Documente Academic
Documente Profesional
Documente Cultură
Compu
ters II
Slide 2
Compu
ters II
Slide 3
Thus, there are three equations, one for each of the groups
defined by the dependent variable.
Compu
ters II
Slide 4
Compu
ters II
Slide 5
Compu
ters II
Slide 6
Compu
ters II
Slide 7
Compu
ters II
Slide 8
The significance test for the final model chi-square (after the
independent variables have been added) is our statistical
evidence of the presence of a relationship between the
dependent variable and the combination of the independent
variables.
Compu
ters II
Slide 9
-2 Log
Likelihood
284.429
265.972
Chi-Square
18.457
df
Sig.
6
.005
ters II
Slide
10
ters II
Slide
11
ters II
Slide
12
1
2
3
62
93
12
167
103
270
153a
Marginal
Percentage
37.1%
55.7%
7.2%
100.0%
ters II
Slide
13
1
15
7
5
16.2%
2
47
86
7
83.8%
3
0
0
0
.0%
Percent
Correct
24.2%
92.5%
.0%
60.5%
ters II
Slide
14
Numerical problems
ters II
Slide
15
ters II
Slide
16
ters II
Slide
17
Parameter Estimates
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
HIGHWAYS
a
AND BRIDGES
TOO LITTLE
ABOUT RIGHT
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
2.478
1.709
1
.191
The
reference
category
plays
the
same
role in
.019
.020
.906
1
.341
multinomial logistic regression that it plays in
.071
.108
.427
1 variable:
.514it is
the dummy-coding
of a nominal
-1.373
.620 that4.913
1 with .027
the category
would be coded
zeros
for all of
the dummy-coded
variables
that
3.639
2.456
2.195
1
.138 all
other categories are interpreted against.
.003
.020
.017
1
.897
.172
.110
2.463
1
.117
-1.657
.613
7.298
1
.007
1.019
1.073
.253
1.003
1.188
.191
95% C
Lower B
ters II
Slide
18
Effect
Intercept
AGE
EDUC
CONLEGIS
-2 Log
Likelihood of
Reduced
Model
268.323
268.625
270.395
275.194
Chi-Square
2.350
2.652
4.423
9.221
df
2
2
2
2
Sig.
.309
.265
.110
.010
HIGHWAYS
a
AND BRIDGES
1
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
1.019
1.073
.253
.980
.868
.075
1.0
1.3
.8
1.003
1.188
.191
.963
.958
.057
1.0
1.4
.6
ters II
Interpreting relationship of individual independent
variables to the dependent variable
Slide
19
Effect
Intercept
AGE
EDUC
CONLEGIS
-2 Log
Likelihood
of respondents who had less confidence in congress (higher
Survey
values correspond to lower confidence) were less likely to be in the
Reduced
group ofChi-Square
survey respondents
who
thought we spend too little money
Model
df
Sig.
on
highways
and
bridges
(DV
category
268.323
2.350
2
.309 1), rather than the group of
survey respondents who thought we spend too much money on
268.625
.265
highways and2.652
bridges (DV 2category
3).
270.395
4.423
2
.110
For each unit9.221
increase in confidence
in Congress, the odds of being
275.194
2
.010
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
Sig.
.191
.341
.514
.027
.138
.897
.117
.007
Exp(B)
1.019
1.073
.253
.980
.868
.075
1.0
1.3
.8
1.003
1.188
.191
.963
.958
.057
1.0
1.4
.6
ters II
Interpreting relationship of individual independent
variables to the dependent variable
Slide
20
Effect
Intercept
AGE
EDUC
CONLEGIS
-2 Log
Likelihood of
Reduced
Model
268.323
268.625
270.395
275.194
Chi-Square
2.350
2.652
4.423
9.221
df
2
2
2
2
Sig.
.309
.265
.110
.010
HIGHWAYS
a
AND BRIDGES
1
B increase
Std. Error
Wald
df
Sig.odds of
Exp(B)
For each unit
in confidence
in Congress,
the
being
Intercept
in the group
of survey
respondents
we spend
about the
3.240
2.478
1.709 who thought
1
.191
of money
on highways
decreased
by1.019
AGE right amount
.019
.020
.906 and bridges
1
.341
80.9%. (0.191 1.0 = 0.809)
EDUC
.071
.108
.427
1
.514
1.073
CONLEGIS
-1.373
.620
4.913
1
.027
.253
Intercept
3.639
2.456
2.195
1
.138
AGE
.003
.020
.017
1
.897
1.003
EDUC
.172
.110
2.463
1
.117
1.188
CONLEGIS
-1.657
.613
7.298
1
.007
.191
1.0
1.3
.8
.963
.958
.057
1.0
1.4
.6
ters II
Slide
21
Effect
Intercept
AGE
EDUC
POLVIEWS
SEX
-2 Log
Likelihood of
Reduced
Model
327.463a
333.440
329.606
334.636
338.985
Chi-Square
.000
5.976
2.143
7.173
11.521
df
Sig.
0
2
2
2
2
.
.050
.343
.028
.003
1.
1.
.
.
1.
1.
1.
1.
ters II
Slide
22
Effect
Intercept
AGE
EDUC
POLVIEWS
SEX
-2 Log
Likelihood of
Reduced
Model
Chi-Square
df
Sig.
327.463a
.000
0
.
333.440
5.976who were2male (code
.050 1 for sex) were less likely
Survey
respondents
to 329.606
be in the group
of
survey
respondents
2.143
2
.343 who thought we spend too
little money on childcare assistance (DV category 1), rather than the
334.636
2
.028 we spend too much
group
of survey 7.173
respondents who
thought
money
on childcare
3).
338.985
11.521assistance2 (DV category
.003
1.
1.
.
.
.965
.824
.406
.040
.
1.
1.
1.
1.
ters II
Slide
23
ters II
Slide
24
Problem 1
11. In the dataset GSS2000, is the following statement true, false, or an incorrect application of
a statistic? Assume that there is no problem with missing data, outliers, or influential cases, and
that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "age" [age], "highest year of school completed" [educ] and "confidence in Congress"
[conlegis] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on highways and bridges" [natroad]. These predictors differentiate
survey respondents who thought we spend too little money on highways and bridges from survey
respondents who thought we spend too much money on highways and bridges and survey
respondents who thought we spend about the right amount of money on highways and bridges
from survey respondents who thought we spend too much money on highways and bridges.
Among this set of predictors, confidence in Congress was helpful in distinguishing among the
groups defined by responses to opinion about spending on highways and bridges. Survey
respondents who had less confidence in congress were less likely to be in the group of survey
respondents who thought we spend too little money on highways and bridges, rather than the
group of survey respondents who thought we spend too much money on highways and bridges.
For each unit increase in confidence in Congress, the odds of being in the group of survey
respondents who thought we spend too little money on highways and bridges decreased by
74.7%. Survey respondents who had less confidence in congress were less likely to be in the
group of survey respondents who thought we spend about the right amount of money on
highways and bridges, rather than the group of survey respondents who thought we spend too
much money on highways and bridges. For each unit increase in confidence in Congress, the odds
of being in the group of survey respondents who thought we spend about the right amount of
money on highways and bridges decreased by 80.9%.
1.
2.
3.
4.
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
25
Dissecting problem 1 - 1
11. In the dataset GSS2000, is the following statement true, false, or an incorrect application of a
statistic? Assume that there is no problem with missing data, outliers, or influential cases, and
that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "age" [age], "highest year of school completed" [educ] and "confidence in Congress"
[conlegis] were useful predictors for distinguishing between groups based on responses to "opinion
about spending on highways and bridges" [natroad]. These predictors differentiate survey
respondents who thought we spend tooFor
little
these
money
problems,
on highways
we willand bridges from survey
respondents who thought we spend tooassume
much money
on
highways
and bridges and survey
that there is no problem
respondents who thought we spend about
the
right
amount
of
money
with missing data, outliers, or on highways and bridges
from survey respondents who thought we
spend too
much
money
on highways and bridges.
influential
cases,
and
that the
validation
analysis
confirm
Among this set of predictors, confidence
in Congress
was will
helpful
in distinguishing among the
the
generalizability
of
the
groups defined by responses to opinion about spending on highways and bridges. Survey
respondents who had less confidence inresults
congress were less likely to be in the group of survey
respondents who thought we spend too little money on highways and bridges, rather than the
group of survey respondents who thought
we spend
toowe
much
In this
problem,
aremoney
told to on highways and bridges. For
each unit increase in confidence in Congress,
theasodds
offor
being
use 0.05
alpha
the in the group of survey respondents
who thought we spend too little moneymultinomial
on highwayslogistic
and bridges
decreased by 74.7%. Survey
regression.
respondents who had less confidence in congress were less likely to be in the group of survey
respondents who thought we spend about the right amount of money on highways and bridges,
rather than the group of survey respondents who thought we spend too much money on highways
and bridges. For each unit increase in confidence in Congress, the odds of being in the group of
survey respondents who thought we spend about the right amount of money on highways and
bridges decreased by 80.9%.
1.
2.
3.
4.
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
26
Dissecting problem 1 - 2
The variables listed first in the problem
statement are the independent variables
(IVs): "age" [age], "highest year of school
[educ]
and "confidence
in statement true, false, or an incorrect application of
11. Incompleted"
the dataset
GSS2000,
is the following
a statistic?
Assume
that there is no problem with missing data, outliers, or influential cases, and
Congress"
[conlegis].
that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "age" [age], "highest year of school completed" [educ] and "confidence in
Congress" [conlegis] were useful predictors for distinguishing between groups based on
responses to "opinion about spending on highways and bridges" [natroad]. These predictors
differentiate survey respondents who thought we spend too little money on highways and
bridges from survey respondents who thought we spend too much money on highways and
bridges and survey respondents who thought we spend about the right amount of money on
variable
used
to define
highways andThe
bridges
from
survey
respondents who thought we spend too much money on
highways andgroups
bridges.
is the dependent
variable (DV): "opinion about
Among this set of predictors, confidence in Congress was helpful in distinguishing among the
spending on highways and
groups defined
by responses to opinion about spending on highways and bridges. Survey
bridges"
respondents who had [natroad].
less confidence in congress were less likely to be in the group of survey
respondents who thought we spend too little money on highways and bridges, rather than the
group of survey respondents who thought we spend too much money on highways and bridges.
For each unit increase in confidence in Congress, the odds of being in the group of survey
only supports
directdecreased
or
respondents who thought we spend too little money SPSS
on highways
and bridges
by
simultaneous
entry
of
independent
74.7%. Survey respondents who had less confidence in congress were less likely to be in the
variables
multinomial
logistic
group of survey respondents who thought we spend about
the in
right
amount of
money on
regression,
so
we
have
no
choice
of
highways and bridges, rather than the group of survey respondents who thought
we spend
too
variables.
much money on highways and bridges. For each unit method
increasefor
in entering
confidence
in Congress, the odds
of being in the group of survey respondents who thought we spend about the right amount of
money on highways and bridges decreased by 80.9%.
ters II
Slide
27
Dissecting problem 1 - 3
SPSS multinomial logistic regression models the relationship by
comparing each of the groups defined by the dependent variable to the
group with the highest code value.
survey
respondents
who
thought
wethought
spend about
the about
right the right amount of
the odds of being in the group of survey respondents
who
we spend
amount
of decreased
money versus
survey respondents who thought we
money on highways and
bridges
by 80.9%.
spend too much money on highways and bridges.
ters II
Slide
28
Dissecting problem 1 - 4
responses to "opinion about spending on highways and bridges" [natroad]. These predictors
problem
identifies
a difference
forspend
both of
the
comparisons
differentiate This
survey
respondents
who
thought we
too
little
money on highways and
groups modeled
the multinomial
regression.
bridges from among
survey respondents
whobythought
we spendlogistic
too much
money on highways and
bridges and survey respondents who thought we spend about the right amount of money on
highways and bridges from survey respondents who thought we spend too much money on
highways and bridges.
Among this set of predictors, confidence in Congress was helpful in distinguishing among the
groups defined by responses to opinion about spending on highways and bridges. Survey
respondents who had less confidence in congress were less likely to be in the group of
survey respondents who thought we spend too little money on highways and bridges, rather
than the group of survey respondents who thought we spend too much money on highways
and bridges. For each unit increase in confidence in Congress, the odds of being in the
group of survey respondents who thought we spend too little money on highways and
bridges decreased by 74.7%. Survey respondents who had less confidence in congress were
less likely to be in the group of survey respondents who thought we spend about the right
amount of money on highways and bridges, rather than the group of survey respondents
who thought we spend too much money on highways and bridges. For each unit increase in
confidence in Congress, the odds of being in the group of survey respondents who thought
we spend about the right amount of money on highways and bridges decreased by 80.9%.
ters II
Slide
29
Dissecting problem 1 - 5
11. In the dataset GSS2000, is the following statement true, false, or an incorrect application of a
statistic? Assume that there is no problem with missing data, outliers, or influential cases, and that the
validation analysis will confirm the generalizability of the results. Use a level of significance of 0.05 for
evaluating the statistical relationships.
The variables "age" [age], "highest year of school completed" [educ] and "confidence in Congress"
[conlegis] were useful predictors for distinguishing between groups based on responses to "opinion about
spending on highways and bridges" [natroad]. These predictors differentiate survey respondents who
thought we spend too little money on highways and bridges from survey respondents who thought we
spend too much money on highways and bridges and survey respondents who thought we spend about the
right amount of money on highways and bridges from survey respondents who thought we spend too much
money on highways and bridges.
Among this set of predictors, confidence in Congress was helpful in distinguishing among the groups
defined by responses to opinion about spending on highways and bridges. Survey respondents who had less
confidence in congress were less likely to be in the group of survey respondents who thought we spend too
little money on highways and bridges, rather than the group of survey respondents who thought we spend
too much money on highways and bridges. For each unit increase in confidence in Congress, the odds of
being in the group of survey respondents who thought we spend too little money on highways and bridges
decreased by 74.7%. Survey respondents who had less confidence in congress were less likely to be in the
group of survey respondents who thought we
about
right amount
of money
on highways and
In spend
order for
thethe
multinomial
logistic
regression
bridges, rather than the group of survey respondents who thought we spend too much money on highways
question
be true,the
theodds
overall
relationship
must
and bridges. For each unit increase in confidence
in to
Congress,
of being
in the group
of survey
be statistically
there
must and
be no
respondents who thought we spend about the
right amountsignificant,
of money on
highways
bridges decreased
by 80.9%.
evidence of numerical problems, the classification
accuracy rate must be substantially better than
could be obtained by chance alone, and the
stated individual relationship must be statistically
significant and interpreted correctly.
ters II
Slide
30
ters II
Slide
31
ters II
Slide
32
In this analysis, there are no nonmetric independent variables. Nonmetric independent variables would be
moved to the Factor(s) list box.
ters II
Slide
33
ters II
Slide
34
Third, click
on the
Continue
button to
complete the
request.
ters II
Slide
35
Click on the OK
button to request
the output for the
multinomial logistic
regression.
ters II
Slide
36
LEVEL OF MEASUREMENT - 1
11. In the dataset GSS2000, is the following statement true, false, or an incorrect application of
a statistic? Assume that there is no problem with missing data, outliers, or influential cases, and
that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "age" [age], "highest year of school completed" [educ] and "confidence in Congress"
[conlegis] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on highways and bridges" [natroad]. These predictors differentiate
survey respondents who thought we spend too little money on highways and bridges from survey
respondents who thought we spend too much money on highways and bridges and survey
respondents who thought we spend about the right amount of money on highways and bridges
from survey respondents who thought we spend too much money on highways and bridges.
Among this set of predictors, confidence in Congress was helpful in distinguishing among the
groups defined by responses to opinion about spending on highways and bridges. Survey
respondents who had less confidence in congress were less likely to be in the group of survey
respondents who thought we spend too little money on highways and bridges, rather than the
Multinomial
logistic
regression
requires
that money
the
group of survey respondents
who
thought
we spend
too much
on highways and bridges.
dependent variable be non-metric and the
For each unit increase
in confidence
in Congress,
of being in the group of survey
independent
variables
be metricthe
or odds
dichotomous.
respondents who thought we spend too little money on highways and bridges decreased by
74.7%. Survey respondents
hadspending
less confidence
in congress
were less likely to be in the
"Opinionwho
about
on highways
and
group of survey respondents
who thought
we spend
about the
the nonright amount of money on
bridges" [natroad]
is ordinal,
satisfying
highways and bridges,
rather
than
the group of survey
respondents
metric
level
of measurement
requirement
for the who thought we spend too
much money on highways
and variable.
bridges. For each unit increase in confidence in Congress, the odds
dependent
of being in the group of survey respondents who thought we spend about the right amount of
contains
three
categories:
survey respondents
money on highwaysItand
bridges
decreased
by 80.9%.
who thought we spend too little money, about
1. True
the right amount of money, and too much money
on highways and bridges.
2. True with caution
ters II
Slide
37
LEVEL OF MEASUREMENT - 2
"Age" [age] and "highest year of
school completed" [educ] are interval,
the metric
or dichotomous
11. satisfying
In the dataset
GSS2000,
is the following statement true, false, or an incorrect application of a
level of
measurement
requirement
for with missing data, outliers, or influential cases, and that
statistic?
Assume
that there
is no problem
variables.
the independent
validation analysis
will confirm the generalizability of the results. Use a level of significance of
The variables "age" [age], "highest year of school completed" [educ] and "confidence in
Congress" [conlegis] were useful predictors for distinguishing between groups based on responses
to "opinion about spending on highways and bridges" [natroad]. These predictors differentiate
survey respondents who thought we spend too little money on highways and bridges from survey
respondents who thought we spend too much money on highways and bridges and survey
respondents who thought we spend about the right amount of money on highways and bridges from
survey respondents who "Confidence
thought we spend
too much
money on
in Congress"
[conlegis]
is highways
ordinal, and bridges.
thought we spend too little money on highways and bridges decreased by 74.7%. Survey respondents
who had less confidence in congress were less likely to be in the group of survey respondents who
thought we spend about the right amount of money on highways and bridges, rather than the group
of survey respondents who thought we spend too much money on highways and bridges. For each
unit increase in confidence in Congress, the odds of being in the group of survey respondents who
thought we spend about the right amount of money on highways and bridges decreased by 80.9%.
ters II
Slide
38
1
2
3
62
93
12
167
103
270
153a
Marginal
Percentage
37.1%
55.7%
7.2%
100.0%
Multinomial logistic
regression
requires that the minimum ratio
in 146
(95.4%) subpopulations.
of valid cases to independent variables be at least 10 to 1. The
ratio of valid cases (167) to number of independent variables
(3) was 55.7 to 1, which was equal to or greater than the
minimum ratio. The requirement for a minimum ratio of cases
to independent variables was satisfied.
The preferred ratio of valid cases to independent variables is
20 to 1. The ratio of 55.7 to 1 was equal to or greater than the
preferred ratio. The preferred ratio of cases to independent
variables was satisfied.
ters II
Slide
39
-2 Log
Likelihood
284.429
265.972
Chi-Square
18.457
df
Sig.
6
.005
ters II
Slide
40
NUMERICAL PROBLEMS
Parameter Estimates
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
ters II
Slide
41
Effect
Intercept
AGE
EDUC
CONLEGIS
-2 Log
Likelihood of
Reduced
Model
268.323
268.625
270.395
275.194
Chi-Square
2.350
2.652
4.423
9.221
df
2
2
2
2
Sig.
.309
.265
.110
.010
ters II
Slide
42
Parameter Estimates
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
Sig.
.191
.341
.514
.027
.138
.897
.117
.007
Exp(B)
95% Confiden
Exp
Lower Bound
1.019
1.073
.253
.980
.868
.075
1.003
1.188
.191
.963
.958
.057
ters II
Slide
43
Parameter Estimates
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
Sig.
.191
.341
.514
.027
.138
.897
.117
.007
The value of Exp(B) was 0.253 which implies that for each unit
increase in confidence in Congress the odds decreased by 74.7%
(0.253 - 1.0 = -0.747).
The relationship stated in the problem is supported. Survey
respondents who had less confidence in congress were less likely
to be in the group of survey respondents who thought we spend
too little money on highways and bridges, rather than the group of
survey respondents who thought we spend too much money on
highways and bridges. For each unit increase in confidence in
Congress, the odds of being in the group of survey respondents
who thought we spend too little money on highways and bridges
decreased by 74.7%.
Exp(B)
95% Confiden
Exp
Lower Bound
1.019
1.073
.253
.980
.868
.075
1.003
1.188
.191
.963
.958
.057
ters II
Slide
44
Parameter Estimates
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
Sig.
.191
.341
.514
.027
.138
.897
.117
.007
Exp(B)
95% Confiden
Exp
Lower Bound
1.019
1.073
.253
.980
.868
.075
1.003
1.188
.191
.963
.958
.057
ters II
Slide
45
95% Con
HIGHWAYS
a
AND BRIDGES
1
Intercept
AGE
EDUC
CONLEGIS
Intercept
AGE
EDUC
CONLEGIS
B
3.240
.019
.071
-1.373
3.639
.003
.172
-1.657
Std. Error
2.478
.020
.108
.620
2.456
.020
.110
.613
Wald
1.709
.906
.427
4.913
2.195
.017
2.463
7.298
df
1
1
1
1
1
1
1
1
Sig.
.191
.341
.514
.027
.138
.897
.117
.007
The value of Exp(B) was 0.191 which implies that for each unit increase in
confidence in Congress the odds decreased by 80.9% (0.191-1.0=-0.809).
The relationship stated in the problem is supported. Survey respondents
who had less confidence in congress were less likely to be in the group of
survey respondents who thought we spend about the right amount of
money on highways and bridges, rather than the group of survey
respondents who thought we spend too much money on highways and
bridges. For each unit increase in confidence in Congress, the odds of
being in the group of survey respondents who thought we spend about the
right amount of money on highways and bridges decreased by 80.9%.
Exp(B)
Lower Bou
1.019
1.073
.253
.9
.8
.0
1.003
1.188
.191
.9
.9
.0
ters II
Slide
46
1
2
3
Marginal
Percentage
37.1%
55.7%
7.2%
100.0%
62
93
12
Valid
167
Missing
103
Total
270
The proportional by chance accuracy rate
was computed by
Subpopulation
153agroup based on
calculating the proportion of cases for each
the number
of dependent
cases in variable
each group
inone
thevalue
'Case
Processing
a. The
has only
observed
Summary', and then squaring and summing the proportion of
in 146 (95.4%) subpopulations.
cases in each
group (0.371 + 0.557 + 0.072 = 0.453).
ters II
Slide
47
Classification
Predicted
Observed
1
2
3
Overall Percentage
1
15
7
5
16.2%
2
47
86
7
83.8%
3
0
0
0
.0%
Percent
Correct
24.2%
92.5%
.0%
60.5%
ters II
Slide
48
True
True with caution
False
Inappropriate application of a
statistic
ters II
Slide
49
Among this set of predictors, confidence in Congress was helpful in distinguishing among the
groups defined by responses to opinion about spending on highways and bridges. Survey
respondents who had less confidence in congress were less likely to be in the group of survey
respondents who thought we spend too little money on highways and bridges, rather than the
group of survey respondents who thought we spend too much money on highways and bridges.
For each unit increase in confidence in Congress, the odds of being in the group of survey
respondents who thought we spend too little money on highways and bridges decreased by
74.7%. Survey respondents who had less confidence in congress were less likely to be in the
group of survey respondents who thought we spend about the right amount of money on
highways and bridges, rather than the group of survey respondents who thought we spend too
much money on highways and bridges. For each unit increase in confidence in Congress, the
odds of being in the group of survey respondents who thought we spend about the right amount
of money on highways and bridges decreased by 80.9%.
1.
2.
3.
4.
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
50
Problem 2
1. In the dataset GSS2000, is the following statement true, false, or an incorrect application of
a statistic? Assume that there is no problem with missing data, outliers, or influential cases,
and that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on space exploration" [natspac]. These predictors differentiate survey
respondents who thought we spend too little money on space exploration from survey
respondents who thought we spend too much money on space exploration and survey
respondents who thought we spend about the right amount of money on space exploration from
survey respondents who thought we spend too much money on space exploration.
Among this set of predictors, total family income was helpful in distinguishing among the
groups defined by responses to opinion about spending on space exploration. Survey
respondents who had higher total family incomes were more likely to be in the group of survey
respondents who thought we spend about the right amount of money on space exploration,
rather than the group of survey respondents who thought we spend too much money on space
exploration. For each unit increase in total family income, the odds of being in the group of
survey respondents who thought we spend about the right amount of money on space
exploration increased by 6.0%.
1.
2.
3.
4.
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
51
Dissecting problem 2 - 1
1. In the dataset GSS2000, is the following statement true, false, or an incorrect
application of a statistic? Assume that there is no problem with missing data, outliers, or
influential cases, and that the validation analysis will confirm the generalizability of the
results. Use a level of significance of 0.05 for evaluating the statistical relationships.
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
For these [natspac].
problems, we
will predictors differentiate survey
"opinion about spending on space exploration"
These
assume
that there
no problem
respondents who thought we spend too
little money
on is
space
exploration from survey
with
missing
data,
or
respondents who thought we spend too
much
money
on outliers,
space exploration
and survey
respondents who thought we spend about
the right
amount
of money
on space exploration from
influential
cases,
and that
the
survey respondents who thought we spend
too much
money
on space exploration.
validation
analysis
will confirm
the generalizability of the
results
Among this set of predictors, total family
income was helpful in distinguishing among the
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
52
Dissecting problem 2 - 2
The variables listed first in the problem
statement are the independent variables
(IVs):
"highest
year ofis school
completed"
1. In the
dataset
GSS2000,
the following
statement true, false, or an incorrect application of a
[educ],
"sex"
[sex]
and
familywith missing data, outliers, or influential cases, and that
statistic?
Assume
that
there
is "total
no problem
the validation
will confirm the generalizability of the results. Use a level of significance of
income" analysis
[income98].
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on space exploration" [natspac]. These predictors differentiate survey
respondents who thought we spend too little money on space exploration from survey respondents
who thought we spend too much money on space exploration and survey respondents who thought we
spend about the right amount of money on space exploration from survey respondents who thought
we spend too much money on space exploration.
The variable used to define
groups
is set
theof
dependent
Among
this
predictors, total family income was helpful in distinguishing among the groups
variable
"opinion
about about spending on space exploration. Survey respondents who had
defined
by (DV):
responses
to opinion
higher
total
family
incomes
were more likely to be in the group of survey respondents who thought
spending on space
weexploration"
spend about[natspac].
the right amount of money on space exploration, rather than the group of survey
respondents who thought we spend too much money on space exploration. For each unit increase in
total family income, the odds of being in the group of survey respondents who thought we spend
about the right amount of money on space exploration increased by 6.0%.
1.
2.
3.
4.
True
True with caution
False
Inappropriate application of a statistic
ters II
Slide
53
Dissecting problem 2 - 3
SPSS multinomial logistic regression models the relationship
by comparing each of the groups defined by the dependent
variable to the group with the highest code value.
1. In the dataset
is the
followingabout
statement
true, on
false,
an incorrect application of a statistic?
TheGSS2000,
responses
to opinion
spending
theorspace
Assume that there
is
no
problem
with
missing
data,
outliers,
or
influential
cases, and that the validation
program were:
analysis will confirm the generalizability of the results. Use a level of significance of 0.05 for evaluating the
1= Too little, 2 = About right, and 3 = Too much.
statistical relationships.
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income" [income98]
were useful predictors for distinguishing between groups based on responses to "opinion about spending on
space exploration" [natspac]. These predictors differentiate survey respondents who thought we spend
too little money on space exploration from survey respondents who thought we spend too much money
on space exploration and survey respondents who thought we spend about the right amount of money on
space exploration from survey respondents who thought we spend too much money on space
exploration.
Among this set of predictors, total family income was helpful in distinguishing among the groups defined by
responses to opinion about spending on space exploration. Survey respondents who had higher total family
incomes were more likely to be in the group of survey respondents who thought we spend about the right
amount of money on space exploration, rather than the group of survey respondents who thought we spend
analysis
will result
in two
too much money onThe
space
exploration.
For each
unitcomparisons:
increase in total family income, the odds of being in the
group of survey respondents
who
thought we who
spendthought
about the
amount
of money
on space exploration
survey
respondents
weright
spend
too little
money
increased by 6.0%.
versus survey respondents who thought we spend too much
money on space exploration
1. True
survey respondents who thought we spend about the right
2. True with cautionamount of money versus survey respondents who thought we
3. False
spend too much money on space exploration.
ters II
Slide
54
Dissecting problem 2 - 4
Each problem includes a statement about the
The variables
"highest year of school completed" [educ], "sex" [sex] and "total family income"
one independent variable and
[income98]relationship
were usefulbetween
predictors
for distinguishing between groups based on responses to
the
dependent
variable.
The answer [natspac].
to the
"opinion about spending on space exploration"
These predictors differentiate survey
is based
the stated
relationship,
respondentsproblem
who thought
weonspend
too little
money on space exploration from survey
the relationships
between
otheron space exploration and survey
respondentsignoring
who thought
we spend too
muchthe
money
independent
variables
and
the
dependent
respondents who thought we spend about the right variable.
amount of money on space exploration from
survey respondents who thought we spend too much money on space exploration.
Among this set of predictors, total family income was helpful in distinguishing among the
groups defined by responses to opinion about spending on space exploration. Survey
respondents who had higher total family incomes were more likely to be in the group of
survey respondents who thought we spend about the right amount of money on space
exploration, rather than the group of survey respondents who thought we spend too much
money on space exploration. For each unit increase in total family income, the odds of
being in the group of survey respondents who thought we spend about the right amount of
money on space exploration increased by 6.0%.
1.
2.
3.
4.
True
This problem identifies a difference for only one
True with caution
of the two comparisons based on the three values
False
of the dependent variable.
Inappropriate application of a statistic
Other problems will specify both of the possible
comparisons.
ters II
Slide
55
Dissecting problem 2 - 5
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on space exploration" [natspac]. These predictors differentiate survey
respondents who thought we spend too little money on space exploration from survey
respondents who thought we spend too much money on space exploration and survey
respondents who thought we spend about the right amount of money on space exploration from
survey respondents who thought we spend too much money on space exploration.
Among this set of predictors, total family income was helpful in distinguishing among the
groups defined by responses to opinion about spending on space exploration. Survey
respondents who had higher total family incomes were more likely to be in the group of survey
respondents who thought we spend about the right amount of money on space exploration,
rather than the group of survey respondents who thought we spend too much money on space
exploration. For each unit increase in total family income, the odds of being in the group of
survey respondents who thought we spend about the right amount of money on space
exploration increased by 6.0%.
1.
2.
3.
4.
ters II
Slide
56
LEVEL OF MEASUREMENT - 1
1. In the dataset GSS2000, is the following statement true, false, or an incorrect application of
a statistic? Assume that there is no problem with missing data, outliers, or influential cases,
and that the validation analysis will confirm the generalizability of the results. Use a level of
significance of 0.05 for evaluating the statistical relationships.
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on space exploration" [natspac]. These predictors differentiate
survey respondents who thought we spend too little money on space exploration from
survey respondents who thought we spend too much money on space exploration and
survey respondents who thought we spend about the right amount of money on space
exploration from survey respondents who thought we spend too much money on space
exploration.
Among this set of predictors, total family income was helpful in distinguishing among the
Multinomial
requires
theexploration. Survey
groups defined by responses
tologistic
opinionregression
about spending
onthat
space
dependent variable be non-metric and the
respondents who had
higher total
family be
incomes
more likely to be in the group of survey
independent
variables
metricwere
or dichotomous.
respondents who thought we spend about the right amount of money on space exploration,
rather than the group
of survey
respondents
thought
we spend too much money on space
"Opinion
about
spending onwho
space
exploration"
exploration. For each
unit increase
in total
family
income,
the odds of being in the group of
[natspac]
is ordinal,
satisfying
the
non-metric
survey respondentslevel
who of
thought
we spend
about the for
right
measurement
requirement
theamount of money on space
dependent
exploration increased
by 6.0%.variable.
1.
2.
3.
4.
True
who thought we spend too little money, about
the right amount of money, and too much money
True with caution
on space exploration.
False
Inappropriate application of a statistic
ters II
Slide
57
LEVEL OF MEASUREMENT - 2
"Highest year of school
"Sex" [sex] is dichotomous,
completed" [educ] is interval,
satisfying the metric or
satisfying the metric or
dichotomous level of measurement
dichotomous
level ofGSS2000, is the following statement
1. In the dataset
true, false, or an incorrect application of a
requirement for independent
measurement
requirement
for is no problem with missing
statistic? Assume
that there
data, outliers, or influential cases, and
variables.
independent
variables.analysis will confirm the generalizability
that the validation
of the results. Use a level of
The variables "highest year of school completed" [educ], "sex" [sex] and "total family income"
[income98] were useful predictors for distinguishing between groups based on responses to
"opinion about spending on space exploration" [natspac]. These predictors differentiate survey
respondents who thought we spend too little money on space exploration from survey
respondents who thought we spend too much money on space exploration and survey
respondents who thought we spend about the right amount of money on space exploration from
survey respondents
who family
thought
we spend
too much
"Total
income"
[income98]
is money
ordinal,on space exploration.
satisfying the metric or dichotomous level of
requirement for independent
Among this set of measurement
predictors,Iftotal
familythe
income
was helpful
in distinguishing among the groups
variables.
we follow
convention
of treating
defined by responses
to
opinion
about
spending
on
space
exploration.
ordinal level variables as metric variables, the level Survey respondents who
had higher total family
incomes were
more likely
beanalysis
in the group
of survey respondents who
of measurement
requirement
forto
the
is
thought we spend satisfied.
about theSince
rightsome
amount
of
money
on
space
exploration,
rather than the group
data analysts do not agree
of survey respondents
who
thought
we
spend
about
the
right
amount
of
money
on space
with this convention, a note of caution should be
exploration. For each
unit in
increase
in total family income, the odds of being in the group of
included
our interpretation.
survey respondents who thought we spend about the right amount of money on space
exploration increased by 6.0%.
1. True
2. True with caution
ters II
Slide
58
ters II
Slide
59
ters II
Slide
60
Select the
dichotomous
variable sex.
ters II
Slide
61
ters II
Slide
62
ters II
Slide
63
Third, click
on the
Continue
button to
complete the
request.
ters II
Slide
64
Click on the OK
button to request
the output for the
multinomial logistic
regression.
ters II
Slide
65
1
2
3
1
2
33
90
85
94
114
208
62
270
138a
Marginal
Percentage
15.9%
43.3%
40.9%
45.2%
54.8%
100.0%
Multinomial
logistic
regression requires that the minimum ratio
(81.2%)
subpopulations.
of valid cases to independent variables be at least 10 to 1. The
ratio of valid cases (208) to number of independent
variables( 3) was 69.3 to 1, which was equal to or greater than
the minimum ratio. The requirement for a minimum ratio of
cases to independent variables was satisfied.
The preferred ratio of valid cases to independent variables is
20 to 1. The ratio of 69.3 to 1 was equal to or greater than the
preferred ratio. The preferred ratio of cases to independent
variables was satisfied.
ters II
Slide
66
-2 Log
Likelihood
354.268
334.967
Chi-Square
19.301
df
Sig.
6
.004
ters II
Slide
67
NUMERICAL PROBLEMS
Parameter Estimates
SPACE EXPLORATION
a
PROGRAM
1
Intercept
EDUC
INCOME98
[SEX=1]
[SEX=2]
Intercept
EDUC
INCOME98
[SEX=1]
[SEX=2]
B
Std. Error
-4.136
1.157
.101
.089
.097
.050
.672
.426
b
0
.
-2.487
.840
.108
.068
.058
.034
.501
.317
b
0
.
Wald
12.779
1.276
3.701
2.488
.
8.774
2.521
2.932
2.492
.
df
95% Confidence
Exp(B)
Lower Bound
U
Sig.
Exp(B)
1
.000
Multicollinearity
in the multinomial
logistic regression
is
1
.259 solution
1.106
detected1 by examining
the
.054
1.102
standard errors for the b
1
.115
1.959
coefficients.
A standard
error
0
.
.
larger than
2.0 indicates
numerical
problems,
such
as
multicollinearity
1
.003
among the independent variables,
1 for a dummy-coded
.112
1.114
zero cells
1
.087 because
1.060 all of
independent
variable
the subjects
have
the
same
value
1
.114
1.650
for the variable, and 'complete
0
.
.
.929
.998
.850
.
.975
.992
.886
.
ters II
Slide
68
Effect
Intercept
EDUC
INCOME98
SEX
-2 Log
Likelihood of
Reduced
Model
334.967a
337.788
340.154
338.511
Chi-Square
.000
2.821
5.187
3.544
df
Sig.
0
2
2
2
.
.244
.075
.170
ters II
Slide
69
had higher total family incomes were more likely to be in the group of survey respondents who
thought we spend about the right amount
of money
on space
exploration,
ratherinthan the group
There was
no evidence
of numerical
problems
of survey respondents who thoughtthe
wesolution.
spend too much money on space exploration. For each
unit increase in total family income, the odds of being in the group of survey respondents who
thought we spend about the right amount
of money on space exploration increased by 6.0%.
However, the individual relationship between
1.
2.
3.
4.
True
not statistically significant.
True with caution
The answer to the question is false.
False
Inappropriate application of a statistic
ters II
Slide
70
No
Inappropriate
application of
a statistic
Yes
Ratio of cases to
independent variables at
least 10 to 1?
Yes
Run multinomial logistic regression
No
Inappropriate
application of
a statistic
ters II
Slide
71
Overall relationship
statistically significant?
(model chi-square test)
No
False
Yes
Standard errors of
coefficients indicate no
numerical problems (s.e.
<= 2.0)?
Yes
No
False
ters II
Slide
72
Overall relationship
between specific IV and DV
is statistically significant?
(likelihood ratio test)
No
False
Yes
Yes
No
False
ters II
Slide
73
No
False
Yes
No
Yes
One or more IV's are
ordinal level treated as
metric?
No
True
Yes