Download Research Methods - Solon City Schools

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

History of statistics wikipedia, lookup

Regression toward the mean wikipedia, lookup

Bootstrapping (statistics) wikipedia, lookup

Misuse of statistics wikipedia, lookup

Psychometrics wikipedia, lookup

Categorical variable wikipedia, lookup

Transcript
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: After Lebron made his
decision to go to Miami, you say
it was obvious he was going
there—That’s where Wade went
Monday Morning
Quarterbacking!!!
Overconfidence
• We tend to think we
know more than we do.
• 82% of U.S. drivers consider
themselves to be in the top 30% of
their group in terms of safety.
• 81% of new business owners felt
they had an excellent chance of
their businesses succeeding. When
asked about the success of their
peers, the answer was only 39%.
(Now that's overconfidence!!!)
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
– Curious eagerness
– Skeptically scrutinize competing ideas
– Open-minded humility before nature
Critical Thinking
• Critical Thinking - thinking that does
not blindly accept arguments and
conclusions
– “Smart thinking”
– Four elements
• Examines assumptions
• Discerns hidden values
• Evaluates evidence
• Assesses conclusions
– 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, a
hypothesis is a testable prediction
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
3Types of Research
• Descriptive
To describe behavior
• Correlational
To predict behavior
Experimental
To explain behavior
3 Types of Descriptive Research
• The Case Study
• The Survey
• Naturalistic Observation
Case Studies
• A detailed picture of
or a few subjects.
one
– Sources of Info
• Direct Observation
• Interviews
• Documents
• Provide unique oppty to
study unusual cases
• Offers suggestions for
further study
• Results can’t be generalized
• Can’t establish cause and
effect
Example: The ideal case
study was John and Kate.
Really interesting, but are
they typical of all
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 chimpanzees
Survey Method
•Method for understanding the self
reported beliefs of a certain group
•personal facts, attitudes, past
behaviors and opinions.
•Questionnaire or Structured interview
•Works with large groups of data
•Can be Descriptive or Correlational
•Cheap and fast, most used research
method in psychology
Example: U.S. Census
Conducting a Survey
• Population - all the possible
subjects in a group you want to
study
• Example: men and women 18-80
• 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
representative of a large population
• Helps avoid false generalizations
• Example: Use of a random computer
generator to select 1000 people in the
phone book for a survey to be sent to.
Survey Method: The Bad
• Low Response
Rate
• Response Bias
- People Lie or
just
misinterpret
themselves.
• Wording
Effects
How accurate would a survey be about the
frequency of diarrhea?
Correlational Method
• Correlation - expresses a
relationship between two
variables.
– Measures how well one variable
predicts the other
• Can be efficient
• Can make predictions
• Can use pre-existing or
archival data
• Make it difficult to assess
the impact of a third variable
• Does not show causation
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
• The variables go in
the SAME direction.
Example: Studying and
grades hopefully has a
positive correlation.
Negative Correlation
• The variables go in
opposite directions.
Example: Heroin use and
grades probably has a
negative correlation.
Correlation Coefficient
A number that measures
the relationship between 2
variables.
• Range is from -1 to +1
•
– The relationship gets weaker
the closer you get to zero.
• 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
The table below lists the scores of eight research participants on a test to
measure anxiety, as well as the typical number of cigarettes each person
smokes daily. Scores on the anxiety test can range anywhere from a low of 0
(indicating very low anxiety) to a high of 30 (indicating very high anxiety).
Research
Participant
1
2
3
4
5
6
7
8
Anxiety
Cigarettes
Test Score____________
8
11
9
3
15
11
14
16
21
26
12
10
22
24
17
18
Construct a scatterplot to represent the correlation between smoking and
anxiety. Describe the direction of the correlation and give two possible
explanations for it.
Correlation
Illusory Correlations
• Illusory Correlation
–Perceived non-existent
correlation
–A random coincidence
• Speaking at a college graduation ceremony,
Professor Robson compared college graduates
with adults who are less educated. She
correctly noted that college graduates pay
more taxes, vote more frequently, engage in
more volunteer activities in their
communities, and are less likely to go to jail
than less-educated adults. The professor
concluded that colleges obviously do great
things for society. How might you reasonably
challenge the way the professor reached her
conclusion?
Psychologists who carefully watch the behavior
of chimpanzee societies in the jungle are using a
research method known as
96%
0%
in
.. .
y
pl
sa
m
nd
om
Ra
ca
se
st
ud
o.
..
Th
e
lis
t ic
en
ta
t
im
ur
a
y
su
rv
e
Ex
pe
r
Th
e
4%
0%
io
. ..
0%
Na
t
1. The survey
2. Experimentation
3. Naturalistic
observation
4. The case study
5. Random sampling
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
96%
or
. ..
0%
ve
l
yc
ar
i. .
.
Po
sit
i
de
nd
e
pe
0%
nt
v
nt
va
.. .
or
...
In
de
Ne
g
at
iv
re
ly
sa
l
Ca
u
0%
el
yc
lat
...
4%
De
pe
n
1. Causally related
2. Negatively
correlated
3. Independent
variables
4. Dependent variables
5. Positively correlated
The belief that weather conditions signal
the onset of arthritis pain best illustrates
100%
1.
2.
3.
4.
rc
on
f id
yc
o.
..
ov
e
An
illu
so
r
bo
pl
a
0%
en
ce
0%
ef
...
.. .
ds
igh
t
hi
n
Th
e
0%
Th
e
Ra
nd
om
sa
m
pl
in
.. .
0%
ce
Random sampling
The hindsight bias
The placebo effect
An illusory
correlation
5. overconfidence
Which of the following correlation coefficients
expresses the strongest degree of relationship
between two variables?
1. +0.10
84%
2. -.67
3. 0.00
4. -0.10
5. +0.59
8%
.5
9
0%
-.1
0
4%
-.6
7
.1
4%
Experimental Method
• Manipulation of one or more
variables
• Shows cause and effect
• Have high internal validity –
confidence that IV
influences DV
• Replication allows for
verification of results
• Can eliminate bias through
double-blind procedures
• Confounding variables
• Time and $
Smoking causes health issues.
Theory
• An explanation that
organizes
observation and
predicts behaviors
or events
• A hunch
• Example:
– Participating in class
leads to better AP
scores
Hypothesis
• A testable
prediction.
• Example:
Students who
participate in AP
Psych class will score
higher on the AP
exam
Operational Definitions
• A statement that tells
a person clearly
• Meaning of hypothesis
• how to perform an
observation
• measure a variable
• Must be clear and
precise
• Must be measurable
• Allow experiment to
be replicated
• Example:
Participation is the
number of times a
person raises their hand
in class to answer or ask
a question about AP
Psych content
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:
Participation
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:
•Score on the AP Exam
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!
Examples:
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
– 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
• Examples:
Experimental - The group
that gets to participate
Control group – The group
that does not get to
participate
2 other confounding variables
• 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)
• Minimizes placebo effect and
experimenter bias
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
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
14
16
17
an
d.
..
0%
es
es
sa
m
pl
sa
di
e
st
u
15
es
.. .
...
..
ef
fe
.
an
d
se
13
0%
Hy
po
th
5
0%
nd
om
4
0%
Ra
3
Ca
u
2
Fa
c
1
ts
a
nd
th
eo
...
0%
Ca
se
1. Facts and theories
2. Cause and effects
3. Case studies and
surveys
4. Random samples and
representative samples
5. Hypotheses and
operational definitions
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
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
16
17
fid
.. .
rc
on
ov
e
sa
m
pl
-b
bl
e
do
u
15
0%
in
.. .
..
li.
...
rre
co
14
0%
18
Th
e
4
0%
nd
om
3
0%
Th
e
2
Ill
us
or
y
1
Na
t
ur
a
lis
t ic
o.
..
0%
Ra
1. Naturalistic
observation
2. Illusory correlation
3. The double-blind
procedure
4. Random sampling
5. The overconfidence
effect
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?
1.
2.
3.
4.
0%
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
of
t
er
m
sr
st
ud
en
ts
io
n
Ti
0%
em
e.
..
0%
fd
ay
0%
or
iza
t
M
em
Lis
t
of
te
rm
s
0%
m
eo
List of terms
Memorization
Time of day
Number of terms
remembered
5. students
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?
3
4
5
6
7
8
9
10
21
22
23
24
25
26
27
28
29
30
11
12
13
15
16
18
ie
Pa
t
De
pr
e
17
0%
nt
s
0%
ss
io
n
0%
c..
.
bo
ce
Pl
a
14
be
r
2
0%
Nu
m
1
An
tid
ep
re
ss
an
. ..
0%
of
su
i
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:
• Frequency polygons,
bargraphs or
histograms.
Types of Data
• Examples: Yes/No answers,
class level in school (Sr.,
Junior etc.)
• Ordinal Data – identifies the • Any Ranking of items –
class rank, music hits
order in which data falls in a
set
• Nominal Data – identifies
categories
• Interval data – data that
falls within a number line
with a zero starting point
• Weight and Height
• Ratio data – data that falls
in a number line where zero
is just another number
• Temperature
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 $_______________________ 100,000
The mean salary also looks good at about $________________ 110,000
But the mode salary is only $___________________________ 25,000
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.
Examples:
Height, Weight, IQ Scores
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 If most students scored well on a
scores)
test, what would the distribution look
like?
If most students scored poorly?
What does the data tell us?
Measures of variability
• Range: distance from highest
to lowest scores.
• Standard Deviation: the
variance of scores around the
mean.
• The higher the variance or SD,
the more spread out the
distribution is.
• 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
35/5 = 7
10-7 = +3, 3-7 = -4, 7-7 =0,
8-7= +1, 7-7=0
3x3=9, 4x4=16, 0x0=0,
1x1=1, 0x0=0
9+16+0+1+0 = 26
26/5 = 2.28
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?
Scores
• A unit that measures the
distance of a score from
the mean in units of
standard deviations
• Observation – Mean
Standard Deviation
• 10-15 = -1
5
• Equals 0 at the mean
• A positive z score =
number above the mean,
negative z score = 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
Normal Distribution
What is the probability an observation is less than the
z score or more than the z score?
Normal Distribution
Calculating the % of the observed
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.
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 – the
observed difference between the
means of the experimental and
control group are not due to
chance
• Measured by P-value= .05
– 5% likely the results are due to chance
or
– 95% confidence level the results are
due to the independent variable
– You can apply the findings to the
population
•
•
What does a p-value= .80 mean?
that you are only 20% confident the difference in
means is due to the independent variable
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
• Clear purpose
• Treated in a humane
way
• Acquire animals
legally
• Least amount of
suffering possible.
Human Research
•
•
•
•
Informed consent
Confidentiality
No harm
Must debrief
Milgram’s Experiment
• “teachers” were
ordered to shock
“learners” for wrong
answers
• Criticized for
deceiving “teachers”
and subjecting them
to stress
• 65% of participants
followed the orders
to the maximum
shock level
Milgram’s Experiment
• Obedience High
– Experimenter in close
proximity to
“teacher”
– Learner placed in a
different room
– Experiment
associated with
prestigious location
(Yale)
• Obedience Lower
– Experimenter out of
the room giving
orders over a phone
– Other “teachers”
observed other
participants refuse
orders
– Experiment not
associated with
prestigious location