Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Statistical Analysis of Rent Paid by U.S. Households
1. Statistical Analysis of Rent Paid by U.S.
Households
Group Mentor :
Prof. Manish Thakkar
Group Members:
Riddhima Kartik (20151037)
Rishabh Surana (20151038)
2. Objective :
To find the extent of relation/dependency of
variables(dependent and independent) by using
various statistical tools like Correlation, Regression
and Independence test using SPSS.
3. Variables Involved
Dependent Variables :
Rent
Independent Variables :
Household Income
Electricity Cost
Gas Cost
Rooms Per House
Vehicle Per Household
4. Software Used :
SPSS (Statistical Package for the Social Sciences)
Statistical Methods Used :
Normality Test
Correlation with Scatter plot.
Multiple Regression Analysis
5. Condition to Apply Parametric Statistical
Methods-
Parametric statistical methods(ANOVA and Linear Regression etc.) requires that
dependent must be normally distributed.
To check Normality by SPSS:
Steps Involved : Analysis->Descriptive->Explore->Plot->Normality plot-> ok
10. Correlation Test :
Hypothesis considered:
Hypothesis 1
H0 : There is no significant correlation between Rent and Household Income.
Ha : There is significant correlation between Rent and Household Income.
Hypothesis 2
H0 : There is no significant correlation between Rent and Electricity Cost.
Ha : There is significant correlation between Rent and Electricity Cost.
Hypothesis 3
H0 : There is no significant correlation between Rent and Gas Cost.
Ha : There is significant correlation between Rent and Gas Cost.
11. Correlation between:
Rent and Household Income = .000
Rent and Electricity Cost = .009
Rent and Gas Cost = .013
Therefore, we reject all three null
Hypothesis and can say that these
factors have significant impact
over rent
Confidence Level taken is 95%
To check Strength of Correlation we can see Scatter plot ->
15. Strength with Household Income: Moderate, not very strong.
Strength with Electricity Cost: Weak
Strength with Gas Cost: Weak
Steps Involved :-(Graph->chart-builder->drag the simple scatter graph):- If all
the scatter points are in same straight line then we have strong relationship
between them.
16. Multiple Regression :
Ho : Variation in Y(Dependent variable(Rent)) is unrelated to variation in
X(Independent variable(Household Income, Electricity Cost, Gas Cost)).
Or
Ho: Correlation between X(Household Income, Electricity Cost, Gas Cost) and
Y(Rent) is 0.
Ha: Correlation between X(Household Income, Electricity Cost, Gas Cost) and
Y(Rent) is not 0.
19. Result
R^2 value is 28.6, which means that 28.6% variation in Rent is due to
Household Income, Electricity Cost, Gas Cost and rest because of other
factors.
Constant = 1168.89 -> This much will be the basic Rent that households have
to pay irrespective of other factors.
p-value for all the variables are quite lower than our significance level that is
0.05. So, Null hypothesis is rejected and we can conclude that :
Conclusion : Correlation between X(Household Income, Electricity Cost, Gas
Cost) and Y(Rent) is not 0.
Y = 1168.89 + .003A + 2.415B + 4.481C