Question: You are given a large sample of data to analyse using regression analysis.
a) Firstly, you wish to discover whether using simple linear regression analysis is appropriate. Which two graphs could you create with your data to help you? How would you interpret them?
b) You also decide to consider using a quadratic regression model. Write out the regression model for this case. The independent variable is x and the dependent variable is y.
c) How would you test whether the coefficient of the quadratic term was statistically significant at the 5% level?
d) How would you compare the two models in order to choose the most appropriate one?
e) Suppose that you wished to study differences in income in the UK.
One of the factors which you believe is important is gender. You are given statistical data from a national survey. Respondents can state that they are male or female or refuse to disclose their gender. How would you incorporate gender into your regression equation or would you choose to discard it from your analysis?
f) What pitfall should you avoid with part e) above?