STA 6127 – HW#1 – Due Jan. 26
Multiple Regression
Download the Japanese emigration dataset. Import the dataset into SPSS. The variables in
the dataset (and their “columns”) are:
Prefecture 1-20
Emigrants 23-25 /* Per 1 million residents */
% land cultivated by tenant farmers 32-33
% change in ratio of tenant farmlands 37-41
Average area of arable land per farm 45-49
# of government labors working in Hawaii 52-57 /* in 100s */
Existence of pioneer immigrants 65 /* 1=Yes, 0=No */
The response variable is EMIGRANT RATE. The k=5 predictor variables can be broken
into 3 “groups”: POVERTY (the 2 variables concerning tenant farming), POPULATION
PRESSURE (area of land per farm), and INFORMATION (the 2 variables concerning
workers in Hawaii and the Dummy variable for Pioneer immigrants).
1. Obtain partial regression plots for each of the predictor variables and interpret.
a) Analyze Regression Linear Fill in dependent and independent variables
Plots Produce all Partial Plots Continue OK
2. Write out the fitted equation.
3. Test whether any of the predictors are associated with the EMIGRANT RATE at the
=0.05 significance level
4. Test whether the EMIGRANT RATE is associated with POPULATION PRESSURE,
after controlling for the variables measuring POVERTY and INFORMATION at the
=0.05 significance level.
5. Obtain the pairwise Pearson correlations of each predictor variable and the response
variable.
a) Analyze Correlate Bivariate select all variables
6. Obtain all partial correlations involving Y and each predictor (e.g.
5,4,3,21 XXXXYX
r
)
a) Analyze Correlate Partial Variables:Y, X1 Controlling for: X2,X3,X4,X5
