Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the...
Transcript of Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the...
![Page 1: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/1.jpg)
communicating uncertainty
Charlotte BoysJanuary 30, 2020
MSc Scientific Computing, University of Heidelberg
![Page 2: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/2.jpg)
overview
1. Introduction
2. How to Lie with Uncertainty
3. How to Communicate Uncertainty
4. Case studies
5. Conclusion
1
![Page 3: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/3.jpg)
introduction
![Page 4: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/4.jpg)
sources of uncertainty (spiegelhalter)
Aleatory UncertaintyInevitable unpredictability of the future due to unforeseeablefactors, fully expressed by classical probabilities.
2
![Page 5: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/5.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 6: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/6.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 7: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/7.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 8: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/8.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 9: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/9.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 10: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/10.jpg)
sources of uncertainty (spiegelhalter)
Epistemic UncertaintyModelled and quantified uncertainty about the structure andparameters of statistical models, expressed, for example,through Bayesian probability distributions or sensitivityanalyses.
Ontological UncertaintyUncertainty about the ability of the modelling process todescribe reality, which can only be expressed as a qualitativeand subjective assessment of the model, conveying withhumility the limitations of our knowledge.
3
![Page 11: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/11.jpg)
how to lie with uncertainty
![Page 12: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/12.jpg)
tobacco industry
• Brown and Williamson Tobacco company internal memo,1969:
“Doubt is our product, since it is the best means ofcompeting with the ‘body of fact’ that exists in themindof the general public. It is also a means of establishingcontroversy.”
4
![Page 13: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/13.jpg)
fossil fuel industry
American Petroleum Institute, 1998
5
![Page 14: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/14.jpg)
wilful misunderstanding?
Daily Mail, March 16. 2013
6
![Page 15: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/15.jpg)
wilful misunderstanding?
Daily Mail, March 16. 2013
7
![Page 16: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/16.jpg)
wilful misunderstanding?
Skeptical Science, April 17. 2013
8
![Page 17: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/17.jpg)
wilful misunderstanding?
Kevin Pluck, December 17. 2019
9
![Page 18: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/18.jpg)
unemployment headlines
BBC, January 24. 2018
10
![Page 19: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/19.jpg)
behind the headlines
ONS, UK Labour Market January 2018
11
![Page 20: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/20.jpg)
how to communicate uncertainty
![Page 21: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/21.jpg)
how to communicate uncertainty
in theory
12
![Page 22: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/22.jpg)
Our aim when communicating risk anduncertainty is to inform decision-making.
12
![Page 23: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/23.jpg)
leiss’s phases of risk communication
Three Phases in the Evolution of Risk Communication Practice, 199613
![Page 24: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/24.jpg)
not-so rational decision makers
Context plays a key role in how we understand risk anduncertainty as a lay audience. The following (Slovic 2000,Ropeik 2010) have been termed fear factors:
• Uncontrollable, novel, or not understood• Having catastrophic potential, or dreadful consequencessuch as fatality
• Bearing an inequitable distribution of risks and benefits• Delayed in their manifestation of harm
14
![Page 25: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/25.jpg)
not-so rational decision makers
Context plays a key role in how we understand risk anduncertainty as a lay audience. The following (Slovic 2000,Ropeik 2010) have been termed fear factors:
• Uncontrollable, novel, or not understood
• Having catastrophic potential, or dreadful consequencessuch as fatality
• Bearing an inequitable distribution of risks and benefits• Delayed in their manifestation of harm
14
![Page 26: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/26.jpg)
not-so rational decision makers
Context plays a key role in how we understand risk anduncertainty as a lay audience. The following (Slovic 2000,Ropeik 2010) have been termed fear factors:
• Uncontrollable, novel, or not understood• Having catastrophic potential, or dreadful consequencessuch as fatality
• Bearing an inequitable distribution of risks and benefits• Delayed in their manifestation of harm
14
![Page 27: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/27.jpg)
not-so rational decision makers
Context plays a key role in how we understand risk anduncertainty as a lay audience. The following (Slovic 2000,Ropeik 2010) have been termed fear factors:
• Uncontrollable, novel, or not understood• Having catastrophic potential, or dreadful consequencessuch as fatality
• Bearing an inequitable distribution of risks and benefits
• Delayed in their manifestation of harm
14
![Page 28: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/28.jpg)
not-so rational decision makers
Context plays a key role in how we understand risk anduncertainty as a lay audience. The following (Slovic 2000,Ropeik 2010) have been termed fear factors:
• Uncontrollable, novel, or not understood• Having catastrophic potential, or dreadful consequencessuch as fatality
• Bearing an inequitable distribution of risks and benefits• Delayed in their manifestation of harm
14
![Page 29: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/29.jpg)
polarisation and risk perception
Climate-science communication and the measurement problem (Kahan, 2015)Perception of risk for polarised and non-polarised issues. N=1800
15
![Page 30: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/30.jpg)
So what can we do?
15
![Page 31: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/31.jpg)
new aims for uncertainty communication
Onara O‘Neill,Philosopher
The key to trust is not moretransparency, but IntelligentTransparency. Information shouldbe:
• accessible• comprehensible• usable• assessable
16
![Page 32: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/32.jpg)
new aims for uncertainty communication
Onara O‘Neill,Philosopher
The key to trust is not moretransparency, but IntelligentTransparency. Information shouldbe:
• accessible
• comprehensible• usable• assessable
16
![Page 33: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/33.jpg)
new aims for uncertainty communication
Onara O‘Neill,Philosopher
The key to trust is not moretransparency, but IntelligentTransparency. Information shouldbe:
• accessible• comprehensible
• usable• assessable
16
![Page 34: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/34.jpg)
new aims for uncertainty communication
Onara O‘Neill,Philosopher
The key to trust is not moretransparency, but IntelligentTransparency. Information shouldbe:
• accessible• comprehensible• usable
• assessable
16
![Page 35: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/35.jpg)
new aims for uncertainty communication
Onara O‘Neill,Philosopher
The key to trust is not moretransparency, but IntelligentTransparency. Information shouldbe:
• accessible• comprehensible• usable• assessable
16
![Page 36: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/36.jpg)
how to communicate uncertainty
lessons from risk communication
17
![Page 37: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/37.jpg)
the problem with words
• Wide variability in interpretation, even within groups(Willems et al., N = 881)
• Asymmetry in interpretation• No reliable ‘translation’ between verbal phrases andnumerical values representing probabilities
18
![Page 38: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/38.jpg)
the problem with words
• Wide variability in interpretation, even within groups(Willems et al., N = 881)
• Asymmetry in interpretation
• No reliable ‘translation’ between verbal phrases andnumerical values representing probabilities
18
![Page 39: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/39.jpg)
the problem with words
• Wide variability in interpretation, even within groups(Willems et al., N = 881)
• Asymmetry in interpretation• No reliable ‘translation’ between verbal phrases andnumerical values representing probabilities
18
![Page 40: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/40.jpg)
the problem with words
• What percentage of people taking a drug can we expect toexperience a ’common’ side effect?
• Mean estimate: 34% (N = 120)• Pharmacological definition: 1− 10% of patients• Recommended: ”Common: may affect up to 1 in 10 people”
19
![Page 41: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/41.jpg)
the problem with words
• What percentage of people taking a drug can we expect toexperience a ’common’ side effect?
• Mean estimate: 34% (N = 120)
• Pharmacological definition: 1− 10% of patients• Recommended: ”Common: may affect up to 1 in 10 people”
19
![Page 42: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/42.jpg)
the problem with words
• What percentage of people taking a drug can we expect toexperience a ’common’ side effect?
• Mean estimate: 34% (N = 120)• Pharmacological definition: 1− 10% of patients
• Recommended: ”Common: may affect up to 1 in 10 people”
19
![Page 43: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/43.jpg)
the problem with words
• What percentage of people taking a drug can we expect toexperience a ’common’ side effect?
• Mean estimate: 34% (N = 120)• Pharmacological definition: 1− 10% of patients• Recommended: ”Common: may affect up to 1 in 10 people”
19
![Page 44: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/44.jpg)
comparing risks
• A survey from Galesic & Garcia-Retamero (2010) asked thefollowing question:
Which of the following numbers represents the biggest risk ofgetting a disease? 1 in 100, 1 in 1,000, or 1 in 10?
→ 72% of 1,000 respondents in the United States and 75% of1,000 respondents in Germany answered correctly.
→ Keep the denominator fixed when making comparisons!
20
![Page 45: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/45.jpg)
comparing risks
• A survey from Galesic & Garcia-Retamero (2010) asked thefollowing question:
Which of the following numbers represents the biggest risk ofgetting a disease? 1 in 100, 1 in 1,000, or 1 in 10?
→ 72% of 1,000 respondents in the United States and 75% of1,000 respondents in Germany answered correctly.
→ Keep the denominator fixed when making comparisons!
20
![Page 46: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/46.jpg)
comparing risks
• A survey from Galesic & Garcia-Retamero (2010) asked thefollowing question:
Which of the following numbers represents the biggest risk ofgetting a disease? 1 in 100, 1 in 1,000, or 1 in 10?
→ 72% of 1,000 respondents in the United States and 75% of1,000 respondents in Germany answered correctly.
→ Keep the denominator fixed when making comparisons!
20
![Page 47: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/47.jpg)
understanding the reference class
BBC Weather
21
![Page 48: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/48.jpg)
relative risk vs. absolute risk
BBC, 26. October 201522
![Page 49: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/49.jpg)
relative risk vs. absolute risk
Spiegelhalter, 2017
• Of 100 people who don’teat bacon, 6 can beexpected to develop bowelcancer.
• Of 100 people who eatbacon every day of theirlives, 7 can be expected todevelop bowel cancer.
23
![Page 50: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/50.jpg)
relative risk vs. absolute risk
Spiegelhalter, 2017
• Of 100 people who don’teat bacon, 6 can beexpected to develop bowelcancer.
• Of 100 people who eatbacon every day of theirlives, 7 can be expected todevelop bowel cancer.
23
![Page 51: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/51.jpg)
relative risk vs. absolute risk
Spiegelhalter, 2017
• Of 100 people who don’teat bacon, 6 can beexpected to develop bowelcancer.
• Of 100 people who eatbacon every day of theirlives, 7 can be expected todevelop bowel cancer.
23
![Page 52: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/52.jpg)
relative risk vs. absolute risk
Spiegelhalter, 2017
• Of 100 people who don’teat bacon, 6 can beexpected to develop bowelcancer.
• Of 100 people who eatbacon every day of theirlives, 7 can be expected todevelop bowel cancer.
23
![Page 53: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/53.jpg)
positive and negative framing
• “10% of patients get a blistering rash”
→ 1.82 on a risk scale of 1 - 5• “90% of patients do not get a blistering rash”
→ 1.43 on a risk scale of 1 - 5
• “10% of patients experience a blistering rash, and 90% donot”
24
![Page 54: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/54.jpg)
positive and negative framing
• “10% of patients get a blistering rash”
→ 1.82 on a risk scale of 1 - 5
• “90% of patients do not get a blistering rash”
→ 1.43 on a risk scale of 1 - 5
• “10% of patients experience a blistering rash, and 90% donot”
24
![Page 55: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/55.jpg)
positive and negative framing
• “10% of patients get a blistering rash”→ 1.82 on a risk scale of 1 - 5
• “90% of patients do not get a blistering rash”→ 1.43 on a risk scale of 1 - 5
• “10% of patients experience a blistering rash, and 90% donot”
24
![Page 56: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/56.jpg)
positive and negative framing
• “10% of patients get a blistering rash”→ 1.82 on a risk scale of 1 - 5
• “90% of patients do not get a blistering rash”→ 1.43 on a risk scale of 1 - 5
• “10% of patients experience a blistering rash, and 90% donot”
24
![Page 57: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/57.jpg)
case studies
![Page 58: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/58.jpg)
case studies
climate and weather forecasting
25
![Page 59: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/59.jpg)
ensemble models
Wikipedia: Representative Concentration Pathway
26
![Page 60: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/60.jpg)
communicating uncertainty from ems
The following communication priorities need to be balanced:
• richness
→ amount of information communicated
• robustness
→ the extent to which the trustworthiness of the model iscommunicated
• saliency
→ interpretability and usefulness for the target audience
27
![Page 61: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/61.jpg)
communicating uncertainty from ems
The following communication priorities need to be balanced:
• richness→ amount of information communicated
• robustness
→ the extent to which the trustworthiness of the model iscommunicated
• saliency
→ interpretability and usefulness for the target audience
27
![Page 62: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/62.jpg)
communicating uncertainty from ems
The following communication priorities need to be balanced:
• richness→ amount of information communicated
• robustness→ the extent to which the trustworthiness of the model is
communicated
• saliency
→ interpretability and usefulness for the target audience
27
![Page 63: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/63.jpg)
communicating uncertainty from ems
The following communication priorities need to be balanced:
• richness→ amount of information communicated
• robustness→ the extent to which the trustworthiness of the model is
communicated• saliency
→ interpretability and usefulness for the target audience
27
![Page 64: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/64.jpg)
cone of uncertainty
National Hurricane Center, US28
![Page 65: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/65.jpg)
sharpiegate
The New Yorker, September 6. 2019
29
![Page 66: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/66.jpg)
spaghetti plots
New York Times, September 5. 2017
30
![Page 67: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/67.jpg)
heatmaps and spaghetti plots
Hugo Bowne-Anderson, September 18. 2017
31
![Page 68: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/68.jpg)
case studies
coronavirus
32
![Page 69: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/69.jpg)
model uncertainty
MRC Centre for Global Infectious Disease Analysis
33
![Page 70: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/70.jpg)
communicating uncertainty
34
![Page 71: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/71.jpg)
communicating uncertainty
35
![Page 72: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/72.jpg)
communicating uncertainty
36
![Page 73: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/73.jpg)
communicating uncertainty
37
![Page 74: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/74.jpg)
communicating uncertainty
NY Times, January 23. 2020
38
![Page 75: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/75.jpg)
conclusion
![Page 76: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/76.jpg)
final remarks
Information about uncertainty should:
• Be accessible, comprehensible, usable, and assessable.• Carefully consider audience, context and framing.• Use combinations of words, numbers and visuals tominimise misunderstanding.
• Be communicated with humility about the extent of ourknowledge; demonstrating trustworthiness, rather thandemanding trust.
39
![Page 77: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/77.jpg)
final remarks
Information about uncertainty should:
• Be accessible, comprehensible, usable, and assessable.
• Carefully consider audience, context and framing.• Use combinations of words, numbers and visuals tominimise misunderstanding.
• Be communicated with humility about the extent of ourknowledge; demonstrating trustworthiness, rather thandemanding trust.
39
![Page 78: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/78.jpg)
final remarks
Information about uncertainty should:
• Be accessible, comprehensible, usable, and assessable.• Carefully consider audience, context and framing.
• Use combinations of words, numbers and visuals tominimise misunderstanding.
• Be communicated with humility about the extent of ourknowledge; demonstrating trustworthiness, rather thandemanding trust.
39
![Page 79: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/79.jpg)
final remarks
Information about uncertainty should:
• Be accessible, comprehensible, usable, and assessable.• Carefully consider audience, context and framing.• Use combinations of words, numbers and visuals tominimise misunderstanding.
• Be communicated with humility about the extent of ourknowledge; demonstrating trustworthiness, rather thandemanding trust.
39
![Page 80: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/80.jpg)
final remarks
Information about uncertainty should:
• Be accessible, comprehensible, usable, and assessable.• Carefully consider audience, context and framing.• Use combinations of words, numbers and visuals tominimise misunderstanding.
• Be communicated with humility about the extent of ourknowledge; demonstrating trustworthiness, rather thandemanding trust.
39
![Page 81: Communicating UncertaintyEpistemic Uncertainty Modelled and quantified uncertainty about the structure and parameters of statistical models, expressed, for example, through Bayesian](https://reader031.fdocuments.in/reader031/viewer/2022011918/5fffeae0e4c46220626fa162/html5/thumbnails/81.jpg)
selected references
Elisabeth M Stephens, Tamsin L Edwards, and DavidDemeritt, Communicating probabilistic information fromclimate model ensembles—lessons from numerical weatherprediction, Wiley interdisciplinary reviews: climate change3 (2012), no. 5, 409–426.David Spiegelhalter, Risk and uncertainty communication,Annual Review of Statistics and Its Application 4 (2017),31–60.Sanne JW Willems, Casper J Albers, and Ionica Smeets,Variability in the interpretation of dutch probabilityphrases-a risk for miscommunication, arXiv preprintarXiv:1901.09686 (2019).
40