The Scientific Method

Download Report

Transcript The Scientific Method

The Scientific Method
Chapter 1
Overview
What is Science?
The Scientific Method
From Hypothesis to Theory
What is Science?
The goal of science is to understand natural
phenomena
2 types of science:
• Discovery science
• Hypothesis-based science
Four-eyed butterfly fish
Discovery Science
Describes nature
Uses inductive reasoning
Observations used to infer conclusions
(but no complete certainty as to validity)
e.g. This red jellybean tastes like cherry.
Thus, all red jellybeans taste like cherry.
=
?
Hypothesis-Based Science
Uses deductive reasoning
Previously-known premises used to
form conclusions (as long as
premises are correct, conclusion
will be too)
e.g. Dogs are mammals.
All mammals have hair or fur.
Thus, all dogs have hair or fur.
Critical Thinking
Using data and facts to
make inferences, draw
conclusions, solve
problems, etc.
Should be unbiased
thought but rarely is
The Scientific Method
Investigation of biological systems
• Observation
• Hypothesis
• Predictions
• Test
• Analysis of results
• Conclusion
Formulating a Hypothesis
Proposes the answer to a question about an
observation
e.g. Exposure to fertilizer increases
reproduction of bean plants
A Good Hypothesis …
Must be testable and falsifiable
A hypothesis can be falsified but can never be proven
true: evidence can only support
Scientific knowledge is ever expanding as new
methodologies are applied to old questions,
etc.
Examples of Good Hypotheses:
• Cactus spines reduce herbivory
• UVB radiation causes limb deformities in
amphibians
Examples of Untestable Hypotheses:
• Women are more moral than men
• Dogs are smarter than cats
Testing a Hypothesis
Scientific experiment
(either falsifies or supports a hypothesis)
• Make predictions
• Determine variables
• Design an experimental procedure
• Carry out procedure
Making Predictions
Based on an experiment designed to test a
hypothesis
If/then statements:
“If the hypothesis is true, then the results of the
experiment will be …”
e.g. If exposure to fertilizer increases reproduction
of bean plants, then the number of pods per
plant should increase in plants exposed to
fertilizer compared with control plants.
If results match prediction, hypothesis is
supported
If not, the hypothesis is falsified
Knowledge of process has been increased
either way, regardless of findings
Determining Variables
Factor, trait, or condition that can exist in
different amounts or types
Independent variable
The variable of interest
Is deliberately manipulated (changed)
Only one is usually chosen
(need to know which factor is affecting the
dependent variable)
Dependent variable
Is measured or observed during an
experiment
Value is caused by and depends on the
value of the independent variable
Control variable
Is kept constant
(so that will not affect outcome of
experiment)
e.g. Exposure to fertilizer increases
reproduction of bean plants
Independent: fertilizer
Dependent: number of bean pods
Control: Other things that would affect bean
reproduction e.g. amount of water,
temperature, amount of light, etc.
Examples
The relationship between drug dosage &
frequency / intensity of symptoms
Independent: amount of drug administered (dosage)
Dependent: Frequency / intensity of symptoms
The relationship between plant growth rate &
duration of light exposure
Independent: Amount of time of exposure to light
Dependent: Plant growth rate
Designing a Procedure
Methodology for an experiment
Quantitative (counts/measurements) methods
usually better than qualitative (descriptive)
(allows re-testing of hypothesis by same or others)
Control Treatment
Independent variable either held at a
standard value or omitted
Results compared between control and
experimental treatments
Allows scientist to decide if predicted effect is
due to independent variable or random
chance
Examples
Researchers want to test the effects of a drug
for treating a medical condition. To ensure
that the drug actually works, they would use a
control group of people …
… who take a placebo drug
Researchers want to see the effects of drinking
caffeinated water. The control group…
… would drink plain water
Level of Treatment
Values set for independent variable
(based on prior knowledge/previous
research)
Can be:
• One specific value (e.g. specific temperature)
• Range of values (e.g. range of temperatures)
Some experiments have no level of treatment
• e.g. gender
Replication
Procedure is repeated several times to see if
results are consistent
Gives an idea of how much variation is
present
Results from different trials are averaged
Sample Size
Larger sample size = greater confidence
Sampling Error
Mean & standard deviations of sample differ
from that of population from which taken
Larger sample size = lower sampling error
Exposure to fertilizer increases reproduction
of bean plants
Level of treatment:
– Specific amount of fertilizer OR
– Range of amounts/concentration of fertilizer
Control:
– Beans grown without applying fertilizer
Replication and sample size:
– Depends … the more the better
Presenting & Analyzing Results
Analysis:
Simple to advanced statistics
9
8
7
6
5
Presentation:
4
3
2
1
0
1
2
3
4
5
12
10
Distance
Tables and/or figures
• Show relationships between
independent and dependent
variables
• Bar graphs vs. line graphs
8
6
4
2
0
0
1
2
Time
More on this in next week’s lab
3
4
Drawing Conclusions
Results are interpreted:
Was hypothesis supported or falsified?
If supported:
Additional experiments suggested to further support
hypothesis
(using same or different methods)
If falsified:
Alternative hypotheses for testing suggested
Exposure to fertilizer increases reproduction
of bean plants
If more pods found on plants treated with
fertilizer:
Hypothesis is supported
Could try different amounts/concentrations
of fertilizer, different fertilizers, etc.
Could also see if combination of variables
affects reproduction e.g. fertilizer & water,
fertilizer & light
Exposure to fertilizer increases reproduction
of bean plants
If no significant difference in number of pods on
beans in either treatment (control or
experimental):
Hypothesis is falsified
Could try testing other factors e.g.
temperature, amount of water, amount of
light, etc.
From Hypothesis to Theory
How does something become a scientific
theory?
When multiple rigorous tests have not
disproved a certain hypothesis
(i.e. an idea with “great explanatory power”)
A theory can be used to interpret other data
and observations
Limits of Science
Only testable hypotheses are valid
Some things are “unexplainable” or
“untestable”
Supernatural, morality, etc.