ap psychology ms. simon september 17-24, 2009 introduction to research methods
TRANSCRIPT
Scientific Theories
• Set of principles that organize and predict behaviors or events
• Link observed facts
• Imply hypotheses that offer testable predictions
Subfields of Psychology
• Basic Psychology- research
• Applied Psychology- research put into practice as therapist
• Psychiatry- a medical field- deals with mental disorders- prescribe medication
There are three main types of research methods in psychology: Descriptive
Correlational
Experimental
Descriptive Study:
1) Case Study- psychologists study one individual in great depth in hopes of revealing universal principles
The Problem with the Case Study:
An individual may be atypical
Cannot generalize results
Difficult to Manipulate Variables
Difficult to quantify data
Naturalistic Observation
Pros ConsRealistic No manipulat-Inexpensive Ion of
variablesFew ethicalconsideration
Observer Bias
Survey Method• Experimenter must identify the
population to study
• Random sampling picking members from a population randomly to ensure a representative sample
Survey Method
Pros ConsCan gather lots of data
Honest answers?
Few ethical considerations
Need many participants
Inexpensive Wording Effects
Wording Effects
• In a study by AMNH, 88% of all respondents said that they were interested in plants and trees, but only 39% said they were interested in botany.
• One out of five Americans (22 percent) doubted that the Holocaust had occurred. 12% said they weren’t sure
What to Watch for…
• Order of choices
• Is the time frame specified?
• How personal or direct is the wording?
• Is there a cultural bias?
Correlational Studies
• Correlational studies assess the association between two or more characteristics of interest without ascribing causes
• Is a correlational study an experiment?
Correlational Studies
Pros ConsInexpensive Confounding
variablesFew ethical considerations
Illusory correlation
Time Does notObjective I mply
causation
Illusory Correlation
• When we believe there is a relationship between two things, we are likely to notice and recall instances that confirm our belief
What is the correlation?• 1) good looks/popularity • 2) self-esteem/depression • 3) brain size/intelligence • 4) money/happiness • 5) Education/length of life
Correlation: determining causation There is a strong
correlation between people who ate oatmeal for breakfast as a child and cancer versus people who ate Frosted Flakes for breakfast as a child
Friday’s Exam: What to Study
• About 40% Historical Approaches, 60% Research Methods
• Study mostly from Class Discussion!
• Format: Multiple Choice, One short AP exam style essay question, short answer and/or matching
Research Methods
Descriptive Correlational ExperimentalExplains behavior using natural observations
assess the association between two or more characteristics of interest
Researcher manipulates one variable and observes the effect on another variable
1) Case studies2) Naturalistic3) Surveys
1) Causation2) Illusory correlation3) Correlational Coefficient
Experimental DesignConfounding Variables, double-blind
Experimental Method• Researcher manipulates one
variable (independent variable) and observes the effect on another variable (dependent variable)
• Allows one to show a causal relationship
• Q: How is the experimental group different from the control group?
• A: The experimental group receives or reacts to the independent variable and the control group does not receive the independent variable
• Q: Why is a control group needed?
• A: Provides baseline for comparison
Confounding variable:
external differences between the experimental group and the control group other than those
resulting from the independent variable
Confounding Variables
• Demand Characteristics- participants form an interpretation of the experiment's purpose and unconsciously change their behavior accordingly
How can we control for confounding variables?
1) Random Assignment method of assigning subjects to groups to
minimize pre-existing differences between those groups
This is an example of Between subjects design: Participants in the experimental and control group are different individuals
How can be control for confounding variables?
2) Within subjects design Technique where subjects serve as control and
experimental group.
3) Twin Studies:
Confounding Variables
• Experimenter bias- researcher’s expectations about the outcome of a study influence the results
Q: How can we eliminate experimenter bias?
How can we eliminate experimenter bias?1. Multiple Experimenters
2. Double blind procedure- research design in which neither the experimenter or the participants know who is in the experimental versus control group
Statistics• Statistical significance (p) is the
likelihood that the observed difference between groups results from a real difference rather than chance alone
• What’s a good p value?
But… what’s a standard deviation?
A standard deviation is the degree by which a score varies from the mean
You don’t need to know the equation!
Mean, Median, Mode
6, 7, 8, 9, 10, 10, 10, 11, 11, 11, 11, 12, 12, 13
What’s the mode?
What’s the median?