STAT question

Source of data:

A website service company's survey of customer satisfaction in its physical stores.

Questions:

What are the specific reasons that affect customer satisfaction?

Which model is best for predicting customer satisfaction?

Interval predictions for future customer satisfaction.

Is there an interaction between different variables? (Variable Interaction)

The exploration of lurking variables in the process of building the model.

Comparison of model selection AIC and BIC

The importance of different variables to the overall model.

If 80 or more is satisfaction, explore the variable relationship that affects the satisfaction or dissatisfaction ratio.

Model:

Simple Linear Regression Model

Multiple Regression Model (include Polynomial and Log model)

Logistic Regression Model

Type of variables and structure of data:

Categorical variables: GENDER (Customer gender); ARRIVE (The way of customer go to store); REASON (1 for personal question; 2 for family question; 3 for business question); VERSION (1 for old version of the product; 2 for new version of the product); LOCATION (1 for Within 50 miles; 2 for 50 miles away); MARRIED (1 for married; 2 for not married)

Numeric variables: Satisfaction (Customer satisfaction); AGE (Customer Age); EDU (Customer educational level); TEMP (On the day of the temperature); STAYMINUTES (The length of time a customer stay in the store)