Do the Math
Post Reply
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
12-09-2015, 07:47 PM
Do the Math
According to the The Global Index of Religiosity and Atheism 1% of Americans openly identified as atheists in 2005. By 2012 the number had grown to 5%. If this growth continues at an exponential rate, 23% per year, America will be 50% atheist by 2024.

73% identified themselves as religious in 2005. By 2012 that number had dropped to 60%. If that trend continues at an exponential rate, 3% per year, 40% of Americans will identify as religious by 2024.

I don't expect either of these to happen. The operative word is "exponential." The religious are already alarmed at these figures and are mobilizing. They will slow both rates but not reverse them. Eventually America (and the world) will be atheist.

Sapere aude
Find all posts by this user
Like Post Quote this message in a reply
[+] 2 users Like f stop's post
12-09-2015, 08:05 PM
RE: Do the Math
I always wonder if these types of polls take into account (or try to) the the people who hide their atheism.
Find all posts by this user
Like Post Quote this message in a reply
[+] 5 users Like pablo's post
12-09-2015, 08:43 PM
RE: Do the Math
(12-09-2015 08:05 PM)pablo Wrote:  ... I always wonder if these types of polls take into account (or try to) the the people who hide their atheism ...

I doubt it. I think irreligiosity is much greater than polls portray. Look at how popularly anti-religious humor is received in many entertainment venues, and especially at the overwhelmingly mocking tone of comments in news articles that cover religion related occurrences. My belief is that any half awake semi-literate human mind can't help but perceive the gross absurdities of religion and only fear of social repercussions keeps people from publicly admitting their disbelief - or even privately admitting it.
Find all posts by this user
Like Post Quote this message in a reply
[+] 9 users Like Airportkid's post
12-09-2015, 10:08 PM
RE: Do the Math
(12-09-2015 07:47 PM)f stop Wrote:  According to the The Global Index of Religiosity and Atheism 1% of Americans openly identified as atheists in 2005. By 2012 the number had grown to 5%. If this growth continues at an exponential rate, 23% per year, America will be 50% atheist by 2024.

73% identified themselves as religious in 2005. By 2012 that number had dropped to 60%. If that trend continues at an exponential rate, 3% per year, 40% of Americans will identify as religious by 2024.

I don't expect either of these to happen. The operative word is "exponential." The religious are already alarmed at these figures and are mobilizing. They will slow both rates but not reverse them. Eventually America (and the world) will be atheist.

... how the hell are you getting "exponential growth rate" out of two data points?

Also, I think your math's off. If the growth of those who openly identify as atheists continues at the exponential rate of +37% per year, by 2024 the share of the American population that openly identifies as atheist will be 453%. Big Grin

All teasing about how you're expressing the maths aside, I honestly don't expect the religious to slow these rates. (Well, okay, they'll slow these rates just by being religious, but other than that.) I don't think they know how. I don't think they'll be competent to produce a winning or even stalling strategy. They've selected their population, and especially their leaders, for the ability to believe fantasy and ignore reality. That's going to have a major impact on their ability to problem-solve.

Look, for example, at the large number of Christian leaders who looked at the Pew results this past year, ignored the details that showed that the younger generation is sick of hate based messaging, and promptly declared that they have to be MORE "authentic" in the message they preach, hew more closely to the doctrines and values that are the very ones that this newest generation is rejecting, in order to recover.
Find all posts by this user
Like Post Quote this message in a reply
[+] 4 users Like Reltzik's post
12-09-2015, 10:28 PM
RE: Do the Math
(12-09-2015 08:05 PM)pablo Wrote:  I always wonder if these types of polls take into account (or try to) the the people who hide their atheism.
The data from the poll I cited used only those who openly declared themselves to be atheist.

Sapere aude
Find all posts by this user
Like Post Quote this message in a reply
12-09-2015, 10:52 PM
RE: Do the Math
(12-09-2015 10:08 PM)Reltzik Wrote:  ... how the hell are you getting "exponential growth rate" out of two data points?
The equation for an exponential curve is

P = P'*exp(k*t)

where P = the population at time t, P' = the initial population, k = rate.

Solve for k

P/P' = exp(k*t)

k*t = ln(P/P') = ln(P) - ln(P')

k = [ln(P) - ln(P')] / t

using the two data points

P = 5%, P' = 1%, t = 2012 - 2010 = 7

k = [ln(5) - ln(1)] / 7

k = (1.609 - 0) /7

k = 0.23 = 23%

Sapere aude
Find all posts by this user
Like Post Quote this message in a reply
12-09-2015, 10:54 PM
RE: Do the Math
(12-09-2015 10:08 PM)Reltzik Wrote:  
(12-09-2015 07:47 PM)f stop Wrote:  According to the The Global Index of Religiosity and Atheism 1% of Americans openly identified as atheists in 2005. By 2012 the number had grown to 5%. If this growth continues at an exponential rate, 23% per year, America will be 50% atheist by 2024.

73% identified themselves as religious in 2005. By 2012 that number had dropped to 60%. If that trend continues at an exponential rate, 3% per year, 40% of Americans will identify as religious by 2024.

I don't expect either of these to happen. The operative word is "exponential." The religious are already alarmed at these figures and are mobilizing. They will slow both rates but not reverse them. Eventually America (and the world) will be atheist.
Also, I think your math's off. If the growth of those who openly identify as atheists continues at the exponential rate of +37% per year, by 2024 the share of the American population that openly identifies as atheist will be 453%. Big Grin
It's 23% per year, not 37%.

Sapere aude
Find all posts by this user
Like Post Quote this message in a reply
12-09-2015, 11:11 PM (This post was last modified: 12-09-2015 11:29 PM by Reltzik.)
RE: Do the Math
Actually my question was more "how do we know that this growth curve is exponential rather than linear, quadratic, cubic, logarithmic, logistic, or something else based on just two data points?"

But moving onto the actual numbers...

.... HUH.

.... well, there's good news and there's bad news.

The good news is that I know how to calculate exponents and exponential rates and I can still wield logarithms like the delicate, precision power tools that they are.

The bad news is that I thought that 2012 - 2005 = 5 rather than 7. Facepalm

THIS is why you can't trust people who know higher mathematics.

Anyhow, with your 23% annual continuous growth rate, we should reach 100% of the population at around 2025 and 200% of the population around 2028. (EDIT: Corrected because I was calculating with a baseline of 2015 instead of 2012. Facepalm again.)

Except.... now I'm thinking we both screwed up the formulas. Look at this.

Quote:P = 5%, P' = 1%, t = 2012 - 2010 = 7

k = [ln(5) - ln(1)] / 7

This should actually be

k = [ln(0.05) - ln(0.01)] / 7

Fortunately, the solution is entirely unchanged. Big Grin

....

Okay, fine, I'll stop derailing your "Do the Math" thread with doing math.
Find all posts by this user
Like Post Quote this message in a reply
[+] 5 users Like Reltzik's post
12-09-2015, 11:50 PM
RE: Do the Math
(12-09-2015 10:52 PM)f stop Wrote:  
(12-09-2015 10:08 PM)Reltzik Wrote:  ... how the hell are you getting "exponential growth rate" out of two data points?
The equation for an exponential curve is

P = P'*exp(k*t)

[omitted to save space]

The problem is bigger than that. You can't do any meaningful regression out of two data points, since the standard deviation (i.e. of expected value of Y given X) is undefined for two data points. You can find a linear, exponential, logistical, sinusoidal, etc. and every regression will give you a solution that goes through two points because there's only two of them.
Find all posts by this user
Like Post Quote this message in a reply
[+] 4 users Like ZoraPrime's post
13-09-2015, 06:59 AM
RE: Do the Math
(12-09-2015 11:11 PM)Reltzik Wrote:  Actually my question was more "how do we know that this growth curve is exponential rather than linear, quadratic, cubic, logarithmic, logistic, or something else based on just two data points?"
We don't know. I simply assumed exponential growth/decay because in many real world problems growth/decay turns out to be exponential. It could well be, and probably is, something else. Sheer randomness is a possibility.

(12-09-2015 11:11 PM)Reltzik Wrote:  Anyhow, with your 23% annual continuous growth rate, we reach 100% of the population at around 2025 and 200% of the population around 2028. (EDIT: Corrected because I was calculating with a baseline of 2015 instead of 2012. Facepalm again.)
That's another thing. Even if the exponential assumption holds it probably will hold over a limited domain. Populations, by definition, cannot exceed 100% and the exponential function will eventually predict a population in excess of 100%. There was a joke among engineers back in the 1960s, "The rate we are building highways is growing exponentially. The entire planet will be paved with asphalt by the year 2000."

(12-09-2015 11:11 PM)Reltzik Wrote:  
Quote:P = 5%, P' = 1%, t = 2012 - 2010 = 7

k = [ln(5) - ln(1)] / 7

This should actually be

k = [ln(0.05) - ln(0.01)] / 7

Fortunately, the solution is entirely unchanged. Big Grin
It is unchanged because

ln(0.05) - ln(0.01) = ln(5/100) - ln(1/100)

= [ln(5) - ln(100)] - [ln(1) - ln(100)]

= ln(5) - ln(100) - ln(1) + ln(100)

= [ln(5) - ln(1)] + [ln(100) - ln(100)]

= [ln(5) - ln(1)] + 0

= ln(5) - ln(1)

Sapere aude
Find all posts by this user
Like Post Quote this message in a reply
Post Reply
Forum Jump: