World population is growing 20 million people every 3 months

[K_V_B]

Originally Posted by UAEguy

Suppose you want to determine the uncertainty in the final value of a quantity that is calculated from several measured quantities. The uncertainties in these measured quantities propagate through the calculation to produce uncertainty in the final result. Consider the following example. Suppose you know the average mass of one apple m with the uncertainty Δm. If you want to calculate the mass M of the basket of 100 apples, you will get the value
M ± ΔM = 100 m ± 100 Δm.

Ok. But if I for example measure each apple with an accuracy if 1g, and then add all the weights I found together. How accurate is now my estimate of the weight of the basket?

The point is that stochastic, thus random errors will average to 0. That is why you can by taking more samples increase your accuracy, even if your instrument remains the same.

[boatman61]

Originally Posted by K_V_B

The mistake you are making here is that you assume that if the underlying data has a certain inaccuracy, that any other measure or conclusion you derive from that data necessarily will be less accurate.

That is where your lack of knowledge regarding statistics shows. Because this assumption is wrong.

Achieving good results from bad data, is basic applied statistics. Something you need to be able to do in about any job that involves data. Not just at UN. In fact, if you are a good statistician you are probably not look at working for the UN. You would work a (well paid) job at an insurance company or a bank...
.

Because '******** Baffles Brains'...

[Adelie]

Snippy, ad hominem posts are about to get this thread closed.

Argue the positions not the debater.

[UAEguy]

Originally Posted by K_V_B

Ok. But if I for example measure each apple with an accuracy if 1g, and then add all the weights I found together. How accurate is now my estimate of the weight of the basket?

The point is that stochastic, thus random errors will average to 0. That is why you can by taking more samples increase your accuracy, even if your instrument remains the same.

First lets be very clear that counting population and increasing sample size are two entirely different things. However, to help you understand more clearly what you are talking about...

From stats for Dummies

In statistics, the two most important ideas regarding sample size and margin of error are, first, sample size and margin of error have an inverse relationship; and second, after a point, increasing the sample size beyond what you already have gives you a diminished return because the increased accuracy will be negligible.

The relationship between margin of error and sample size is simple: As the sample size increases, the margin of error decreases. This relationship is called an inverse because the two move in opposite directions. If you think about it, it makes sense that the more information you have, the more accurate your results are going to be (in other words, the smaller your margin of error will get). (That assumes, of course, that the data were collected and handled properly.)

Suppose that the Gallup Organization’s latest poll sampled 1,000 people from the United States, and the results show that 520 people (52%) think the president is doing a good job, compared to 48% who don’t think so. First, assume you want a 95% level of confidence, so you find z* using the following table.

z*-Values for Selected (Percentage) Confidence
Levels
Percentage Confidence z*-Value
80 1.28
90 1.645
95 1.96
98 2.33
99 2.58
From the table, you find that z* = 1.96.
The number of Americans in the sample who said they approve of the president was found to be 520. This means that the sample proportion,

is 520 / 1,000 = 0.52. (The sample size, n, was 1,000.) The margin of error for this polling question is calculated in the following way:

See Dummies EQ1

According to this data, you conclude with 95% confidence that 52% of all Americans approve of the president, plus or minus 3.1%.

Using the same formula, you can look at how the margin of error changes dramatically for samples of different sizes. Suppose in the presidential approval poll that n was 500 instead of 1,000. Now the margin of error for 95% confidence is

See Dummies EQ2

which is equivalent to 4.38%. If n is increased to 1,500, the margin of error (with the same level of confidence) becomes

See Dummies EQ3

or 2.53%. Finally, when n = 2,000, the margin of error is

See Dummies EQ4

or 2.19%.

Looking at these different results, you can see that larger sample sizes decrease the margin of error, but after a certain point, you have a diminished return. Each time you survey one more person, the cost of your survey increases, and going from a sample size of, say, 1,500 to a sample size of 2,000 decreases your margin of error by only 0.34% (one third of one percent!) — from 0.0253 to 0.0219. The extra cost and trouble to get that small decrease in the margin of error may not be worthwhile. Bigger isn’t always that much better!

[SV THIRD DAY]

It's never been a Population issue...it's a Population Control Issue and the Progressives LOVE to control...they love Eugenics, they love selective breeding, they love weeding out the lesser desirables (their words not mine). The more things change...the more they stay the same and new generations get sucked into the old ideas that have been proven wrong time and time again.

[boatman61]

Originally Posted by Adelie

Snippy, ad hominem posts are about to get this thread closed.

Argue the positions not the debater.

I've not seen any threats to close the thread yet..
And I am arguing a position.. my position is that most of what's churned out is politically expedient ********..
Like diesel cars being promoted as being better for the environment 20years ago to boost profits.. building more basic engines and raising more profits from the more basic fuel which in Europe soon became more expensive than petrol..
Today the 'gods' say its bad for our health and we should scrap them and buy electric.
Stats are like polls.. the figures/questions are always shaped to get the answers you want..
Unless your a pollster or statistician.. in which case ^^^^^^^^^^ sucks..

[StuM]

Originally Posted by UAEguy

First lets be very clear that counting population and increasing sample size are two entirely different things. However, to help you understand more clearly what you are talking about...

From stats for Dummies

That says it all.

I, and I suspect a number of other people here, have actually studied applied statistics at tertiary level and used them extensively in work over many years. We do not base our knowledge on a "for dummies" book.

Let me explain briefly how a census works, even in a third world country like PNG.

1. The government spends millions of dollars every 10 years doing a complete enumeration as best they can. That involves employing many, many teams of enumerators who supposedly cover the whole country. That enumeration does not only count heads, it also records other demographics such as gender, age, no of children per parent etc, etc,

In reality, because of various factors, that probably gives a population figure which is well below the true number because places are missed, counts are inaccurate etc. Suppose that gives us a margin of error of 20%.

Now comes the important part, and this is where statistical sampling comes into effect and drastically reduces that margin of error.

A subset of locations is carefully selected to be representative of the various types of population areas and of the country as a whole and a second enumeration is done on those locations to validate the original data and determine the level of under/over counting and confirm the other demographics. Including in that second enumeration is comparison with data collected from local records such as church registers, health clinic records, local community school enrolments, discussions with local clan/tribal chiefs etc, etc.

That second "statistical sampling" can them be used to determine with a high degree of accuracy what the actual error was in the original census figures. The original census figures can then be adjusted in line with that calculated error to arrive at a fairly accurate population total.

Statistical analysis of the other demographics such as ages, especially in conjunction with the previous census data can then be used to determine annual birth and death rates with a high degree of accuracy.

Between censuses, selective sampling a representative locations can then be used to monitor changing trends in births, deaths etc and these can also be extrapolated to whole regions and the whole country, again with a high degree of accuracy.

Believe me, even those kids frolicking in the sea on a remote tropical island in a third world country do eventually get taken into account in population statistics.

[DumnMad]

Originally Posted by UAEguy

Looking at these different results, you can see that larger sample sizes decrease the margin of error, but after a certain point, you have a diminished return.

Exactly, the larger the sample size the smaller the %age error.
Thats why gathering all the worlds samples and adding them together we get a much better estimate of world population.
Even though a few of those samples(population estimates) are pretty rough (+ or -)they are a small portion of the total and don't alter the projected world estimates significantly. In this case increasing the sample size doen't cost much because the sampling/census/estimates for each country has already been paid for.

[UAEguy]

Originally Posted by StuM

That says it all.

I, and I suspect a number of other people here, have actually studied applied statistics at tertiary level and used them extensively in work over many years. We do not base our knowledge on a "for dummies" book.

Well StuM, sometimes "for Dummies" it is the only way to show someone that simply increasing the size of the sample does not get them to perfect accuracy, especially when you don't know what it is that you're measuring in the first place. The example of calculating the weight of an apple is completely irrelevant since you know you are measuring the entire weight of the apple each time.

And yes, there are many of us here with tertiary degrees besides you. Some even have a couple.

And yes, adding demographics improves the calculations. That was never debated.

The question remains, how can the UN with a known error in their ability to estimate population at +/-3% publish statements that the population is decreasing in a range of 0-2.5%? That is the question. Show us how that is done without an uncertainty greater than the result.

Please StuM...show us.

[daletournier]

Originally Posted by UAEguy

Well StuM, sometimes "for Dummies" it is the only way to show someone that simply increasing the size of the sample does not get them to perfect accuracy, especially when you don't know what it is that you're measuring in the first place. The example of calculating the weight of an apple is completely irrelevant since you know you are measuring the entire weight of the apple each time.

And yes, there are many of us here with tertiary degrees besides you. Some even have a couple.

And yes, adding demographics improves the calculations. That was never debated.

The question remains, how can the UN with a known error in their ability to estimate population at +/-3% publish statements that the population is decreasing in a range of 0-2.5%? That is the question. Show us how that is done without an uncertainty greater than the result.

Please StuM...show us.

I thought StuM was the only one that had a degree...lol, lol,lol.

Sent from my vivo Y35 using Cruisers Sailing Forum mobile app

[Jason Flare]

A company that prints phone books hires people in our area to distribute them locally door to door. Some people, instead of distributing them toss them over the banking in the woods.

Granted, the phone book company calls a few houses to find out if they actually got a phone book as a way to verify delivery, but it's not fool proof since phone books still get dumped in dumpsters and over bankings.

Human nature being what it is I doubt humans being hired to go out into the countryside and count people are any more motivated to do an accurate job than the people hired to deliver phone books.

[Jason Flare]

Count your islands first before you count the people.

How many islands in the Philippines. (Notice the word, "approximately.")


How many islands are there in the Philippines

[Jason Flare]

Maybe Greece. They have islands

How many islands in Greece?

Greece has an extremely large number of islands, with estimates ranging from somewhere around 1,200[1] to 6,000,[2] depending on the minimum size to take into account. The number of inhabited islands is variously cited as between 166[3] and 227.[2]

I like the dispute over populated Greek islands.

https://en.m.wikipedia.org/wiki/List...ands_of_Greece

Maybe they hired the same people to count the islands that they hired to deliver the phone books.

[StuM]

Originally Posted by UAEguy

The question remains, how can the UN with a known error in their ability to estimate population at +/-3% publish statements that the population is decreasing in a range of 0-2.5%? That is the question. Show us how that is done without an uncertainty greater than the result.

I've got 200 bags of wet sand each of which weigh between 20 and 40 Kg. I record the weigh of each bag to the nearest Kg.

When I add the weights together I come up with a total of 6138 Kg.

What is the margin of error on that total at a 95 and 99.7% confidence level? (You should be able to work that out from your Statistics for Dummies book. Note, you are not working here with "samples", you are working with multiple "measurements". Even though each bag's weigh is only known to +/- 2.5 to 5%, the final MoE is a LOT less than 3% (180Kg).)

After those bags have been sitting in the sun for a day, I weigh them again. Next day, I do it again. I repeat this every day for a month

At the end of the month, I plot the daily total weights and derive a best fit curve. How accurately can I determine the rate at which the roughly 600 Kg of sand have lost water through drainage and evaporation?

[IdoraKeeper]

Originally Posted by Dockhead

Indeed. And significant net population growth occurs only in populations at a particular, fairly narrowly defined state of development. Below that state, and people die too much. Above that state, and people don't have enough kids.

So significant worldwide net population growth is a short-term phenomenon unless a large part of the world gets stuck in the stage of development India is in now. China is already well out of it and has jumped directly to first world demographic problems (the stupid one child policy accelerated the development of that problem). If India and a few big countries in Africa and South America will get to the stage China has reached just in the last 10 years, then we will start having a declining world population and a whole different set of demographic problems.

And with or without that, the developed countries already have a big problem with demographics, with not enough young people to pay the taxes to support growing numbers of old people living longer and longer. And not enough people to do the work needed. The gap is currently being filled with immigration, but what happens when those countries start needing their own productive young people? Big trouble, that's what. We will have to have a big increase in the rate of productivity growth to fill the gap.

The world has some pretty big demographic challenges, but overpopulation is not one of them.

As to "consumerism" -- there's not really any growth in the consumption of stuff, in developed countries. What people consume as they get richer and richer is rather experience, like travel. Why we cruisers are the perfect examples of this. So it's not the "powers that be" and "vested economic interests" which need economic growth -- it's us, who want to work less and have richer experiences, more time to play and explore and more means to do so. We cruisers are not the only ones who want that -- it's a basic human drive.

Very true but the advent of AI directed robots, many more people will have more free time. Taxes on robot income will pay for guaranteed annual income...