Science Data Collection

Definition of Science:
- A systematic way of learning about the world.
- Involves testing hypotheses through observations.
- Utilizes systematic, repeatable methods.
- Conclusions drawn are tentative and falsifiable.
- Importance of peer review and documentation for validation.

Inductive Logic:
- Used to build theories from experiments.
- Involves developing hypotheses based on specific observations.
Deductive Logic:
- Used to predict outcomes of experiments based on hypotheses.
- Allows systematic testing of alternative explanations.
- Quote: "Once you have eliminated the possible, whatever remains, however unlikely, must be true" – Sherlock Holmes.
- Highlights the relationship between specific observations and general predictions.

Hypothesis Defined:
- A proposed explanation for an observation.
- Based on available data and inductive reasoning.
- Leads to specific predictions in the form of if/then statements.
- Must be testable through experiments and falsifiable via additional observations.

Key Distinctions:
- Hypotheses cannot be proven; they can only be supported or refuted.
- Theory Defined:
  - A broad and general mechanism supported by a larger body of evidence than a single hypothesis.
  - Examples: Theory of Evolution, Theory of Relativity.

Purpose of Experiments:
- To test hypotheses by isolating and manipulating variables.
- Observations of outcomes determine hypothesis support.
- Experiments must be repeatable to be considered valid.
- Controls needed to account for variables: comparing experimental groups to control groups.

Independent Variable:
- What is manipulated by the experimenter (treatment variable).
Dependent Variable:
- What is measured as a result in both experimental and control groups (response variable).
Control Group:
- No manipulation or 'fake' manipulation to mimic the experimental group.
Experimental Group:
- Receives the manipulative treatment.

Understanding Correlation:
- Variables may demonstrate correlation or lack thereof.
- Types of Correlation:
  - Positive Correlation:
    - One variable increases as the other does, shown by a positive slope on best-fit line.
  - Negative Correlation:
    - One variable decreases as the other increases, shown by negative slope on best-fit line.
- Correlation can be strong or weak.

Correlation vs. Causation:
- Correlation does not imply causation.
- Samples must be representative of larger populations.
- Randomly chosen samples reduce bias.
- Larger sample sizes and repeated studies improve statistical significance, signaling real differences rather than results from random chance.

Natural Experiments:
- Even experiments with limited controls can yield valuable scientific insights.
- Example: Studies from the eruption of Mt. St. Helens.
Reflection on Science:
- Science does not exist in isolation; context and variables always matter.