Warning controversial subject regarding math question
- Thread starter Gorcon
- Start date
- Joined
- Apr 12, 2005
- Messages
- 11,339
1) The fact that you keep writing 1.256th when you mean 1/256 does not suggest sufficient ability to interpret such data.
2) The first significant problem that popped out at me is precision. "265 million" is not an exact value. It has only 3 significant digits. It could be, in reality, anywhere from 265,500,000 to 264,500,000. If these are exact, we get:
265,500,000 * 1% / 256 ==> 10,371
264,500,000 * 1% / 256 ==> 10,332
or, using your number and maintaining significant digits at all intermediate values:
265,000,000 * 1% / 256 = 10,400 -- Opinions may vary on exactly how to do this.
In any case, your result of 10,156 does not contribute positively to your eventual argument.
3) Your conclusion doesn't really jump out from your calculations. This is clearly the result of "confirmation bias". That doesn't make it wrong, per se, it just makes it less trustworthy.
4) A LOT more organization, making your argument logical-looking, might be beneficial.
5) Good luck with your crusade. It is an important issue the we should settle rationally, somehow.
2) The first significant problem that popped out at me is precision. "265 million" is not an exact value. It has only 3 significant digits. It could be, in reality, anywhere from 265,500,000 to 264,500,000. If these are exact, we get:
265,500,000 * 1% / 256 ==> 10,371
264,500,000 * 1% / 256 ==> 10,332
or, using your number and maintaining significant digits at all intermediate values:
265,000,000 * 1% / 256 = 10,400 -- Opinions may vary on exactly how to do this.
In any case, your result of 10,156 does not contribute positively to your eventual argument.
3) Your conclusion doesn't really jump out from your calculations. This is clearly the result of "confirmation bias". That doesn't make it wrong, per se, it just makes it less trustworthy.
4) A LOT more organization, making your argument logical-looking, might be beneficial.
5) Good luck with your crusade. It is an important issue the we should settle rationally, somehow.
Last edited:
Following up on tkhunny, I suggest that you use an excel spreadsheet for your data. There are several reasons for that advice.
First, it will preclude errors of arithmetic (though not coding errors, which have to be found by spot checking results.)
Second, it will result in fairly pretty presentations, which should not, but very definitely do, influence acceptance of quantitative arguments.
Third, there are all sorts of tools in excel that may be useful to finding relationships.
Now for a second point. You probably should start by calculating all the ratios potentially relevant for each year before doing any summarization and use excel to look for material trends. The reason is that if there are material trends over time in the data, then any kind of average will not be relevant. To make that clear by an example. If on average 50 some years ago, 1% of murders were with a legally owned handgun and currently 7% of murders are with a legally owned handgun, no one is going to pay much attention to an argument that on average only 4% of murders are with a legally owned handgun. What people will focus on is that things are now seven times worse and will conclude that something must be done.
For a third (highly technical point), if averages are relevant, you probably need to be using the harmonic mean rather than the arithmetic mean to calculate your averages.
See https://towardsdatascience.com/on-a...c-harmonic-means-in-data-analysis-2a703e21ea0
And do not worry that the topic is controversial. So long as you ask questions about the math in your argument, no one should get triggered. As tkhunny indicated, most of us highly respect rational argument.
First, it will preclude errors of arithmetic (though not coding errors, which have to be found by spot checking results.)
Second, it will result in fairly pretty presentations, which should not, but very definitely do, influence acceptance of quantitative arguments.
Third, there are all sorts of tools in excel that may be useful to finding relationships.
Now for a second point. You probably should start by calculating all the ratios potentially relevant for each year before doing any summarization and use excel to look for material trends. The reason is that if there are material trends over time in the data, then any kind of average will not be relevant. To make that clear by an example. If on average 50 some years ago, 1% of murders were with a legally owned handgun and currently 7% of murders are with a legally owned handgun, no one is going to pay much attention to an argument that on average only 4% of murders are with a legally owned handgun. What people will focus on is that things are now seven times worse and will conclude that something must be done.
For a third (highly technical point), if averages are relevant, you probably need to be using the harmonic mean rather than the arithmetic mean to calculate your averages.
See https://towardsdatascience.com/on-a...c-harmonic-means-in-data-analysis-2a703e21ea0
And do not worry that the topic is controversial. So long as you ask questions about the math in your argument, no one should get triggered. As tkhunny indicated, most of us highly respect rational argument.
Steven G
Elite Member
- Joined
- Dec 30, 2014
- Messages
- 14,405
Are you saying that most of us on this forum are rational? WOW!
Steven G
Elite Member
- Joined
- Dec 30, 2014
- Messages
- 14,405
Did you use the word triggered on purpose?!
I do want to confirm that as long as you ask about math we do not care about the topic it is coming from.
Not at all. I merely said we respect rational argument, not that we actually exhibit it,
- Joined
- Apr 12, 2005
- Messages
- 11,339
I found an algebra book, circa 1953 or so. I was a bit humored at the story problem to practice Distant/Rate/Time concepts. It was something like this:
A bomber has enough fuel to fly for 17 hours. It can fly at 430 mph fully loaded. After dropping its payload on the target, it can achieve 470 mph. How far away from the target can the bomber start its flight and just make it back home with an empty tank?
Do you suppose that would fly in a recent algebra book? Like all science, it can be influenced by political and social pressure, although we like to think that it is independent of such things.
A bomber has enough fuel to fly for 17 hours. It can fly at 430 mph fully loaded. After dropping its payload on the target, it can achieve 470 mph. How far away from the target can the bomber start its flight and just make it back home with an empty tank?
Do you suppose that would fly in a recent algebra book? Like all science, it can be influenced by political and social pressure, although we like to think that it is independent of such things.