Transcript Research Methods

A look at
psychological
research
General principles
• The specious attraction of anecdotes
• The concern for precise measurement
• Operational definitions – definitions which
specify the procedure used to measure
something
• Also, a way to give a study’s focus of concern a
numerical value
• How do you measure anger, love, intelligence?
Research designs
• Naturalistic observations
• A careful examination of
an individual’s or animal’s
behavior in more or less
natural conditions
No manipulation, just observation
Often the first step in generating hypothesis’
Case histories
• A thorough description of a
person or small group of
people with unusual or
noteworthy qualities
Another source of hypothesis’
Just a super-sized anecdote?
surveys
• A study of the prevalence of certain
beliefs, behaviors, or attitudes based upon
people’s responses to specific questions
• Many problems:
sampling
nonchalance
the questions
bias
correlations
• A measure of the relationship between two
variables
• Variable – a measurable item that can vary in
magnitude
• Correlational study – a procedure in which the
investigator measures the relationship between
two variables without controlling either one
Correlations
examples
• Between class attendance and final grade
• Between hours worked and $ earned
• Between smiles given and smiles received
• Between miles run/week and 5k time
• Between hours on internet and final grade
• Between exercise sessions and weight
• Between hair color and grade point average
Correlations
graphic
examples
The correlation
coefficient
• R = the mathematical relationship between
two variables, ranging from -1 to 1
• Positive correlations approach 1
• Negative correlations approach -1
• A R of 0 means that there is no
relationship between the two variables
• Look to R’s absolute value when
assessing its strength
Illusory
correlations
• An apparent relationship based upon
casual observations of unrelated or weakly
related events
• Do people really get
wild under a full moon?
Does the weather affect
arthritis symptoms?
Correlations and
causation
• No matter how it might seem, correlations
do not tell us about cause and effect
• We never really know if changes in one
variable affects the other, or
• If a third, lurking, variable controls them
both.
• Correlations help us predict but not
explain or control behavior or experience.
experiments
experiments
• A study in which the investigator
manipulates at least one variable while
measuring at least one other variable
• By adjusting one variable or factor, while
keeping all other factors constant, we can
determine if that factor causes changes in
the other(s)
More definitions:
variables
• Independent variable – the variable which
is manipulated, or adjusted, by the
investigator
• Dependent variable – the variable which is
measured by the investigator to determine
the effects of the independent variable
• Thoughtfully quantified through the
operational definition
Even more:
groups
• Experimental group – receives the
treatment (independent variable) that the
experiment was designed to test
• Control group – handled exactly the same
as the experimental group except for the
independent variable
• Control groups usually receive a placebo
Hazards to be
prepared for
• Biased groups
• to make sure that the experimental and control
groups are as similar as possible, before being
introduced to the independent variable we use
random assignment
•
every participant must have an equal chance of
being placed in either the experimental or the
control group
more hazards
• Even unconsciously, investigators want to see
their hypothesis confirmed and unintentionally
might distort the results
• Also, the experiment’s participants might try to
help prove the hypothesis
Solution – Double blind – a procedure in which
both the observer and the participant are
unaware of which participants received which
treatment and the experiment’s goals
review
•
•
•
•
•
•
•
Hypothesis
Select Method & Operational Definition
Random Assignment of groups
Run experiment (double blind)
Collect and analyze results (data)
Peer review
Replication