicious cycle. That cycle can lead to a recession. Mr. Trump has been clear that he hopes for a recession this year, because then he can beat Mr. Biden over the head with it. He has little/no concern that a recession would hurt millions of people; he just wants to win. So he (and his supporters) pound on the message “the economy is bad.” They focus on the year-over-year number.
Mr. Biden, on the other hand, is going to pound on a positive economic message. A positive economic message can avert a recession. It can also sway voters to see him in a better light. Mr. Biden wants to achieve both things: a win politically, and a healthy economy. He tends to focus on the month-by-month number.
You see how statistics can be used to paint the message you want to paint?
Now let’s take the crime claims. To listen to most conservatives and Republicans, you would think that crime is a festering problem in large cities and blue states. They cite huge crime numbers in large cities as their proof. But they ignore a simple fact: if you bring more people together, you are likely to get more crime. Imagine the crime rate was the same everywhere, let’s say one crime per 100 people per year. That would mean that a small town of 500 people would see five crimes in a year, but a city of five million people would see 50,000 crimes. Those who want to paint large cities in a bad light would emphasize the 50,000 crimes per year, and ignore the per capita (per person) rate that shows that crime is the same everywhere: for every 100 people, there will be one crime.
What if we look at crimes per capita for a moment? What would that tell us. It actually tells us that crimes in so-called “red states” (in this case, meaning states that went for Trump in 2020) is higher than it is for “blue states” (i.e., states that went for Biden in 2020). And it is higher by a fairly significant amount. Indeed, red states have had a higher murder rate per capita than blue states for every year from 2000 to 2020. That gap has widened from 9% in 2003 to 43% in 2020. Note that this includes the years Trump was in office. Collectively, the per capita red state murder rate exceed the blue state murder rate for those 21 years by 23%. In fact, if the blue state murder rate had been as high as the red state murder rate, there would be over 45,000 more dead people in the blue states. Even if you take the major cities out of the red states, the red state murder rate exceeds the blue state murder rate by 12%.
Why is this the case? Well, many studies have correlated crime to poverty and education, and the rate of poverty is higher in red states than blue states, as is the percentage of people who do not go beyond a high school education and the percentage of people who do not complete high school. Red states also generally have more lax gun laws, and more guns per capita.
All of this is captured in this article dated from last January (https://www.thirdway.org/report/the-...murder-problem)
So what is missing from the article? There is one number I would love to see explored: shootings per capita instead of “murders per capita.” You see, “murder” only happens when someone dies. Some recent work has shown that “deaths per shooting” is lower in urban and suburban areas because hospitals and ambulance services are nearer and victims can receive treatment more quickly. How would that affect the data? I think I’m safe assuming that there are more shootings than there are deaths. I don’t know how the data per capita would change the red state vs. blue state analysis.
I am not going to guess. All I would be doing is assuming an outcome I want to see. I prefer to just look at the data and see what it tells us. That is, I believe, what we ALL should be doing.
, then I win.
Comment