MegaMaester

Scientific Thinking · Module 2

Evaluating Scientific Evidence

Module 1 covered how evidence is produced. This module covers how to judge evidence someone else produced — which is the situation almost everyone is in, almost all of the time.

It moves from reading a graph honestly, through confounding and statistics, to the social machinery that makes science self-correcting: peer review, replication, and the failure modes that appear when those break down.

Lessons
7 lessons
Estimated time
~5-7 hours
Assessment
Module quiz included

Not started yet

Lessons

Lesson 1 · 17 min

Data and Graphs

Lesson 2 · 18 min

Correlation, Causation and Confounding

Lesson 3 · 18 min

Statistical Thinking for Everyone

Lesson 4 · 17 min

Peer Review and Replication

Lesson 5 · 18 min

Pseudoscience and Scientific Misconduct

Lesson 6 · 17 min

Scientific Thinking in Everyday Life

Lesson 7 · 25 min

Capstone Investigation

Check your understanding

Finish the lessons, then test yourself. Your best score is saved on this device. You can also open this quiz on its own page.

Evaluating Scientific Evidence — Module Assessment

12 questions · pass mark 75%

  1. 1.When is a truncated y-axis legitimate?
  2. 2.A trial chart shows average symptoms falling. What missing information could explain it entirely?
  3. 3.What two properties make a variable a confounder?
  4. 4.Simpson's paradox describes a case where:
  5. 5.What does a p-value of 0.03 mean?
  6. 6.A test is 99% accurate for a condition affecting 1 in 10,000. A positive result means roughly:
  7. 7.Why is testing twenty outcomes at p < 0.05 a problem?
  8. 8.What does peer review reliably catch?
  9. 9.Why is replication stronger evidence than publication?
  10. 10.Which is the clearest marker of pseudoscientific reasoning?
  11. 11.What does 'the evidence is unclear' represent in an investigation?
  12. 12.Name three of the Bradford Hill considerations used to build a causal case without experiment.
Answer every question to submit.