RESEARCH
METHODS
AP Psychology
Ms. Brown
Myers – Ch. 1
LIMITS OF INTUITION
& COMMON SENSE
“The naked intellect is an extraordinarily inaccurate instrument.”
- Madeleine L’Engle (1977)
Why Science?
• How can we best understand why people think, feel, and act as
they do?
Hunches?
• Common sense?
• Intuition?
•
• Psychology needs a scientific approach to separate “common
sense” and “hunches” from actual credible evidence.
Differ from person to person
• Not always 100% correct
• Always seem correct after the fact
•
Hindsight Bias
• The tendency to believe, after learning an outcome, that one would
have foreseen it.
•
“I knew it all along phenomenon”
• Hindsight is 20/20 – after the fact, it is easy to see why an event
occurred, someone acted in a certain way, etc
• Psychological findings often seem already known or like common
sense because we constantly observe psychological phenomena or
“knew it all along.”
Overconfidence
• The tendency to overestimate our abilities, including knowledge (the
illusion of knowledge)
Faster
• Smarter
• Stronger
• More aware
• More observant
•
• Even after being presented with evidence that contradicts previous
assumptions about ourselves, most claim, “Well I was close.”
Draw a bike.
A
B
C
D
A
I bet you were pretty confident you knew how to draw a
bike…
Our brains allow us to be overconfident to shield us from
everything we DON’T know – “the illusion of knowledge”
effect.
How many of you are claiming, “I knew it all along!”
(hindsight bias!)
The Scientific Method
• A method in which scientists make observations, form theories, and
then refine theories in the light of observations.
1) Theories
Theory - an explanation using an
organizes data and predicts
observations (NOT fact)
Hypothesis - a testable prediction,
often implied by a theory
3) Research and
Observations
2) Hypothesis
Lead to…
Replication – repeating the study
to see if the basic finding extends
to other participants in other
circumstances
DESCRIPTIVE
METHODS
Description is the starting point of any science.
These types of research methods describe behaviors/attitudes, not
explain them.
Case Study
• An observation technique in which one person is studied in depth with
the hope of revealing universal principles.
• Pros:
•
Insight into specific cases that possibly could not be studied due to ethics
• Cons:
Individual studies are hard to generalize to large populations
• Some events/circumstances cannot be replicated (ethics)
•
• Ex: brain lesion studies, instances of socially isolated (“feral”)
children
Survey
• A technique for ascertaining the self-reported attitudes or behaviors
of people, usually by questioning a representative, random sample of
them.
• Must have a representative and random sampling from the
population for generalization to be possible
Population – all of the cases in a group from which samples may be
drawn for study
• Representative – reflective of the population
• Random sample – a sample in which all individuals have an equal chance
of inclusion in the study
• Generalization – the ability to reflect results from the random sample on
the entire population
•
Survey
• Pros:
•
Can study large populations from a representative sample
• Cons:
•
Self-reporting is not always reliable
•
Social-desirability effect – the tendency of respondents to answer questions in
a manner that will be viewed favorably by others.
Samples are not always representative
• Results are largely based on how questions are worded (“aid to the
needy” instead of “welfare”)
•
• Ex: Gallup polls, Kinsey Report on sexuality
Naturalistic Observation
• Observing and recording behavior in naturally occurring situations
without trying to manipulate and control the situation.
•
The observer must not manipulate or stage the situation.
• Pros:
•
Observe people/animals in real, not artificial, environments
• Cons:
•
No control over events or variables
• Ex: videotaping mothers and children together in different cultures,
recording students’ self-seating patterns in the lunchroom
CORRELATIONAL
METHODS
Describing behavior is the first step to predicting it.
When observed variables seem to relate to each other, it is said they
correlate.
Correlational Research
• A measure of the extent to which two factors vary
together, and thus of how well either factor predicts the
other.
• How are two things related?
• How strong is this relationship?
• Can the relationship shape predictions?
• Scatterplots – a graphed cluster of dots, each of which
represents the values of two variables. The amount of
scatter suggests the strength of the correlation (little
scatter = more correlation)
Positive Correlation
• Two variables rise or fall
together
• The taller you are, the more you
weigh
• The more you smoke, your risk of
cancer increases
• As temperature rises, crime rate
increases
• As the ocean level decreases, the
fish population decreases
Negative Correlation
• Two variables relate
inversely to each other –
as one rises, the other
falls.
• The more you brush your
teeth, your risk of cavities
decreases
• The more years spent in jail,
the lower the education level
• The more you hold a baby,
the less it cries
• The more hours spent
watching TV, the less time
spent doing HW
No Correlation
• Two variables do not seem to be related
• People born later in the year and intelligence level
Correlation coefficient
• The mathematical expression of the relationship,
ranging from -1 to +1
• Measures how well either one predicts the other and how
strong that relationship/prediction is (0 = no relationship)
• r = +0.37
r  correlation coefficient (relationship)
• +  indicates direction of relationship (positive or negative)
• 0.37  Indicates strength of relationship (0.00 weak to 1.00 strong)
•
r = +1.00
r = -1.00
r=0
Practice
Positive or Negative Correlation?
•
•
•
Strong , Moderate, or Weak
Correlation?
Those with higher rates of
depression tend to have higher risks
of suicide.
• +0.71
The louder the music while studying,
the lower the exam performance.
• -0.13
The more you observe aggression,
the more aggression you display.
•
•
Fairly strong relationship
Fairly weak relationship
• +0.46
•
Moderate relationship
Correlation IS NOT Causation
• Correlations cannot fully predict future behaviors/attitudes, regardless of
how strong the correlation coefficient is.
•
Ex: low self esteem is correlated with depression, however this does not mean
that low self-esteem directly causes depression.
Could cause
(1) Low self esteem
Depression
OR
(2) Depression
Could cause
Low self esteem
OR
(3) Distressing
events or biological
predisposition
Could cause
Depression
Low self esteem
Correlational Research
• Pros:
• Can measure the extent of a relationship
• Cons:
• Correlation is not causation (just because two things are related
does not mean one causes the other)
EXPERIMENTAL
METHODS
Because many factors influence everyday behaviors/attitudes,
psychologists need to isolate and control variables to establish cause
and effect relationships.
They do this using experiments.
Experiments
• A research method in which an investigator
manipulates one or more factors (independent
variables) to observe the effect on some behavior or
mental process (the dependent variable). By random
assignment of participants, the experimenter aims to
control other relevant factors.
• Manipulate the factors of influence
• Hold other variables constant
• Unlike correlational studies that uncover naturally occurring
relationships, an experiment manipulates a factor to
determine its effect.
Variables – anything that can vary
• Independent variable
•
The variable manipulated by the experimenter
• Dependent variable
•
The outcome being studied as a result of the ind. variable
• Ex: Hypothesis – Pill X can reduce the symptoms of schizophrenia.
Ind. - Drug
• Dep. – symptoms of schizophrenia
•
• Mnemonic (memory aid): make a if-then statement  “IF the independent
variable THEN the dependent variable.”
Operational Definition
• Specific definition of the independent and dependent
variables
• Very specific especially when variables are vague
• Ex: Hypothesis – Pill X can reduce the symptoms of
schizophrenia.
• Ind – Drug
•
Operationalize: Pill X, placebo
• Dep – symptoms of schizophrenia
•
Operationalize: the amount of dopamine activity in the brain, the
prevalence of disorganized speech
Research groups
• Experimental group
• Participants that receive the independent variable
• Control group
• Participants NOT exposed to the independent variable
• Serves as a comparison for evaluating the effect of the ind.
variable (sets a base-line)
• Ex: Hypothesis – Pill X can reduce symptoms of
schizophrenia. Group A receives Pill X, while Group B
receives a placebo (fake drug).
• Exp. – Group A receiving Pill X (ind. variable)
• Control – Group B receiving placebo
Placebos
• Placebo – a substance or treatment that has no effect apart from a
person’s belief in it.
• Placebo effect – a person receiving the placebo may report to
positive effects due to a belief in the drug/treatment
• Single blind study – participants do not know if they are in the
experimental or control group
• Double blind study – participants nor researchers know who is in
the experimental or control group
•
Reduces researcher bias - process where the scientists performing the
research unconsciously influence the results, in order to portray a certain
outcome.
Experiments
• Pros:
Variables can be controlled and manipulated
• Can determine cause-and-effect
• Can be replicated
•
• Cons:
Labs can not always duplicate real-life environments
• Can be expensive
• Ethics can prohibit certain experiments
•
COMPARING RESEARCH METHODS
Research Method
Basic Purpose
How Conducted
What is
Manipulated?
Weakness
Descriptive
To observe and
record behavior
Case studies,
surveys, or
naturalistic
observations
Nothing
No control of
variables; single
cases may be
misleading
Correlational
To detect naturally
occurring
relationships; to
assess how well one
variable predicts
another
Compute statistical
association,
sometimes among
survey responses
Nothing
Does not specify
cause and effect
Experimental
To explore cause
and effect
Manipulate one or
Independent
more factors;
variable(s)
random assignment
to groups
Crash Course – Research Methods
Sometimes not
feasible; results may
not generalize to
other contexts; not
ethical to manipulate
certain variables
STATISTICAL
REASONING
Isn’t this AP Psych, not AP Stats?
• After gathering dating with research methods, it needs to be
organized and summarized using statistics so we can make
inferences about the data.
• We make sense out of data using statistics.
Measures of Central Tendency
• a single score that represents a whole set of
scores/data
• Mean – the average of the scores/values
• Median – the middle score /value in a distribution
• Mode – the most frequent score/value
• Eg: Test scores – 60, 70, 70, 80, 80, 80, 80, 80, 90,
90, 100
• Mean: 80
• Mode: 80
• Median: 80
Measures in Variation
• Allows us to see the variation or difference in a set of scores/values.
Range - the difference between the highest and lowest scores/values.
• Standard deviation – a measure of how much scores/values vary around
the mean.
•
σ=
Sum of (deviations from the mean)2
Number of scores
Range and Standard Deviation
• Eg: Test scores- 60, 70, 70, 80, 80, 80, 80, 80, 90, 90, 100
• Mean: 80
• Mode: 80
• Median: 80
• Range: 100-60
•
= 40
The measure of dispersion of data
• Standard Deviation:
•
11
The degree to which scores differ from each other and vary around the
mean value for the set (will always fall between 0 and ½ of range)
As the standard deviation approaches 0, the closer the scores are to each
other.
• As the standard deviation approaches ½ of range, the more dispersed the
score are.
•
Mean,
Median,
Mode
Distribution of Quiz Scores
6
Frequency
5
4
3
2
1
0
0
20
40
Quiz Scores
60
80
100
120
Standard Deviation = 11 points (how much scores
vary from each and around the mean)
Normal Distributions
• In a typical distribution of data:
68% of scores will be within 1 standard deviation above or below the mean
• 95% of scores are within 2 standard deviations above or below the mean.
• 99.7% of scores are within 3 standard deviations above or below the mean.
•
Not a perfect distribution, therefore the
quiz is not standardized .
Mean,
Median,
Mode
Distribution of Quiz Scores
6
Frequency
5
4
3
2
1
0
0
20
40
Quiz Scores
60
80
100
-1 to +1 St.Dev. = 77%
-2 to +2 St.Dev. = 100%
120
Statistical Significance
• a statistical statement of how likely it is that an obtained result
occurred by chance.
•
Uses a fancy formula that we do not need to know… HOWEVER…
When a researcher says…
It means…
“When the results from the
experimental group and the control
group were compared, they were
found to be statistically significant.”
The results could not have
statistically occurred by chance….
There is a direct relationship
between the variables.
“When the results from the
experimental group and the control
group were compared, they were
NOT found to be statistically
significant.”
There is not enough statistical
evidence to say that the variables
are in fact directly related and the
results could have just occurred by
chance.