Transcript Research Methods - elizabethmarquardt


The other choices are intuition and common
sense

If you drop a bullet off of a 3-foot tall table
and fire another bullet straight across an
empty football field, which one will hit the
ground first?



How many murders are there in Michigan
annually?
How many murders are there in Detroit
annually?
613 and 377 respectively



The earth is flat
The earth is the center of the universe
Bathing makes you more susceptible to
illness

When shown a picture of an “effect,” most
people later claimed to remember being
shown a picture of the “cause”


“I knew it all along”
Makes research findings seem like common
sense

Most people tend to be overconfident in their
abilities and guesses about the future

“There is no reason for anyone to have a
computer in their home.”
◦ Ken Olson, President of Digital Equipment
Company, 1977
“Reagan doesn’t have that presidential look.”
United Artists Executive when asked
whether Reagan should be offered the
starring role in The Best Man, 1964

“A severe depression like that of 1920-21 is
outside the range of probability.”
◦ Harvard Economic Society, Weekly Letter, November
16, 1929

“Nuclear powered vacuum cleaners will
probably be a reality within 10 years.”
◦ Alex Lewyt, vacuum manufacturer, 1955



People have a tendency to search for
information that confirms their original idea
Ex: horoscopes
“Your vitality is low. It may be difficult to fight adversity,
Scorpio. Try not to get discouraged. Instead, realize that this
is part of the natural cycle. Awareness of your energy patterns
is half the battle. Don't force yourself into feeling something
you don't. It's important that you not overexert yourself or
take on more responsibilities than you can handle.”
http://my.horoscope.com/astrology/tomorrow-horoscope-scorpio.html

Curiosity
Openness
Skepticism
Humility

Critical thinking






1. Come up with a theory that seems to
explain and predict phenomena
2. Theories lead to hypotheses – testable
predictions
3. Come up with a way to test your
hypothesis



Uses clear operational definitions – defines
variables concretely
Can be replicated with similar results
Is generalizable – the results hold true in
multiple situtations




Case studies
Surveys
Naturalistic Observation
All descriptive research gives us correlation at
the most, NEVER causation


In-depth observations, often over a long
period of time
Study of one person or a small group of
people (or animals)





Identifying new questions for research
Studying rare phenomena
Give vivid, memorable illustrations
Psychobiographies
What are some drawbacks to using case
studies?


One-time, less-in-depth look at a lot of
people
Asks questions in person or on paper



Participants may be ignorant of issue
Answers are affected by gender/race of
questioner
Word choice, grammatical structure, answer
choices, and order of answers can all affect
answers





Are balanced – “Do you agree or disagree?”
Do not assume knowledge – ex: explain
policy
Use everyday language
Use neutral language – avoid positive or
negative connotations
Be specific


Are you in favor of direct retaliatory measures
against Franco’s piracy?
What’s wrong with this?



Representative sampling allows us to get
useful information about an entire population
by asking only part of the population
Random sampling is the best way to get a
representative sample
30 = minimum for sample sizes


Takes place in natural setting (descriptive) or
in laboratory
Advantages/disadvantages?
Can help predict events, but does not explain them
Positive correlation means variables move together (same
direction)
Negative correlation means variables move in opposite
directions



Ranges from -1 to 1 (perfectly negative to
perfectly positive)
The farther away from 0 in either direction,
the stronger the correlation
Describes how often two events occur
together


Controlling the environment (changing
something while keeping everything else
constant) allows researchers to determine
cause and effect
How do we keep everything else constant?


Using a large sample and randomly assigning
people means that their individual differences
don’t matter as much
Participants can be assigned to control group
(no change in environment) or experimental
group (some change administered)

Blind - participants don’t know whether they
are control or experimental
◦ Important because of placebo effect


http://www.youtube.com/watch?v=wsFTgirK
XHk
Double blind- researchers don’t know
whether they are working with experimental
or control
◦ Why is this important?



Independent variables
can be varied
independently of
anything else
What researchers
control
Ex: Patient was given a
drug



Dependent variables
depend on the
independent variables
Results measured by
experimenters
Ex: Patient’s condition
improved



Confounding variables could throw off the
results of an experiment
Ex: one patient has a healthier diet than
another
Random assignment controls for confounding
variables

What are the advantages and disadvantages
of experimental research?
Central Tendency, Variation,
and Significance




Summarize data
Mode- most commonly occurring (French for
“fashionable”)
Median – the middle number of a set (like the
median of a highway is in the middle of the
lanes)
Mean – average: sum of all scores/number of
scores
Mean, median, and probably mode are very close to each other
Ex: intelligence, height
Data is skewed when the mean and median are different
because the mean is being pulled by an outlier, or a score that
is very different from the median
Ex: income
Ex: Age in nursing
homes when
grandchildren go visit



Tell us how consistent scores are
Range – difference between highest score and
lowest score
Standard deviation – tells us how much scores
vary around the mean

Game
Game
Game
Game

Average = 25 points



1
2
3
4
–
–
–
–
10
50
10
30
points
points
points
points

Game
Game
Game
Game

Average = 25 points



1- 28
2 – 24
3 – 22
4 – 26
points
points
points
points
Player 2 is more consistent and
has a smaller standard deviation
Player 1
Player 2




68% of scores fall within 1 standard deviation
95% of scores fall within 2 standard
deviations
99.7% fall within 3 standard deviations
Anything outside 3 standard deviations is an
outlier



Representative samples
Smaller variation (standard deviation)
More cases included in the sample



Needs a relatively large difference between
reliable samples
Means that difference is probably not due to
chance (<5% odds of being due to chance)
Is not the same as practical significance
Statistically
significant
Not
significant



Cause the least possible harm
Treat humanely
Can harm with good reason
1.
2.
3.
4.
Get informed consent
Protect from harm and discomfort
Maintain confidentiality
Debrief afterwards