Transcript Research Methods - Solon City Schools

Research Methods
It is actually way more exciting
than it sounds!!!!
Why do we have to learn this
stuff?
Psychology is first and foremost a science.
Thus it is based in research.
Be aware however of two hurdles that tend to skew our
logic when we research
Hindsight Bias
• The tendency to
believe, after
learning the
outcome, that you
knew it all along.
Example:
Monday Morning
Quarterbacking!!!
Overconfidence
• We tend to think we
know more than we do.
• Examples:
Overconfidence
“There
is no reason for anyone to have a computerin their home.” (Ken Olson, president
of Digital Equipment Company, 1977)
“Heavier-than-air flying machines are impossible.”(Lord Kelvin, British mathematician,
physicist, and president of the British Royal Society, 1895)
“Reagan doesn’t have the presidential look.”(United Artists executive when asked
whether Ronald Reagan should be offered the starring role in the movie The Best
Man, 1964)
“A severe depression like that of 1920–1921 is outside the range of probability.”
(Harvard Economic Society, Weekly Letter, November 16, 1929)
“Man will never reach the Moon, regardless of all future scientific advances.” (Lee
DeForest, inventor of the vacuum tube, 1957)
The Scientific Attitude
• Three main components
Critical Thinking
• Critical Thinking - thinking that does
not blindly accept arguments and
conclusions
– “Smart thinking”
– Four elements
– Empirical Approach
Scientific Method
1. Observe some aspect of the universe.
2. Invent a theory (an explanation) that is
consistent with what you have observed.
3. Use the theory to make predictions,
4. Test those predictions by experiments or
further observations.
5. Modify the theory in the light of your results.
6, Go to step 3.
Thinking that she had outperformed most of her
classmates, Glenda was surprised to receive just an
average grade on her psychology test. Glenda's
experience best illustrates
9
10
21
22
23
24
25
26
27
28
29
30
12
13
15
16
ce
fid
en
rre
...
co
.. .
ec
bo
ce
14
or
re
.. .
11
17
18
0%
rc
on
8
0%
Ov
e
7
0%
Ill
us
or
y
6
at
iv
5
pl
a
4
Th
e
3
hi
n
2
Th
e
1
0%
ef
...
0%
Ne
g
The hindsight bias
The placebo effect
Negative correlation
Illusory correlation
Overconfidence
ds
igh
t
1.
2.
3.
4.
5.
19
20
According to Emily's grandfather, Adolf Hitler's obvious
emotional instability made it clear from the beginning days of his
international conflicts that Germany would inevitably lose World
War II. The grandfather's claim best illustrates
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
15
pl
in
sa
m
n
us
io
17
0%
...
of
...
ce
...
16
ill
yc
so
r
14
do
m
8
0%
18
ra
n
7
an
6
0%
rc
on
f id
5
ov
e
4
or
r
ht
3
illu
nd
sig
hi
2
th
e
1
0%
e.
..
0%
en
the hindsight bias.
illusory correlation.
overconfidence.
an illusion of control
random sampling.
...
1.
2.
3.
4.
5.
19
20
3 Types of Research
• Descriptive
• Correlational
Experimental
3 Types of Descriptive Research
• The Case Study
• The Survey
• Naturalistic Observation
Case Studies
• A detailed picture of
one or a few
subjects.
• Offers suggestions
for further study
• Can mislead if is
subject is atypical.
Example: The ideal case study was
John and Kate. Really interesting,
but are they typical of all families?
Naturalistic Observation
• Watch subjects in
their natural
environment.
• Do not manipulate
the environment.
• May be done when it
is not ethical to
manipulate variables
• Does not show cause
and effect.
Example: Jane Goodall’s
research on chimpanzes
Survey Method
•Works with large groups of data
•Can be Descriptive or Correlational
•Cheap and fast, most used research
method in psychology
Example:
Sampling
• Population - all the possible
subjects in a group you want to
study
• Example:
• Random Sampling – a portion of
the population that fairly
represents the population
because each person has an
equal chance of getting chosen
• Ensures that the participants are Population
representative of a large
population
• Helps avoid false generalizations
• Example:
random sample
Survey Method: The Bad
• Low Response
Rate
• People Lie or
just
misinterpret
themselves.
• Wording
Effects
How accurate would a survey be about the
frequency of diarrhea?
Wording Surveys
•
• Does Ohio spend
too much, too
little or the
right amount on
prisons and
corrections?
• Is Ohio’s
spending $1.3
billion on prisons
and corrections
too much, too
little or the
right amount?
Correlational Method
• Correlation –
• Measures how well one
variable predicts the other
• Does not show causation
• Surveys and Naturalistic
observation can use
correlational method.
As more ice cream is eaten,
more people are murdered.
Does ice cream cause murder, or murder cause people to eat ice cream?
Types of Correlation
Positive Correlation
Example::.
Negative Correlation
Example:
Correlation Coefficient
• A number that measures
the strength and direction
of a relationship between 2
variables.
• Range is from -1 to +1
–
• Scatterplot – a visual
representation of the
relationship between the
variables
– shown as a graphed cluster of
dots
Which is a stronger
correlation?
• -.13 or +.38
• -.72 or +.59
• -.91 or +.04
Which of the following would be a
negative correlation, and which
would be a positive
correlation?
Education and years in jail
Weight and hours of TV watched
Education and income
Holding babies and crying
Food and calories ingested
Correlation
Correlation
Correlation
Correlation
Illusory Correlations
• Illusory Correlation
–Perceived non-existent
correlation
–A random coincidence
Psychologists who carefully watch the behavior
of chimpanzee societies in the jungle are using a
research method known as
2
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
nd
om
sa
m
pl
st
ud
in
.. .
y
0%
Ra
Na
t
0%
ca
se
ur
a
lis
t ic
en
ta
t
im
14
0%
o.
..
y
su
rv
e
Ex
pe
r
Th
e
1
0%
io
. ..
0%
Th
e
1. The survey
2. Experimentation
3. Naturalistic
observation
4. The case study
5. Random sampling
15
16
17
18
19
20
If psychologists discovered that wealthy people are less
satisfied with their marriages than poor people are, this
would indicate that wealth and marital satisfaction are
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
14
15
ve
l
yc
ar
i. .
.
17
Po
sit
i
de
16
0%
or
. ..
0%
nt
v
nt
va
.. .
or
...
pe
Ne
g
13
0%
De
pe
n
3
In
de
2
at
iv
re
ly
sa
l
Ca
u
1
0%
el
yc
lat
...
0%
nd
e
1. Causally related
2. Negatively
correlated
3. Independent
variables
4. Dependent variables
5. Positively correlated
18
19
20
The belief that weather conditions signal
the onset of arthritis pain best illustrates
1.
2.
3.
4.
1
2
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
16
17
rc
on
f id
en
ce
0%
ov
e
An
illu
so
r
bo
pl
a
Th
e
15
0%
yc
o.
..
.. .
ds
igh
t
14
0%
ef
...
0%
hi
n
Ra
Th
e
nd
om
sa
m
pl
in
.. .
0%
ce
Random sampling
The hindsight bias
The placebo effect
An illusory
correlation
5. overconfidence
18
19
20
Which of the following correlation coefficients
expresses the strongest degree of relationship
between two variables?
1. +0.10
2. -.67
3. 0.00
4. -0.10
5. +0.59
1
2
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
14
15
16
17
0%
.5
9
0%
-.1
0%
0
0%
-.6
7
.1
0%
18
19
20
Experimental Method
• Manipulation of one
or more variables
•
• Laboratory v. Field
Experiments
Smoking causes health issues.
Theory
• An explanation that
organizes
observation and
predicts behaviors
or events
• A hunch
• Example
Hypothesis
• A testable
prediction.
• Example:
Independent Variable
• Whatever is being
manipulated in the
experiment.
If there is a drug in an
experiment, the drug is
almost always the
independent variable.
Example:
Dependent Variable
• Whatever is being
measured in the
experiment.
• It is dependent on the
independent variable.
The dependent variable
would be the effect
of the drug.
Example:
Operational Definitions
• A statement that
tells a person clearly
how to perform an
observation or
measure a variable
• Must be clear and
precise
• Must be measurable
• Allow experiment to
be replicated
• Example:
Beware of
Confounding Variables
• A factor other than the
independent variable
that might produce an
effect on the
experiment
• Confounds often arise
due to differences
between the groups that
exist before the
independent variable is
imposed!
Example:
motivation,
eagerness to
please, lifestyle,
age, intelligence….
*
Other confounding variables
in an experiment could be
experimenter bias or the
placebo effect but we’ll talk
about those later…
Assignment
• Random Assignment
Example:
– Assigning participants in an
experiment to experimental and
control groups by chance
– Minimizes differences between
two groups
– Different than Random Sample
– Reduces the impact of
confounding variables
• Experimental Group – the group
that’s exposed to the treatment
• Control Group – the group not
exposed to the treatment
Bias
• Experimenter Bias - expectations
by the experimenter that are
subtly communicated to the
participants
• Placebo effect – an experimental
effect caused by expectations of
participants or caused by a
substance which the recipient
assumes is the independent
variable but is not (ie. A drug
that has no effect is assumed to
be the drug that has an effect).
Double-blind Procedure
•
both the researcher
and the participants
are ignorant about who
is receiving the
treatment and who is
receiving a placebo ( a
dummy medication)
Replication
• Repeating the
research study to
see whether the
basic finding
extends to other
participants and
circumstances.
• What helps the
researcher to insure
the study can be
replicated?
Researchers use experiments rather than
other research methods in order to
distinguish between
1. Facts and theories
2. Cause and effects
3. Case studies and
surveys
4. Random samples and
representative samples
5. Hypotheses and
operational definitions
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
14
16
17
d.
..
an
es
es
sa
m
pl
sa
di
e
st
u
15
es
.. .
...
..
ef
fe
.
an
d
se
13
Hy
po
th
6
nd
om
5
Ra
4
Ca
se
3
Ca
u
2
Fa
c
1
ts
a
nd
th
eo
...
20% 20% 20% 20% 20%
18
19
20
Abdul has volunteered to participate in an experiment evaluating the
effectiveness of aspirin. Neither he nor the experimenters know whether the
pills he takes during the experiment contain aspirin or are merely placebos.
The investigators are apparently making use of
1. Naturalistic
observation
2. Illusory correlation
3. The double-blind
procedure
4. Random sampling
5. The overconfidence
effect
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
16
17
rc
on
ov
e
sa
m
pl
-b
bl
e
do
u
15
fid
.. .
in
.. .
..
li.
...
rre
co
14
18
Th
e
5
nd
om
4
Ra
3
Th
e
2
Ill
us
or
y
1
Na
t
ur
a
lis
t ic
o.
..
20% 20% 20% 20% 20%
19
20
In the hypothesis “Students who study a list of terms in the
morning, just after waking up, will recall more terms than
students who study the list just before falling asleep,” what is
the dependent variable?
2
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
14
15
16
Nu
m
1
be
r
st
ud
en
ts
of
t
er
m
sr
em
e.
..
io
n
Ti
M
em
or
iza
t
te
rm
of
Lis
t
m
eo
List of terms
Memorization
Time of day
Number of terms
remembered
5. students
s
1.
2.
3.
4.
fd
ay
20% 20% 20% 20% 20%
17
18
19
20
In a study conducted on depression, some patients were given
an anti-depressant and others were given a placebo to test
whether the drug reduced suicidal thoughts. What is the
independent variable in this study?
20% 20% 20% 20% 20%
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
16
17
18
nt
s
ie
Pa
t
ss
io
n
of
su
i
15
De
pr
e
c..
.
bo
ce
Pl
a
14
be
r
2
Nu
m
1
An
tid
ep
re
ss
an
. ..
1. Anti-depressant
2. Placebo
3. Number of suicidal
thoughts
4. Depression
5. Patients
19
20
Statistics
• Recording the
results from our
studies.
• Must use a common
language so we all
know what we are
talking about.
Descriptive Statistics
• Just describes sets
of data.
• You might create a
frequency
distribution.
• Examples:
Types of Data
• Nominal Data – identifies
categories
• Ordinal Data – identifies the
order in which data falls in a
set
• Interval data – data that
falls within a number line
with a zero point
• Ratio data – data that falls
in a number line where zero
is just another number
• Examples:
Measures of Central Tendency
• Mode (occurs the most)
• Mean (arithmetic average)
• Median (middle score)
Central Tendency
• Mean, Median and Mode.
Let’s look at the salaries of the
employees at Dunder Mifflen Paper
in Scranton:
$25,000-Pam
$25,000- Kevin
$25,000- Angela
$100,000- Andy
$100,000- Dwight
$200,000- Jim
$300,000- Michael
The median salary looks good at $_______________________
The mean salary also looks good at about $________________
But the mode salary is only $___________________________
Watch out for extreme scores or outliers.
Medium is a better measure than the mean when
there are extremes
Normal Distribution
• In a normal
distribution, the
mean, median and
mode are all the
same.
Distributions
• Outliers skew
distributions.
• If group has one high
score, the curve has a
positive skew
(contains more low
scores)
• If a group has a low
outlier, the curve has
a negative skew
(contains more high
scores)
Measures of variability
• Range: distance from highest
to lowest scores.
• Standard Deviation: the
variance of scores around the
mean.
• A higher the variance or SD,
means…
• Do scientists want a big or
small SD?
• Variance - The average of
the squared differences from
the mean – another measure of
how the data is distributed
around the mean
Shaq and Kobe may both
score 30 ppg (same mean).
But their SDs are very
different…meaning?
Calculating Standard Deviation
Step 1 – calculate the mean –
add all of the raw scores and
divide by the # of scores
Step 2 – calculate the deviation
from the mean by
subtracting each of the raw
scores from the mean
Step 3 – square the deviation
from the mean for each
score
• Step 4 – Sum the squared
deviations
• Step 5 – divide the sum of
the squared deviation by the
number of scores and find
the square root
Calculating the Standard
Deviation
Calculating the Standard Deviation
your turn
Scores – 10, 3, 7, 8, 7
Step 1 – calculate the mean – add
the all of the raw scores and
divide by the # of scores
Step 2 – calculate the deviation
from the mean by subtracting
each of the raw scores from the
mean
• Step 3 – square the deviation
from the mean for each score
• Step 4 – Sum the squared
deviations
• Step 5 – divide the sum of the
squared deviation by the number
of scores and find the square
root
Variance
• The average of the
squared differences
from the Mean.
• = Standard Deviation2
• Example:
Standard Deviation = 5
Variance = 25
• *if you know the variance, how
can you calculate the standard
deviation?
Normal Distribution
Normal Distribution
Calculating scores based on
a normal distribution.
Step 1 – calculate the mean
Step 2 - calculate/find the standard deviation
or variance. If you only have the variance you
must calculate the standard deviation
Step 3 draw a normal distribution curve and
find the scores for each standard deviation
from the mean and place them on the graph
Step 4: calculating the % of students who
scored within a range of scores by finding the
corresponding scores on the curve, then add
the percentages % from each standard
deviation.
Scores
• A unit that measures
the distance of a
score from the mean
in units of standard
deviations
• Equals 0 at the mean
• A positive z score
means a number above
the mean.
• A negative z score
means a number below
the mean.
Example: If John scored a 72 on a test
with a mean of 80 and a standard deviation
of 8, John’s z score would be -1
Inferential Statistics
•
The purpose is to discover whether the
finding can be applied to the larger
population from which the sample was
collected.
• Statistical Significance
• P-value= .05
– 5% likely the results are due to
chance
– psychologists like this because it
means that they can have a 95%
confidence level that there was a
reason for the differences and only
5% that they occurred by chance
•
•
What does a p-value= .80 mean?
Statistical Significance
Key Ideas
• The bigger the difference between groups the less likely it is
that it's due to chance.
• The larger the sample size (number of patients) the more likely
it is that the observed difference is close to the actual
difference. This is an example of the "law of large numbers."
– Ie. The smaller the real difference is, the more patients you need
to be likely to detect a statistically significant difference in a
clinical trial. The larger the real difference is, the fewer patients
you need to be likely to be detect a statistically significant
difference in an actual clinical trial.
• The larger the sample size, the smaller an observed difference
has to be in order to be statistically significant.
• The smaller the sample size, the larger an observed difference
would have to be in order to be statistically significant.
APA Ethical Guidelines for
Research
• IRB- Internal Review
Board
• Both for humans and
animals.
Animal Research
Human Research