Multivariate Statistics

Homework: Multiple Regression Assignment

Use the dataset: TEEN.sav

These data were originally published as part of a study of gambling behavior by Ide-Smith & Lea (1988) in the Journal of Gambling Behavior. The dataset contains the following information on each of the 47 teenagers in the study:

- Subject ID#

- Gender (0= male; 1= female )

- Socioeconomic status (SES) a scale based on parents' occupation with higher values meaning higher SES.

- Income (an index of S's available spending money) in a pounds-per-week metric.

- \(\mathrm{N}\) of words correctly defined-out of 12 , a proxy measure of verbal IQ.

- S's estimate, based on responses on questionnaire, of their total expenditures on all forms of gambling - expressed in a pounds-per-year metric.

Each line in the database contains values for all 6 data elements for a single subject \((\mathrm{N}=47)\).

Challenge

1. Check the distributions of all variables in the dataset, looking specifically at the assumptions of normality, linearity and homogeneity of residuals, for suitability for using Regression techniques in this analysis.
2. Calculate the correlation matrix for all meaningful variables in the dataset, and inspect each of the correlations. Write your summary of what the pattern of correlations suggests about creating and testing viable models in subsequent regression analyses.
3. Using the model-based methods discussed in class, fit at least two competing regression models - interpret both models in terms of the variables that are most explanatory of gambling expenditures. Write up your analyses and explain which model is the bestfitting, and which one you prefer and why. Include all appropriate tables and statistical output to make and defend your model choice.

Price: $17.54

Solution: The downloadable solution consists of 12 pages, 554 words and 17 charts.
Deliverable: Word Document