THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

48
THE RISKS OF LIFE THE RISKS OF LIFE Video Conference 5 Media & Misinformation

Transcript of THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Page 1: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

THE RISKS OF THE RISKS OF LIFELIFEVideo Conference 5

Media & Misinformation

Page 2: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 3: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Newspaper HeadlinesNewspaper HeadlinesDaily Mail

“Obesity worse for cancer than smoking”“Is anything safe to eat? Cancer report adds

bacon, ham and drink to danger list”

Guardian“Stay trim and stop eating bacon, cancer report

declares”

The Sun“Bacon butty cancer risk”

“Careless pork costs lives”“Shock Cancer Alert. Save Our Bacon. Storm

as Butties Branded Killers!”

Page 4: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Butties Branded Killers!Butties Branded Killers!

OCTOBER 2007

Page 5: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

The reason behind the The reason behind the headlinesheadlines

World Cancer Research Fund (WCRF) issued a report on the

influences of nutrition and physical activity on Cancer.

They gave 10 recommendations for Cancer prevention

Page 6: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

WCRF Recommendation WCRF Recommendation

“Avoid processed meats such as bacon, ham, salami, corned beef

and some sausages”

Page 7: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

The EvidenceThe EvidenceThe “Convincing Evidence” states

that there is a 20% increased risk of bowel cancer per 50g

of daily processed meat

Page 8: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

What is this really saying?What is this really saying?

Report20% increased risk of bowel cancer

per 50g daily processed meat

Fact Approximately 5 out of every 100 people

will get bowel cancer during their life.

ConclusionEating a huge bacon sandwich every day

increases the number of people getting bowelcancer to 6 out of every 100.

Page 9: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 10: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Simpson’s ParadoxSimpson’s ParadoxYou are suffering from a terrible illness.

There are two drugs you can take to cure this illness:(1) Drug A(2) Drug B

Both drugs are put on trial

Page 11: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Day 2

Total

Page 12: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

Day 2

Total

Page 13: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Day 2

Total

Page 14: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

Day 2

Total

Page 15: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Total

Page 16: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

Total

Page 17: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Total

Page 18: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

Total

Page 19: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Page 20: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Which is the better drug?

Page 21: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Cures 67 out of 100 people

Which is the better drug?

Page 22: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Cures 67 out of 100 people

67%

Which is the better drug?

Page 23: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Cures 67 out of 100 people

67%

Cures 53 out of 100 people

Which is the better drug?

Page 24: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A Drug B

Day 1

Cures 63 out of 90 people

70%

Cures 8 out of 10 people

80%

Day 2

Cures 4 out of 10 people

40%

Cures 45 out of 90 people

50%

Total

Cures 67 out of 100 people

67%

Cures 53 out of 100 people

53%

Which is the better drug?

Page 25: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Drug A is Better!Drug A is Better!

You will reach the incorrect solution if you only consider the daily percentage results.

You are not comparing ‘like with like’.

Sports statistics, exam results are often reported in this way and can be misleading

if the data isn’t interpreted correctly.

Page 26: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 27: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Football League Tables

Page 28: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1

3 points for a win and 1 point for a draw

Page 29: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 25

3 points for a win and 1 point for a draw

Page 30: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 25

Arsenal 13 6 1 6

3 points for a win and 1 point for a draw

Page 31: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 25

Arsenal 13 6 1 6 24

3 points for a win and 1 point for a draw

Page 32: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1

2 points for a win and 1 point for a draw

Page 33: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 17

2 points for a win and 1 point for a draw

Page 34: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 17

Arsenal 13 6 1 6

2 points for a win and 1 point for a draw

Page 35: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Which is the better team?

Played Win Lose Draw Points

Man Utd 12 8 3 1 17

Arsenal 13 6 1 6 18

2 points for a win and 1 point for a draw

Page 36: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Played Win Lose Draw Points

Man Utd 12 8 3 1 25

Arsenal 13 6 1 6 24

3 points for a win and 1 point for a draw

2 points for a win and 1 point for a draw

Played Win Lose Draw Points

Man Utd 12 8 3 1 17

Arsenal 13 6 1 6 18

Page 37: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 38: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Medical and Security Testing

Page 39: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Suppose there are 100 students in your year group, and one of which has stolen another

student’s lunch money. To find the thief everyone undergoes a lie detector test which has an accuracy of 95%. The lie detector says

you are a thief. What is the chance you are actually the thief?

What is the chance you are actually the thief?

17% 84% 5% 95%

Page 40: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Understanding Uncertainty Bayes Theorem animation

Page 41: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 42: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Statistics used in CourtStatistics used in Court

Sally Clark

On the 9th November 1999, Sally Clark, was convicted of

the murder of her first two children.

Page 43: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

What was said in Court?What was said in Court?

Chances of 1 child dying from SIDS = 1 in 6,500

Chance of both children dying from SIDS

= 1 in 73,000 000

21

6500

Was Sally Clark guilty?

Page 44: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

What wasn’t said in court?What wasn’t said in court?

The birth of two children are NOT exclusively

independent events.

Page 45: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

What wasn’t said in court?What wasn’t said in court?Statistics of two babies dying

from the same mother:

One third caused by rare but known natural causes (not SIDS).

One third associated with child abuse.

One third true “SIDS” deaths.

2 in 3 chance of innocence, rather than a 1 in 73 million chance of innocence.

Page 46: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

What wasn’t said in court?What wasn’t said in court?

Probability of a mother murdering

2 of her own children is

1 in 2,200,000,000

Much lower than the probability of

both children dying from SIDS

1 in 73,000,000

Page 47: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.
Page 48: THE RISKS OF LIFE Video Conference 5 Media & Misinformation.

Compulsory InvestigationCompulsory Investigation

Using probability to understand medical and security test

results.