Social Science Reasoning Using Statistics

Download Report

Transcript Social Science Reasoning Using Statistics

• Exam 1 next Wednesday in both lecture & lab
– Pt 1: In lecture: closed book, scantron – bring #2 pencil &
know your Univ. ID#
– Pt 2: In labs: open book, open notes (no collaboration
though)
Annoucements
Reasoning in Psychology
Using Statistics
Where does “probability”
fit in?
• Randomness in
sampling leads to
variability in sampling
error
• “Randomness” in short
run is unpredictable but
in long run is
predictable
• Allows predictions
about likelihood of
getting particular
samples
Population
Possible Samples
Inferential statistics
Reasoning in Psychology
Using Statistics
• Draw lettered tiles from bag
– Bag contains:
• A’s B’s and C’s.
• Both upper and lower case letters
b
c
a
A
• What’s the probability of getting an A
(upper or lower case)?
Prob. of A = p(A) =
Total number of outcomes classified as A
Total number of possible outcomes
(Sample space)
Basics of probability:
Derived from games with all outcomes known
Reasoning in Psychology
Using Statistics
C
B
= ??
• What’s the probability of getting an A (upper or lower
case) on the first pick and another on a second pick?
First Pick: Prob. of A = p(A) = 2/6 = 0.33
a
A
1/6
+
b
c
a
A
C
B
2/6 on 1st pick
1/6
Second Pick: ? – it depends on how you sample
Sampling with replacement
b
c
a
A
2/6
b
C
B
Sampling without replacement
c
a
b
C
B
1/5
Basics of probability
Reasoning in Psychology
Using Statistics
c
C
A
B
1/5
The
probabilities
of selecting
the titles
change from
1st to 2nd
pick
b
• What’s the probability of getting an A (upper or lower
case) on the first pick and another on a second pick?
a
A
Sampling with replacement
Sampling without replacement
1st picks
1st picks
2nd picks
2nd picks
a
a
A
b
B
c
C
a
A
a
A
b
B
c
C
A
a
b
a
A
b
B
c
C
b
a
A
B
a
A
b
B
c
C
B
a
A
b
c
a
A
b
B
c
C
c
a
A
b
B
C
a
A
b
B
c
C
C
a
A
b
B
A
b
B
c
C
b
B
c
C
B
c
C
c
C
C
c
36 total outcomes
4 outcomes of 2 A’s
30 total outcomes
2 outcomes of 2 A’s
Probability?
Probability?
Basics of probability
Reasoning in Psychology
Using Statistics
c
C
B
• Most statistical procedures assume sampling with replacement
• For large populations it turns out not to matter much
•
Example?
• Sampling with replacement can also be a disadvantage in experiments –
•
Why?
Basics of probability
Reasoning in Psychology
Using Statistics
The researchers used a questionnaire to interview a random sample
of 351 people ages 20 to 80 in the Vancouver, British Columbia, area,
asking about their dog ownership and level of physical activity.
Dog owners walked almost twice as many minutes per week as
people who did not own dogs, and the dog owners spent more time in all
forms of mild and moderate physical activity.
People who own dogs
appear to get more exercise
than those who do not …
Neither the sex of the owner nor the size of the dog made a
difference: men and women who owned dogs participated in increased
amounts of moderate exercise compared with those who had no dog. But
no statistically significant difference between the two groups was
evident in the amount of strenuous exercise performed.
The authors concede that the study, published in the February issue
of The American Journal of Preventive Medicine, does not determine
whether owning a dog itself makes people exercise more.
Shane Brown, the study's lead author and a graduate student at the
Behavioral Medicine Laboratory of the University of Victoria, called this
"the million-dollar question."
When the researchers subtracted the time spent walking the dog
from total physical activity, dog owners actually walked less than their
counterparts without dogs.
This, the researchers say, may suggest that the dog makes the
difference. When dog owners choose to engage in moderate physical
exercise, the reason may be that a four-legged member of the household
is insisting on a walk. Now.
Report (NY Times 2006): dog a more reliable exercise partner than a human.
In the news
Reasoning in Psychology
Using Statistics
• Research process
Variables
– Ask research question
– Operational level
Regimens:
– Identify variables &
– Conceptual
level Pounding the
formulate hypothesis
Pavement, Stopping at
Fire Hydrants
– Define your population
Variables
– Select a research
The researchers …, asking about their dog
–
Independent
(explanatory) variables
ownership and level of physical activity.
methodology
– Dependent (response) variable
Methods
– Collect your data from a
–
Control
variables
…. Questionnaires were mailed out in 2004 to
sample
collect information about demographics, dog
– Random
variables
– Analyze
your data
ownership, leisure-time walking, physical
– Confound
variables
activity levels, and theory of planned behavior
–
Draw
conclusions based on
(TPB) constructs.
your data
– Repeat
In the news
Reasoning in Psychology
Using Statistics
• Research process
– Ask research question
Regimens: Pounding the
– Identify variables &
formulate hypothesis
Pavement, Stopping at
Instrument used?
Fire Hydrants
– Define your population
Scales of measurement
– Select a research
The researchers …, asking about their dog
–
Nominal
ownership and level of physical activity.
methodology
– –Ordinal
Methods
Collect your data from a
…. Questionnaires were mailed out in 2004 to
– Interval
sample
collect information about demographics, dog
– –Ratio
Analyze your data
ownership, leisure-time walking, physical
activity levels, and theory of planned behavior
– Draw conclusions based on
(TPB) constructs.
data
– Repeat
In the news
Reasoning in Psychology
Using Statistics
• Research process
– Ask the research question
Is the sample representative?
Regimens: Pounding the
– Identify variables &
Is there
Pavement, Stopping
at bias in sample?formulate hypothesis
Fire Hydrants
– Define population
The researchers used a questionnaire to
– Select research methodology
interview a random sample of 351 people ages
– Collect data from sample
20 to 80 in the Vancouver, British Columbia,
– Analyze data
area, (i.e., mailed out questionnaires)
– Draw conclusions based on
data
Was it simple random sampling?
Stratified random sampling?
– Repeat
Etc.?
In the news
Reasoning in Psychology
Using Statistics
• Research process
– Ask research question
Regimens: Pounding the
– Identify variables &
formulate hypothesis
Pavement,Causal
Stopping
at
Claims??
Fire Hydrants
– Define population
The researchers used a questionnaire to
– Select research methodology
interview a random sample of 351 people ages
– Collect data from sample
20 to 80 in the Vancouver, British Columbia,
– Research
Analyze your
data
Which
Method??
area,
The authors concede that the study… does
Draw conclusions
– –Observational
studybased on
not determine whether owning a dog itself
your data
– Experimental methodology
makes people exercise more.
– Repeat
This may suggest that .. when dog owners
– Quasi-experimental
choose to engage in moderate physical
methodology
exercise, the reason may be that [the dog] is
insisting on a walk.
In the news
Reasoning in Psychology
Using Statistics
• Research process
– Ask research question
Regimens: Pounding the
– Identify variables &
formulate hypothesis
Pavement, Stopping at
Fire Hydrants
– Define population
– Select research methodology
Possible confounds?
– Collect data from sample
Threats to internal validity?
– Analyze your data
If follow-up experiment were
– Draw conclusions based on
designed,
yourwhat
data would you do?
– Repeat
In the news
Reasoning in Psychology
Using Statistics
• Scientific method
• Research methods
– Observation methods
– Experimental methods
– Quasi-experimental (see Course
Packet)
• Pop vs. Sample
• Variables
– Types
– Operational definitions
• Measurements
– Continuous and discrete
– Scales of measurement
• Errors in Measurement
– Validity & Reliability
• Dartboard example
– Improving validity
• Matching design (see CP)
• Double-blind design (CP)
• Block design (CP)
• Sampling
– Sampling error
– Statistics and parameters
– Sampling Methods
• Probability Sampling (3 types)
• Non-probability Sampling (2
types)
• Basic probability
Reviewing Producing Data
Reasoning in Psychology
Using Statistics