Descriptive Research Strategies: The Case Study

Research Strategies in Psychology
Descriptive Research Strategies:
The Case Study
Descriptive Research Strategies:
The Case Study
• “Given a thimbleful of [dramatic] facts we
rush to make generalizations as large as a
tub” - Psychologist Gordon Allport
» Fears after school shootings or kidnappings
» The contradictory case: e.g. “My
grandfather smoked until he was 102”
» Unreliability of anecdotal evidence
Descriptive Research Strategies:
The Survey
• Used in both
descriptive and
correlational studies
• Asks people to report
on behaviour and
opinions.
The Survey:
Wording Effects
• Should cigarette ads or pornography be
allowed on television?
• People prefer “not allowing” to “forbidding” or
“censoring”
• Only 27% of Americans approved of “government
censorship” of media sex and violence, while 66%
approved of “more restrictions on what is shown on
television”
• People prefer “aid the needy” to “welfare”
• EI instead of UI
The Survey:
Sampling
• False consensus affect
• A “sample” is a small group selected
randomly to represent the larger
population
• The larger the sample the more accurately
it represents the larger population
- Top of pg26
Descriptive Research Strategies:
Naturalistic Observations
• Not just about
chimpanzees in
nature.
Correlational Research Strategies:
Correlation
• The Correlation coefficient is a statistical measure of
relationship.
• Is there a relationship between stress and illness?
• Is there a relationship between test scores and school success?
• Scatter plots: pg 28
Correlational Research Strategies:
Correlation and Causation
• Correlation is not causation
• For example:
– Low self esteem and depression have a
positive correlation
• Low self-esteem could cause depression
• Depression could cause low self esteem
• Distressing events or biological disposition could
cause both depression and low self esteem.
Correlational Research Strategies:
Illusory Correlation
• When we believe there is a relationship between
two things, we are likely to notice and recall
instances that confirm our belief.
– Full moon correlations
– Getting cold and wet makes one catch a cold
– Sugar makes kids hyperactive
Correlational Research Strategies:
Perceiving Order in Random
Events
• Shown random data, scientists and psychics
alike can often “see” an interesting pattern.
• If I flip a coin 9 times and ‘heads’ comes up each
time, is it more likely that the tenth time will be
‘tails’?
• When no ones wins a given lottery draw, the
prize of the subsequent draw increases and
more people buy tickets. How does this affect
the chances of a single ticket winning?
• Do professional athletes go on “streaks” or
develop “hot hands” – pg 32
Experimental Research
Strategies:
•
Experiments enable a researcher to
focus on the possible effects of one or
more factors by:
1. Manipulating the factors of interest
2. Holding constant (“controlling”) other factors
•
E.g. There is a strong positive correlation
between breast feeding and later
intelligence. Is the later intelligence
caused by breast feeding? Pg 33
Experimental Research Strategies:
Evaluating Therapies
• If we get better 3 days after taking a
certain pill we tend to think that pill is the
reason we got better.
• 1700’s bloodletting seemed to be very
effective.
• To find if something is actually effective we
need to experiment
Experimental Research Strategies:
Evaluating Therapies
• The Placebo Effect …fake pills often
work because people believe they do.
• Double-blind procedure creates an
experimental condition in which people
receive the treatment and a contrasting
control condition without the treatment.
By randomly assigning people to these
conditions the two groups should
otherwise be identical.
Experimental Research Strategies:
Evaluating Therapies
• The independent variable: the experimental
factor that is manipulated; the variable whose
effects are being studied
• The dependent variable: the variable that may
change in response to manipulations of the
independent variable (e.g. behaviour or mental
process)
• “Experiments aim to manipulate an independent
variable, measure the dependent variable, and
control all other variables” –pg35
• Can Subliminal Tapes Improve Your Life? –pg
35
Statistical Reasoning:
• Bombarded by
numbers
• Harper’s Index
Statistical Reasoning:
Measures of Central Tendency
• Mode – the most frequently occuring
score in distribution
• Mean – the average
• Median – the middle score in a
distribution; half the scores above and half
are below it.
Statistical Reasoning:
Measures of Variation
• Averages based on scores with low
variability are more reliable than averages
based on scores with high variability
• Range: the difference between the highest
and lowest scores in a distribution
• Standard Deviation: a computed
measure of how much scores vary around
the mean score
Statistical Reasoning:
Making Inferences
• When is an observed difference
reliable?
1. Representative samples are better than
biased samples
2. Less-variable observations are more reliable
than those that are more variable.
3. More cases are better than fewer
Statistical Reasoning:
Making Inferences
• When is a difference significant?
– Statistical significance: a statistical
statement of how likely it is that an
obtained result occurred by chance.
– Occurs when the sample averages are
reliable and the difference between
them is relatively large.