|
|
|
||||
|
|
|||||
Page 235 Table 10.1 Means and standard deviations for depression data set.
File Open - open the depression data set
Insert Add variables - add one variable after sex called sex1 and in the box at the bottom, type "= sex - 1" (without the quotes)
Statistics
Basic statistics
Descriptives
Variables - select cesd, health, sex1, age, education and income
"Advanced" tab - uncheck "Valid N" and "minimum and maximum"
Summary
| Mean | Std.Dev. | |
|---|---|---|
| SEX | 0.62 | 0.49 |
| AGE | 44.41 | 18.09 |
| EDUCAT | 3.48 | 1.31 |
| INCOME | 20.57 | 15.29 |
| CESD | 8.88 | 8.82 |
| HEALTH | 1.77 | 0.84 |
Page 235 Table 10.2 Correlation matrix for depression data set.
Statistics
Basic statistics
Correlation matrices
One variable list - select cesd, health, sex, age, education and income
Summary
| SEX | AGE | EDUCAT | INCOME | CESD | HEALTH | |
|---|---|---|---|---|---|---|
| SEX | 1.000 | 0.044 | -0.106 | -0.180 | 0.124 | 0.098 |
| AGE | 0.044 | 1.000 | -0.208 | -0.192 | -0.164 | 0.304 |
| EDUCAT | -0.106 | -0.208 | 1.000 | 0.429 | -0.101 | -0.270 |
| INCOME | -0.180 | -0.192 | 0.429 | 1.000 | -0.158 | -0.183 |
| CESD | 0.124 | -0.164 | -0.101 | -0.158 | 1.000 | 0.212 |
| HEALTH | 0.098 | 0.304 | -0.270 | -0.183 | 0.212 | 1.000 |
Page 237 Table 10.3 Canonical correlation coefficients for first correlation (depression data set)
NOTE: We were unable to find a way to have Statistica display the raw coefficients shown on the left-hand side of the table.
Statistics
Multivariate exploratory techniques
Canonical analysis
Variables - select cesd, health, sex, age, education and income
OK
OK
"Quick" tab
Variables for canonical analysis - select cesd and health for the first variable list
and sex, age, education and income for the second variable list
OK
RESULTS
"Canonical scores" tab
Left and right set canonical weights
These are the standardized coefficients
| Root 1 | Root 2 | |
|---|---|---|
| SEX | 0.025 | -0.396 |
| AGE | 0.871 | 0.443 |
| EDUCAT | -0.383 | 0.448 |
| INCOME | 0.082 | 0.555 |
Page 238 Table 10.4 Computation of correlation between U(1) and V(1)
The U's and V's can easily be added to the Statistica data set. Begin with the RESULTS from the analysis above and click on the "Save canonical scores" button. When asked about saving other variables with the scores, you may want to click on the "Select all" button to save all of the variables into the new data set with the scores.
Page 239 The second canonical correlations give at the top of the page.
These numbers are displayed in the output shown immediately above, called "Root 2".
Page 241 Figure 10.1 Plot of 294 pairs of the canonical variables U(1) and V(1) for the depression data set (canonical correlation = 0.405)
Graphs
2D Scatterplots
"Advanced" tab - set "Fit" to off
Variables - select r_fac1 for the x-axis and l_fac1 for the y-axis
OK
OK
Page 242 Table 10.5 Correlations between canonical variables and corresponding variables (depression data set)
The panels of the table were obtained using the results from the canonical correlation conducted above.
RESULTS
"Factor structures"
Factor structures and redundancies
| Root 1 (U) | Root 2 (V) | |
|---|---|---|
| CESD | -0.281 | -0.960 |
| HEALTH | 0.878 | -0.478 |
| Root 1 (U) | Root 2 (V) | |
|---|---|---|
| SEX | 0.089 | -0.525 |
| AGE | 0.936 | 0.225 |
| EDUCAT | -0.532 | 0.636 |
| INCOME | -0.254 | 0.734 |
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services