Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

caraher

(6,359 posts)
13. Interesting piece, but it understates the badness of the Harvard conclusions
Sun Apr 21, 2013, 02:08 PM
Apr 2013
By failing to include certain spreadsheet cells in its calculations, the study by Harvard economists Carmen Reinhart and Kenneth Rogoff may have overstated the impact that debt burdens have on a nation’s economic growth.


Worse - it claimed an impact where none exists, which is very different from merely overestimating a real effect.

Panko's paper is interesting... it looks like the best way to reduce spreadsheet errors is by having lots of people look at a given spreadsheet. Which is what ultimately uncovered the error in R&R.

Of course, Dean Baker chided them as unprofessional in 2010 when they're paper came out for not being transparent with their data. I'll never know for certain, but I wouldn't be shocked if that were by design. And (he says in tinfoil hat mode) maybe they only released it to a lowly grad student because they didn't think he'd find out the problem...

But I don't really believe that; more likely R&R feel prey to the biggest hazard of research, fooling yourself. When a result contradicts your expectations the instinct is to dig deeper, but when it confirms them there's a strong tendency to just accept them uncritically. It's the stuff that bolsters their own beliefs that truth-seekers need to place under the most careful scrutiny!

Recommendations

0 members have recommended this reply (displayed in chronological order):

Never should have been alive to begin with. yourout Apr 2013 #1
Hadn't Thought About it in Relationship to SS On the Road Apr 2013 #2
Right - take it further FogerRox Apr 2013 #16
Japan Tried Something Similar in the 90s On the Road Apr 2013 #22
Japan then and the US now, very different FogerRox Apr 2013 #27
Recommend. Triana Apr 2013 #3
it won't, he's totally invested in the charade... KG Apr 2013 #4
Exactly. AnotherMcIntosh Apr 2013 #6
As a side note to this, a recent study found that 88% of spreadsheets have errors: TheManInTheMac Apr 2013 #5
Unfortunately, they likely came up with that 88% figure on a spreadsheet (n/t) thesquanderer Apr 2013 #7
I was thinking that too, along with "78% of statistics are made up." TheManInTheMac Apr 2013 #8
Wow! Who knew? avaistheone1 Apr 2013 #9
Interesting piece, but it understates the badness of the Harvard conclusions caraher Apr 2013 #13
Mine don't underpants Apr 2013 #15
Think again. . . . h2ebits Apr 2013 #10
I have been pointing out for a long time that Timothy Geithner was a stealth JDPriestly Apr 2013 #11
I don't think Obama was deceived at all. Lasher Apr 2013 #17
I did not say that Obama was deceived, did I? JDPriestly Apr 2013 #24
I didn't realize Geithner was a protege of Pete Peterson. avaistheone1 Apr 2013 #19
Did you read the article at Nation of Change? JDPriestly Apr 2013 #23
"I first learned of the Administration’s plans to cut Social Security in a deep-background briefing" DJ13 Apr 2013 #12
A Famous Cyborg allinthegame Apr 2013 #14
K&R abelenkpe Apr 2013 #18
So what does that do to this chart? Lasher Apr 2013 #20
Sane minds and people over 62 will KILL Obama's offer to Rethugs Dan Ken Apr 2013 #21
Obama would just come up with some other phony justification to cut it. forestpath Apr 2013 #25
Great piece we should all read but I think Eskow is being a little too charitable. pa28 Apr 2013 #26
Latest Discussions»General Discussion»why the 'spread sheet sca...»Reply #13