Sinclair Davidson Brussels December 7 2017 A bleak perspective But anyone who after the twentieth century still thinks that thoroughgoing socialism nationalism imperialism mobilization central planning regulation zoning price controls tax policy labor unions business cartels governme ID: 754583
Download Presentation The PPT/PDF document "— How to torture data to justify publi..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
—How to torture data to justify public policy
Sinclair DavidsonBrussels December 7, 2017Slide2
A bleak perspective
But anyone who after the twentieth century still thinks that thoroughgoing socialism, nationalism, imperialism, mobilization, central planning, regulation, zoning, price controls, tax policy, labor unions, business cartels, government spending, intrusive policing, adventurism in foreign policy, faith in entangling religion and politics, or most of the thoroughgoing nineteenth-century proposals for government action are still neat, harmless ideas for improving our lives is not paying attention.
Deirdre
McCloskey Slide3
My thoughts
Government has lost its way. Since abandoning its role as an impartial player in society it has taken to abusing the trust we invest in it. The corresponding abuse of evidence-based policy should alarm everyone.Ask yourself – how appropriate is it for a government and its agencies and its financially supported NGOs to ‘
denormalise
’ a segment of its population? Slide4
The Australian government has a track record in dubious
public policy research:FuelWatch (omitted variable bias)
Stimulusgate
(data snooping)
Mining
tax (incorrect interpretation of empirical results)
Carbon
tax (fudging results and definition of costs)
…
Plain packaging.Slide5
How I came to be involved
Initially argued against policy on IP grounds.Then journal editor commissioned a paper.Then watched hysterical response and realised a cover-up was in progress. Slide6
Highlights of plain packaging debate
Household consumption dataThe 3.4% lieCancer Council surveysConflict of interestWho commissioned data (was proper tender process followed?)
Methodological problems
Lack of diagnostics (100% miss on a hit-miss table)
Timing problems
Peer review?
Were contractual obligations met?
Cancer Council Press release and Carly Weeks smear
Post-implementation Review
Data snooping around trend lines
Unusual base case
0.55% result smaller than 0.6% sample error.
Price not included in analysis. Slide7
Lessons
Read the original source literature yourself.Pay attention to methodPay attention to tables Examine the data yourself.Try to replicate the results
What results have not been reported?
Perform robustness tests
Look for Type III errors
.
The correct answer to a different question
Ignore propaganda.
Check facts.
Especially check ‘facts’ that you ‘know’ to be true
Use Freedom of Information laws to access data.
Prepare to be demonised.
“It is dangerous to be right when the government is wrong”
Government relies on statistical ignorance of the population and cynicism towards statistics