Eight Rules 1. False Precision 2. Base Rate Fallacy 3. Correlation Not Causation 4. Single Variable 5. Substantial Differences 6. Longitudinal Study 7. Differences From Chance |
If A causes B: | ||||||||||||||||||
More of A produces more of B; | |||||||||||||||||||
Longer exposure to A produces more of B; | |||||||||||||||||||
Exposure to A, then stopping, produces less of B than continual exposure; | |||||||||||||||||||
The amount of B decreases as the amount of time since exposure to A increases; | |||||||||||||||||||
These relationships are not necessarily linear
|
|||||||||||||||||||
— If you stop smoking your risk decreases — The longer the time since you stopped, the lower your risk — The more you reduce, the lower your risk — The longer the time since you reduced, the lower your risk |
|||||||||||||||||||
TOP | |||||||||||||||||||
|
|||||||||||||||||||
1. | Dose x |
|
|||||||||||||||||
2. | Dose 2x |
|
|||||||||||||||||
3. | Dose 3x |
|
|||||||||||||||||
4. | Dose 4x |
|
|||||||||||||||||
5. | Placebo |
|
|||||||||||||||||
Disadvantage: Everyone in study knows their dose does not change throughout study. How can we fix this?
What if we change this, so that everyone knows they switch (from placebo to drug or drug to placebo). How can we fix this? |
|||||||||||||||||||
TOP | Eight Rules |