On your own exercise 7
Group 1 is the row of students by the wall and group 1a is the continuation of the row on
the door side of the room.
OPEN House.sav
Group Task
Using correlation and
1
regression explain the
influence of COLLEGE on
POVERTY variable.
Analyze> Correlate> Bivariate> enter variables:
POVERTY, COLLEGE,
FEMCHILD,
Analyze> Regression>Linear,
Dependent = POVERTY and
Independent =COLLEGE.
Select OK.
1a
Graphs> Legacy Dialogs >
Scatter/dot> Simple>
Define> Y Axis (dependent
variable) and X Axis is
independent variable
Analyze> Regression>Linear,
Dependent = POVERTY and
Independent =FEMCHILD.
Select OK
Analyze > Regression>
Linear, Dependent
=POVERTY and Independent
=FEMCHILD and COLLEGE.
Select OK
Question
1. Is there as statistically significant correlation
between POVERTY and COLLEGE
educated or the number of FEMCHILD
households in a district? What is your
evidence?
2. Why is one Pearson value positive and the
other negative?
3. What is the PRE value for the influence of
COLLEGE on POVERTY?
4. If a district had a 30% college educated
population what percentage of poverty
would one anticipate? USE MODEL to
predict. Y est. poverty =
5. Produce graph of relationship between
College and Poverty.
1. What is the PRE for POVERTY and
FEMCHILD? How much of a change in the
PRE is realized when adding COLLEGE as
a variable?
2. In the ideal model which factor more fully
explains district poverty and what evidence
do you have for your choice?
3.
Should you retain or reject the null and
based on what evidence?
2
Using correlation and
regression explain the
influence of URBAN on a
support for John MCCAIN.
Analyze> Correlate>
Bivariate> enter variables:
MCCAIN , URBAN, OWNOCC
and SINGUNIT
Analyze> Regression>Linear,
Dependent = MCCAIN And
Independent = SINGUNIT
Select OK.
2a
Graphs> Legacy Dialogs >
Scatter/dot> Simple>
Define> Y Axis (dependent
variable) and X Axis is
independent variable
Analyze> Regression>Linear,
Dependent = MCCAIN and
Independent =URBAN. Select
OK
Analyze > Regression> Linear,
Dependent =_MCCAIN and
Independent =URBAN and
SINGUNIT.. Select OK
1. What is the correlation and the significance
of support for John MCCAIN based on a
district being urbanized [URBAN] and
having single unit occupied [SINGUNIT]
housing?
2. Why is the coefficient or Person’s value for
SINGUNIT and McCain positive?
3. What is the model of the line from which
one could compute MCCAIN support from
districts with high single unit housing
occupancy if the formula began:
Ymccain est. =
4. Graph a scatter plot for this relationship
and comment on whether you think the
model (line of best fit) is strong or weak?
What is your evidence
1. What is the PRE value for the relationship
between McCain and URBAN? What can
one conclude from this value?
2. What change occurs in McCain support
when one includes in the model
SINGUNIT? Is this addition important and
if so why?
3. Based on this output should you retain or
reject the null hypothesis with what
evidence? [- Reject; evidence is sig. =.000
ANOVA and sig. t-tests for both variables.]
4. If one drew a line that best reflected where
URBANIZATION and SINGUNIT crossed
the Y axis, the value of Y at that point
would be equal to what number?
3
Using correlation and
regression explain the
influence of URBAN on a
support for Barrak OBAMA.
1. What is the correlation and the significance
of support for OPBAMA based on a district
being urbanized [URBAN] and having single
unit occupied [SINGUNIT] housing?
Analyze> Correlate>
Bivariate> enter variables:
OBAMA , URBAN, and
SINGUNIT
2. Why is the coefficient or Person’s value for
SINGUNIT and OBAMA negative?
Analyze>
Regression>Linear,
Dependent = OBAMA And
Independent = SINGUNIT
Select OK.
3a
Graphs> Legacy Dialogs >
Scatter/dot> Simple>
Define> Y Axis (dependent
variable) and X Axis is
independent variable
Analyze>
Regression>Linear,
Dependent = OBAMA and
Independent =URBAN.
Select OK
Analyze > Regression>
Linear, Dependent =OBAMA
and Independent =URBAN
and SINGUNIT.. Select OK
3. What is the model of the line from which one
could compute OBAMA support from districts
with high single unit housing occupancy if
the formula began:
Yobama est. =
4. Graph a scatter plot for this relationship and
comment on whether you think the model
(line of best fit) is strong or weak? What is
your evidence?
1. What is the PRE value for the relationship
between OBAMA and URBAN? What can
one conclude from this value?
2. What change occurs in OBAMA support
when one includes in the model SINGUNIT?
Is this addition important and if so why?
3. Based on this output should you retain or
reject the null hypothesis with what
evidence?
4. If one drew a line that best reflected where
URBANIZATION and SINGUNIT crossed the
Y axis the value at that point would be equal
to what number?
4
Using correlation and
regression explain the
influence of race LATINO and
BLACK on MCCAIN support
variable.
Analyze> Correlate>
Bivariate> enter variables:
MCCAIN, BLACK, LATINO
Analyze> Regression>Linear,
Dependent = MCCAIN and
Independent = BLACK and
LATINO Select OK.
4a
Analyze> Regression>Linear,
Dependent = McCain and
Independent =WHITE. Select
OK
Graphs> Legacy Dialogs >
Scatter/dot> Simple>
Define> Y Axis (dependent
variable) and X Axis is
independent variable
Questions, concerns or fuzziness?
1. What is the correlation between the
percentage of Blacks and Latino’s in a
district and its support for McCain?
2. In a regression analysis how much is
explained by the % of black and Latino
populations in a district with regard to
support for McCain? What is the evidence?
3. How much needs to be explained to
understand McCain support?
4. If a congressional district had neither Black
nor Latino residents what would be the
level of support for McCain? What is your
evidence?
1. What is the proportional reduction in error
in estimating McCain support if one knows
the percentage of the district that is White?
2. What would be McCain’s support in a
district that was 70% white? What is your
evidence and calculations?
Ymccain =
3. Display a scatter plot that depicts the
influence of white residents on support for
John McCain.