Warning

 

Close

Confirm Action

Are you sure you wish to do this?

Confirm Cancel
Member Login

Posted: 12/17/2016 2:40:18 PM EST
Link Posted: 12/17/2016 2:48:21 PM EST
He'll be missed...
Link Posted: 12/17/2016 2:53:12 PM EST
Hope and Change. Only was was had.
Link Posted: 12/17/2016 2:59:09 PM EST
Link Posted: 12/17/2016 3:01:10 PM EST
Silicon Valley made out like a bandit.
Link Posted: 12/17/2016 3:01:18 PM EST
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
Link Posted: 12/17/2016 3:01:55 PM EST
Link Posted: 12/17/2016 3:04:13 PM EST
FBHO
Link Posted: 12/17/2016 3:05:04 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By BerlinVet:
He'll be missed...
View Quote

Not by anyone here.
Link Posted: 12/17/2016 3:05:34 PM EST
Holy s#%(
Link Posted: 12/17/2016 3:06:38 PM EST
Wow
Link Posted: 12/17/2016 3:08:20 PM EST
Debt is up. That's good, right?
Link Posted: 12/17/2016 3:09:27 PM EST
too bad the charts stop in 2013
Link Posted: 12/17/2016 3:10:13 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By BerlinVet:
He'll be missed...
View Quote



He'll be missed . . . . like an infectious rash that takes eight years to go away.
Link Posted: 12/17/2016 3:10:17 PM EST
Most of those charts are about four years old, so Obama would blame Bush.  The last few years look better. 
Link Posted: 12/17/2016 3:10:24 PM EST
Those fucking Russians have their hand in everything........
Link Posted: 12/17/2016 3:11:25 PM EST
Bush's fault.
Link Posted: 12/17/2016 3:12:25 PM EST
Thanks Obama!

Thanks idiot liberals!

Link Posted: 12/17/2016 3:22:47 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By P400:
Thanks Obama!

Thanks idiot liberals!

View Quote

Link Posted: 12/17/2016 3:25:40 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By BGENE:

He'll be missed . . . . like an infectious rash cancer that takes eight years to go away.
View Quote

Link Posted: 12/17/2016 3:26:10 PM EST
[Last Edit: 12/17/2016 3:35:07 PM EST by 13010311]
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote


While I agree with your underlying statement, take a look at the links below. Most go back decades, and include 2016. Paints a pretty grim picture...

1:) Student Loans

2:) Food Stamps

3:) Federal Debt

4:) Money Printing

5:) Couldn't find the data...will update when I do. That said: Obamacare premiums to soar 22% in 2017

6:) Labor Force Participation

7:) Worker's Share of the Economy

8:) Median Family Income

9:) Home Ownership
Link Posted: 12/17/2016 3:39:38 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote

Thank you for pointing that out. I had missed it and it'll be useful in other areas.
Link Posted: 12/17/2016 3:45:03 PM EST
Thomas Payne didn't have anything good to say about Obama
Link Posted: 12/17/2016 3:53:00 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote


Exactly. I can't show this to any of my science colleagues, because they are not all indicating the same timeframe. Y-axis issues I can live with, as long as we're looking at equal/preceding timescales to interpret a trend. They need to fix their charting so that the real debate can happen around the validity of the information.
Link Posted: 12/17/2016 4:01:58 PM EST
[Last Edit: 12/17/2016 4:02:53 PM EST by Silverslider]
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote


Blah blah blah. Statistics
You can normalize everything at zero or parse significant pieces
Sure if you start at zero the graph will look less significant.
Simple conclusions to draw.
Everything trends downward
I don't have the data and I am sure all downward trends are statistically significant
If you had the data and could multiply the percentages by the actual number you would see loss in millions
Or
Excuse me, trillions
I forgot some trends do go up. The ones that sucks us dry. Food stamps, debt, etc
Sigma!
Link Posted: 12/17/2016 4:04:20 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By 13010311:


While I agree with your underlying statement, take a look at the links below. Most go back decades, and include 2016. Paints a pretty grim picture...
View Quote



oh, i completely agree that the data the charts represent is an accurate state of affairs--i'm not disputing that at all.

it's the presentation that i'm concerned with. to someone with a quantitative background, it looks like a deliberate distortion.
Link Posted: 12/17/2016 4:08:07 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By Silverslider:

Sure if you start at zero the graph will look less significant.
View Quote


exactly--if the axis starts at zero, it will accurately reflect the difference.

distorting the truth in order to score political points is liberal behavior.
Link Posted: 12/17/2016 4:53:48 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By ScottyS:

Exactly. I can't show this to any of my science colleagues, because they are not all indicating the same timeframe. Y-axis issues I can live with, as long as we're looking at equal/preceding timescales to interpret a trend. They need to fix their charting so that the real debate can happen around the validity of the information.
View Quote View All Quotes
View All Quotes
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By ScottyS:
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.

Exactly. I can't show this to any of my science colleagues, because they are not all indicating the same timeframe. Y-axis issues I can live with, as long as we're looking at equal/preceding timescales to interpret a trend. They need to fix their charting so that the real debate can happen around the validity of the information.

This. The X axes are all over the place and printed so small that it's a bit of effort to figure out what they represent. If you stack several charts the implication is that they follow the same rules and thus can be compared visually.
Link Posted: 12/17/2016 4:59:45 PM EST
Discussion ForumsJump to Quoted PostQuote History

some of those were going to shit before he took over. It's not that he was or wasn't responsible for them, it's the fact he did jack or shit about actually fixing it
Link Posted: 12/17/2016 5:08:18 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote

Here's the full adjusted monetary base chart. 

Looks good 

Link Posted: 12/17/2016 5:21:32 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By ContrarianIndicator:

Here's the full adjusted monetary base chart. 

Looks good 

http://getgold999.com/wp-content/uploads/2015/01/2015-01-21_0846-StLouis-adjusted-monetary-base.png
View Quote



as i wrote, some of the charts are fine. but putting a 0-based y axis graph next to an otherwise identically-formatted non-0-based y axis graph makes the viz even more deceptive. makes it look like a deliberate misrepresentation rather than an honest mistake.

so if anyone gets called out on these figures, they will appear to be either deceptive or incompetent. that's a bad place to be in an argument. normalize that shit like a good statistician, and all these problems go away.
Link Posted: 12/17/2016 5:25:25 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By sirensong:
the charts are powerful, but beware--they violate one of the most fundamental principles of statistics data viz. as a general principle, any time the series on the y axis does not start at 0, the chart is deceptive because it magnifies the visual weight of the difference. people parse charts by comparing visual ratios, not by carefully examining the axis details. for example, chart 1 (student loans) is very straightforward and robust. chart 7 (workers' share oft he economy) is very deceptive, since the graphic appears to represent a change from almost 100% to almost 0%. however, inspecting the y-axis shows it to be a change of only ~6%.

some of those charts are straight out of the "how to lie with statistics" playbook. they're not wrong--they're just distorted in order to make the change look bigger than it actually is. if you put them in front of someone who deals with quantitative data, you're probably going to get called out on it.
View Quote

Yep +1
Link Posted: 12/17/2016 5:31:27 PM EST
bu the told us he reduced the annual deficit faster then any other president.
Link Posted: 12/17/2016 5:32:35 PM EST
Discussion ForumsJump to Quoted PostQuote History
Originally Posted By ContrarianIndicator:

Here's the full adjusted monetary base chart. 

Looks good 

http://getgold999.com/wp-content/uploads/2015/01/2015-01-21_0846-StLouis-adjusted-monetary-base.png
View Quote


I'm not a gold bug by any stretch of the imagination. But if the monetary base as been inflated that much, how is gold not at $5,000/ounce?
Top Top