Confused about statistics?

imoldernu

Gone but not forgotten
Joined
Jul 18, 2012
Messages
6,335
Location
Peru
Yes? No?
While we may speak the same language, being on the same framework of understanding gets confusing. I'd like to start a discussion on this by looking at what I thought was a simple question about my own life expectancy.

How long am I expected to live, at my current age (82)? While I understand that life expectancy is based on many factors, I thought there might be some "average" numbers. Not so:

Mean refers to what we normally call “average,”
Median refers to the middle data point in our set, and
Mode refers to the most common value.

from here: https://obliviousinvestor.com/mean-vs-median-life-expectancy-for-retirement-planning/

But WAIT!:
That's just the very beginning.

So what's wrong with the question "What is the average Net Worth for retirees in the United States?"

Egad! "You pays your money and takes your choice... " ... but that's only the beginning....:(

At what age?
At age of retirement, at the average age, or across the board for all retirees?
For the individual, or for the household.
Male, female or both?
As of what year of statistics? This year? Census year? other?
Government statistics?
Includes actual dollars or projected pension/SS income too?
Current estimates? or based on projected or imputed values.

etc.

But of course that is only one question... Practically speaking, your own personal interest looks to finding a comparison to yourself. This requires even more definition.

How old you are now
Your health history
Your sex
The socio/economic area that you live in and where you plan to live
... and of course, your current status of income and outgo factors and the hundreds of other factors that have to do with things that make you different, like how much attention you spend to keep healthy... medical care, personal healthcare regimen. Then there's genealogy, risk factors, and attitude.
.....................................................................................................

Overboard... yes... agreed! But that's just one question, and one that can be found on every... every retirement website. Not looking for an answer to this but looking for commiseration on the confusion.

To toss out the kind statement that personifies the inherent statistical confusion, and the kind of imputed positive results to be expected.... My own statement.

"My investment in IBonds has always provided me with more than 5% interest earnings, and will continue to do so in the future. Furthermore
future increases in the CPI will be reflected in higher earnings."

How do you handle statistical confusion?
 
... How do you handle statistical confusion?
I am not confused. I ignore it completely. I am a sample space of one and small sample spaces cannot be dealt with statistically, even if the "statistical" articles were not click-bait written primarily by ignorant dolts.
 
I ignore statistics, so I am never confused by them.

OK, that's not quite true. I had a career where I had to understand maximum likelihood methods and even helped teach statistics. Thus, I learned the limitations and have not been intimidated by statistics.

Many people put too much weight on statistics and patterns. Knowing that helps one take advantage of them.
 
Last edited:
I don't understand why the article qualifies as "statistical confusion". I see it as enlightenment. I, for one, have never really thought about the difference between mean and median life expectancy, and why the latter is higher, and I learned something by reading the article.
 
I enjoy reading the stories that use statistics. But if you assume "there are three kinds of lies: lies, damned lies, and statistics," it's easy enough to assign the story a bias, and take it into account. It's also fun to read different stories that use the same statistics to "prove" contradictory opinions/theories. The truth is NOT out there.
 
This thread reminds me of REWahoo's sig line.
 
So, anyway... Here's an example... about life expectancy...

Life Expectancy Male
When I was born: 59
According to SS 2018 @ age 82: 89
Average male US today: 79
at this website https://livingto100.com/calculator/age
the estimate was age 97...

:2funny:

yeah... "numbers is hard"
 
Last edited:
Statistics mean very little to me. Really, how many people die in car accidents, from cancer, from any health issue...do the statistics take all this into account or do they just take the entire population and determine who dies and at what age?

And those are very nice I-bond returns, we have @ 100K in them from 2003 and on. I don't count them for FIRE because they are back up case of disaster. But, if you count net worth at time of retirement, they would be included. So, I think of it that way. Take our net worth at this point in time and extend it out 30 years, we are 60. So, when and if we get to 80, I would re evaluate net worth at that time. And I do this every year anyway to keep track of where we are. The graph of $$ at certain points in time can be way up or way down, or plugging along @ 5% which i'm happy with.
 
So, anyway... Here's an example... about life expectancy...

Life Expectancy Male
When I was born: 59
According to SS 2018 @ age 82: 89
Average male US today: 79
at this website https://livingto100.com/calculator/age
the estimate was age 97...

:2funny:

yeah... "numbers is hard"


Just a note on that website. After you spend 5 minutes giving all sorts of detailed information, it asks for your email so they can send you your score plus health information that will allow you to live to age 100.
 
I laugh and cry.
 
Am on a roll..

Here's something on looking back:
When I first came to ER a few years ago, I felt like a pauper... Even though I retired @53, my wages at that time were so far from those of the"avg" ER poster, that the thoughts of having a retirement cushion of a 1M+ was not conceivable. In fact it turned out to be unimportant.

Fun with numbers from CPI.
https://bls.gov/data/inflation_calculator.htm

So a little history of my own salary compared to the US median wage progression through the years, using the CPI.

First married 1958.. wages $6200, today $58,000 then median $3700
Promotion 1966... wages $15,000, today $119,000 then median $6900
Final, retired 1984... wages $48,000, today $119,000 then median $16,000

Nothing else to do today, so playing with numbers. Already retired? Play with the CPI site.
 
Last edited:
My daddy taught me "Figures don't lie, but liars can figure." That has stuck with me and I recall it every time someone uses numbers to try and prove their point. Context and reliable sources are important items often left out.
 
I'm a retired actuary so yes, I notice the use and abuse of statistics. I ignore those that are so vague they're meaningless ("The average household credit card debt is $$,000", without specifying whether it's over all households or only those that have debt). If it's someone on FB mindlessly reposting something that fits their worldview, I may call them in it if I can find reliable figures to refute them. One posted the salaries of the top 8 health insurance CEOs and said that was the reason for high health insurance costs. Some simple math including the population of the US and the estimate of those companies' market share showed that the 8 compensation packages divided by the likely number of insureds they covered was about $5/year per insured.

I saw another purporting to show what you'd have gotten from SS if you'd averaged $30K/year over a 40-year career which then did a calculation based on $30K each year using current contribution %s applied to $30K/year and accumulated at interest (few people made $30K 40 years ago, the SS base was only $17,700 in 1978 and the contribution percentages were not always at current levels).

There's a lot of bad math out there.
 
The problem comes in when a distribution is not normal... life expectancy is not a normal curve.... that should be known by people who are looking at statistics so you do not confuse anything...


A perfect example is wealth... take the top 7 billionaires in the US... (3 of the top 10 are foreigners)... total assets of $535.5 billion...


Now, just these people take the average for ALL US people to $164... that might not like a lot, but this is from 7 people vs 326 million...


BTW, I still think it holds as this is an old stat I heard many years ago... that those 7 will have more net worth than the bottom half of the US population... IOW, the bottom 163 million people do not have net assets over that $500 billion....
 
So, anyway... Here's an example... about life expectancy...

Life Expectancy Male
When I was born: 59
According to SS 2018 @ age 82: 89
Average male US today: 79
at this website https://livingto100.com/calculator/age
the estimate was age 97...

:2funny:

yeah... "numbers is hard"

I'm not sure if you are actually confused by this, or just having some fun.

It takes a bit of thought, but clearly as you get older, your statistical life expectancy increases (you were not one of the 'statistics').

At birth, the statistical LE includes those who succumb to childhood diseases. But if you reach 80, those are no longer in the calculation. It reminds me of a comedian I heard many years ago:
"My dear, dear Grandmother turned 104 today. I was so worried that she wouldn't be around for another year to celebrate her 105th birthday. But then I read, very few people die between the ages of 104 and 105, so now I feel better!" :LOL:


...

Lots of readers and unfortunately some writers as well are virtually innumerate.

Seems to me that most writers are innumerate. Oh, the things I read, drives me nuts!

-ERD50
 
Back
Top Bottom