1. A way of knowing about the natural world using a process designed to reduce the chance of being misled.

2. Process:

• Observation
  • Go see your friend's fish pond to confirm the absence of fish, etc.
• Question
  • Ask, "Why have my neighbor's fish died in the pond?"
• Literature Review
  • Read the literature to find out the survival requirements for the fish that were in your neighbor's pond and to find out about past reasons fish have died in ponds.
• Multiple Hypotheses
  • Come up with as many possible answers to the question as you can.
    • The water was not the right temperature, there was not enough oxygen, there was not enough food, over-harvesting by your neighbor or predators, toxins, etc.
• Deductions
  • For each hypothesis, determine what evidence (data) would be required to refute the hypothesis or support the hypothesis.
    • Example: The water temperature must be below 40 degrees or above 90 degrees to kill the fish.
• Tests
  • Collect the actual data
    • Example: The water temperature in the pond was always between 71 and 82 degrees.
• Tentative Conclusions
  • Report your methods and results and state which hypotheses seem to be supported and which you were able to rule out tentatively.
    • Example: Water temperature most likely was not the cause of the fish dieoff.
• Peer Review
  • Submit your report to a scientific journal. The journal editors give your report to scientists who critique the scientific soundness and merits of your report.

3. Rules

• Maximize Sample Size
  • Collect data from as large a sample as possible and needed based on the population's variability and the amount of certainty required.
    • Example: if you want to know the average height of PCC students, you can't just measure the height of 10 students, measure the height of 1000s.
• Representative Sample
  • Choose the sample you are going to study randomly or systematically to avoid bias.
    • Example: if you want to know the average height of PCC students, you can't just measure the height of all the basketball players, you need to measure the height of students representing as much of the variability as possible.
• Controlled Studies
  • Use when determining the cause of an effect.
    • Always have a control group to compare to the experimental group and always have just one variable be different between the control group and the experimental group at a time so you can be more sure that any effects seen in the experimental group that are not seen in the control group likely arose from that particular cause.