Ch:3.4 - The Nature of Science

How can we distinguish science from nonscience?

• defining science can be surprisingly difficult
• science comes from the Latin 'scientia,' meaning "knowledge"
• but not all knowledge comes from science

The idealized Scientific Method:

• based on proposing and testing hypothesis
• hypothesis = educated guess
1. make observations
2. ask a question
3. suggest a hypothesis
4. make a prediction
5. perform a test: experiment or additional observation
• Test supports hypothesis - go back to #4 & start from there
• Test does NOT support hypothesis - go back to #3 & start from there
• but science rarely proceeds in this idealized way..... for example:
• sometimes we start by "just looking" then coming up with possible explanations
• sometimes we follow our intuition rather than a particular line of evidence

Hallmarks of Science: #1

• Modern science seeks explanations for observed phenomena that rely solely on natural causes (a scientific model cannot include divine intervention)

Hallmarks of Science: #2

• Science progresses through the creation and testing of models of nature that explain the observations s simple as possible (simplicity = "Occam's razor") 

Hallmarks of Science: #3

• A scientific model must make testable predictions about natural phenomena that would force us to revise or abandon the model if the predictions do not agree with observations

What is a scientific theory?

• The word theory has a different meaning in science than in everyday life
• In science, a theory is NOT the same as a hypothesis
• A scientific theory must:
• explain a wide variety of observations with a few simple principles
• be supported by a large, compelling body of evidence
• NOT have failed any crucial test of its validity
• a model that explains a wide variety of observations in terms of a few general principles and that has survived repeated and varied testing

How can we distinguish science from nonscience?

• Science: seeks explanations that rely solely on natural causes; progresses through the creation and testing of models of nature; models must make testable predictions