1 of 28

Lecture 2

Cause and Effect

DATA 8

Fall 2023

2 of 28

Announcements

  • Lab 01 has been released!
    • Due 11 PM Friday 8/25 for all students
    • Office Hours in Warren Hall 101B/Online from 11AM-2PM Friday
    • More details on Ed
  • New lab sections releasing by Sunday; will announce on Ed
  • Homework 01 has been released!
    • Due 11 PM Wednesday 8/30
    • Turn in by 11 PM Tuesday 8/29 for 5 bonus points!
  • Regular office hours start next week

Data 8, UC Berkeley

3 of 28

A Link

https://www.everydayhealth.com/diet-nutrition/eating-chocolate-regularly-linked-to-lower-heart-attack-risk/

https://journals.sagepub.com/doi/full/10.1177/2047487320936787

Report on a July 2020 article in the European Journal of Preventive Cardiology

Data 8, UC Berkeley

4 of 28

Observation

  • individuals, study subjects, participants, units
    • 336,289 US, Swedish, and Australian adults in several studies

​

  • treatment
    • chocolate consumption

​

  • outcome
    • heart disease

Data 8, UC Berkeley

5 of 28

The first question

​

Is there any relation between chocolate consumption and heart disease?

​

  • association
    • any relation
    • link

Answer: Yes, because those who ate chocolate had less heart disease.

Data 8, UC Berkeley

6 of 28

A Stronger Link?

Other headlines about the same article:

European Society of Cardiology Press Release

Family Safety and Health, National Safety Council

https://www.safetyandhealthmagazine.com/articles/20257-is-eating-chocolate-heart-healthy-study-says-yes

https://www.escardio.org/The-ESC/Press-Office/Press-releases/Chocolate-is-good-for-the-heart

Data 8, UC Berkeley

7 of 28

The next question

Does chocolate consumption lead to a reduction in heart disease?

​

  • causality

​

This question is often harder to answer.

​

​

​

​

​

“Dr. Alice Lichtenstein, an American Heart Association volunteer and professor of nutrition science and policy at Tufts University, was more skeptical of the findings.”

Market Watch

https://www.marketwatch.com/story/eating-chocolate-once-a-week-can-lower-your-risk-of-heart-disease-study-2020-07-23

Data 8, UC Berkeley

8 of 28

Association

9 of 28

London, early 1850s

​

Illustration from Punch (1852)

Data 8, UC Berkeley

10 of 28

Miasma

​

  • Bad smell given off by waste and rotting matter
  • Believed to be the main source of disease

​

​

  • Believers included:
    • Florence Nightingale
    • Edwin Chadwick, Commissioner of the General Board of Health

11 of 28

Suggested Remedies

Cholera, around 1850

  • “fly to clene air”
  • Fire off barrels of gunpowder
  • Fetch clean air from the top of the Eiffel Tower in Paris

​

Covid19, 2020

  • Inject disinfectant
  • Sunlight
  • Hydroxychloroquine
  • Take 6 deep breaths, then cough while covering mouth
  • Cannabis, cocaine, mangoes, onion, garlic, drinking water every 15 minutes, tea, eating ice cream, avoiding ice cream

12 of 28

John Snow, 1813-1858

Data 8, UC Berkeley

13 of 28

14 of 28

15 of 28

16 of 28

17 of 28

18 of 28

Mr. Nicholas Cleary kindly moved aside so I could photograph the wall behind him, and then returned to strike a familiar pose.

– London, 2017

19 of 28

Causation

20 of 28

21 of 28

Comparison

  • treatment group

​

  • control group
    • does not receive the treatment

Data 8, UC Berkeley

22 of 28

Snow’s “Grand Experiment”

“… there is no difference whatever in the houses or the people receiving the supply of the two Water Companies, or in any of the physical conditions with which they are surrounded …”

​

  • The two groups were similar except for the treatment.

Data 8, UC Berkeley

23 of 28

Snow’s table

Supply Area

Number of houses

Cholera deaths

Deaths per 10,000 houses

S&V

40,046

1,263

315

Lambeth

26,107

98

37

Rest of London

256,423

1,422

59

Data 8, UC Berkeley

24 of 28

Key to establishing causality

If the treatment and control groups are similar apart from the treatment, then differences between the outcomes in the two groups can be ascribed to the treatment.

Data 8, UC Berkeley

25 of 28

Confounding

26 of 28

Trouble

If the treatment and control groups have systematic differences other than the treatment, then it might be difficult to identify causality.

​

Such differences are often present in observational studies.

​

When they lead researchers astray, they are called confounding factors.

​

Data 8, UC Berkeley

27 of 28

Randomize!

  • If you assign individuals to treatment and control at random, then the two groups are likely to be similar apart from the treatment.

​

  • You can account – mathematically – for variability in the assignment.

​

  • Randomized Controlled Experiment

Data 8, UC Berkeley

28 of 28

Careful ...

Regardless of what the dictionary says,

in probability theory

​

Random ≠ Haphazard

Data 8, UC Berkeley