Transcript Ethics in Research

Research Methods in Psychology
Group Activity
Friday August 5, 2011
Design Some Research
1.
2.
3.
4.
5.
6.
7.
The What: Choose an area of psychology & social
science that interests you (rating attractiveness among
teens, intelligence level vs. sleep, musical taste and
behavior, customer treatment based on clothing style
The Who: Develop your target group for study—teens,
young adults, males, females, parents, senior citizens,
The Why: rationale and reason for studying, also what
psychological perspective or theory you might be using
(see chapter 1)
The When: will this project take place. Also include
how long with the research take.
The How: Develop a hypothesis, design a research
method, collect and analyze data.
The Results: Report your findings (where would you
publish or show)
The Replication: how many times will you need to
replicate this research to be able to make accurate
predictions,
The Experiment
• Only research method capable
of showing cause and effect
Experimental Method
1. Review Literature of Past
Research
2. Formulate Hypothesis
3. Design Research/Study Method
(naturalistic observation, case
studies, surveys, experiments,
etc)
4. Collect/Analyze the Data
5. Report the Findings (journal,
critique, replicate)
6. Draw Conclusion or Theory on
Explanation of Findings
Hypothesis
• A statement about the relationship between two or more
variables
• Must be testable and refutable
• Instead of proving the hypothesis, science usually tries to
disprove a null hypothesis.
Null Hypothesis (H0):
opposite of hypothesis
Hypothesis Example:
Statistical Significance :
95% not due to chance
H0: Gender does not have an
effect on perceived intelligence
H1: Gender has an effect on
perceived intelligence
Variables
1. Independent Variable (I.V.): manipulated by
experimenter
2. Dependent Variable (D.V.): MEASURED
variable influenced by independent
3. Operational definition
4. Confounding/extraneous variables
Control Group
• Receives no treatment or placebo
• Serves as a basis for comparison
• Serves to eliminate alternative explanations
Population – The larger group of
people from which a sample is drawn
Sample: Representative of the population
Random: Every member of the pop has = chance
Stratified: Sample is put together by picking a
group statistically equal to the population
Control Measures
• Single-Blind: subject unaware of assignment
• Double-Blind: subject and experimenter
unaware of placement
• Randomization
– From population (sample)
– From assignment to groups (assignment)
Statistical Significance
• Probability results are due to chance
• Inferential stats are used to check for
either a 5% or 1% level of significance.
Lottery tickets 14, 3, 27, 41, 18
1, 2, 3, 4, 5
Coin flips HHHHHHH or HHTHTHT
More likely?
Correlation
A statistical value of the relationship between two variables
Positive Correlation
As one number increases, the other increases.
Ex: Study time to GPA
Negative Correlation
As one number increases, the other decreases.
Ex: Absences to GPA
No Correlation
Variables do not affect one another in a significant way
Ex: Height to GPA
CORRELATION IS NOT CAUSATION!!!
• People that floss everyday live 3 years longer than those that do
not.
• Red wine drinkers live longer than those that do not drink red
wine.
• As speed limits increased on America’s highways, the death rate
went down.
• Children who are played Mozart in the womb have higher IQ’s.
• Marijuana users in youth are more likely to have mental illness as
adults.
• As Swim suit sales increased, so did shark attacks.
Other Research Methods
Ex Post Facto (after the fact)
- Independent variable already present
- Not a true independent variable, no cause and effect
- Often used due to ethical concerns
Naturalistic Observation
- Natural setting: behavior is not interfered with or altered
Survey Method
- Gathers data on attitudes and behaviors.
Case Study
- Intense study of an individual
Flaws in Research
1.
2.
3.
4.
5.
6.
Sampling Bias
Overgeneralization
Placebo effect
Hawthorne/Barnum effect
Demand Characteristics
Experimenter Bias
Ethics in Research
• Participants are free to withdraw at any time
• No undo stress
• Subjects informed of significant factors that may
influence their willingness to participate
• Subjects should be debriefed
• Ethical treatment of animals
• Generally research goes before a review board for
approval
Evaluating Research
What are the STRENGTHS/WEAKNESSES of:
Experiment
Correlation
Surveys
Naturalistic Observation
Case Studies