The Importance of Sampling in Research

advertisement
Scientific Method
Create testable
operational
definitions
Be curious.
Question your
world.
Collect data using
DESCRIPTIVE
Research
Conduct more
research
Naturalistic
Observation
Survey
Case
Study
Form a
Hypothesis
Relationship?
CORRELATION
Causation?
EXPERIMENT
Doesn’t support
the hypothesis
Continue to
retest
Use this to
predict and
explain the world
Supports the
hypothesis
Create a Theory
Essential Question
• 1-5 How do psychologists draw
appropriate conclusions about behavior
from research?
“The root of the problem is that
in real life, all scientists ever
observe are samples. And, in
real life, all they want to know
about is populations”
Nancy Darling, Ph.D.
The Importance of Sampling in
Research
• Sample
– Small representative subset of a larger population
• Random sample
– Every subject had equal chance of being selected
• Representative sample
– Characteristics of participants correspond to larger
population
• Example: If we know approximately 15% of the United
States’ population is of Hispanic descent, a sample of 100
Americans also ought to include around 15 Hispanic people
to be representative.
You want to make a claim about a WHS student.
• What is the Population?
– All WHS Students
• What will be the sample?
– The participants in the sample
• How could you obtain a random sample?
– Get the student list from the office and pick every 10th
student.
– Pull names out of a hat?
• Why might you want a representative sample
– A random sample won’t account for small populations
•
•
•
•
Ethnic Groups
AP Students
Transitional Students
Students in Life Skills
Survey
Random Sampling
From a population if
each member has an
equal chance of inclusion
into a sample, we call
that a random sample
(unbiased). If the survey
sample is biased, its
results are wrong.
The fastest way to know about the
marble color ratio is to blindly
transfer a few into a smaller jar and
count them.
Sampling in Surveys
• Women and Love study done by Shere Hite
1974
• 98% Dissatisfied by their Marriage
• 75% Extramarital Affairs
• But to all of those who were
mailed surveys only 4%
responded.
When randomly
sampled
• 93% of women are satisfied in their marriages
• Only 7% had affairs
Making Inferences
When is an Observed Difference Reliable?
1. Representative samples are better than biased
samples.
2. Less variable observations are more reliable
than more variable ones.
3. More cases are better than fewer cases.
Making Inferences
When is a Difference Significant?
When sample averages are reliable and the
difference between them is relatively large, we say
the difference has statistical significance.
For psychologists this difference is measured
through p value below .05
Random Assignment
• 100 Volunteers for an experiment
• How do you choose which ones will be in
the control group?
Essential Question
• 1-5 How do psychologists draw
appropriate conclusions about behavior
from research?
Download