Answer Happy

3. Imagine a researcher has written a report on the factors associated with poverty in England. They have collected data on poverty and a number of possible related factors for a sample of 58 medium sized towns. Their data is shown in the Excel spreadsheet "poverty.xls" which you should download from Canvas. The variables in the spreadsheet are described in Table 7: Table 7: Description of Poverty data TOWN Town identification number POV The poverty rate, defined as the percentage of people classified as poor by the town council based on a set of deprivation indicators. FAMSIZE UR The mean family size, i.e. mean number of people in a family, in the town. The male unemployment rate in the town. Median household income per capita in the town. INCOME TRAIN A dummy variable: TRAIN=1 if the town has a train station, TRAIN=0 if not Your task is to analyse the factors associated with poverty and to make a judgement about the researcher's analysis. a. Draw a scatter plot of the poverty rate, POV, against either INCOME, FAMSIZE or UR. Copy and paste the scatter plot in to your answer file. Comment on what your graph suggests about the relationship between POV and your chosen variable. [5 marks] b. Calculate the correlation coefficient between the poverty rate, POV, and the variable you have chosen in part a, and test if it is statistically significant. Remember to explain your method and to comment on your answer. [5 marks] c. Use Excel to estimate a simple linear regression of the relationship between the poverty rate and your chosen variable from part a, i.e. POVi = a + b Xi + e, where X stands for your chosen variable from part a and i=1,...58. i. Copy and paste your Excel output in to your answer doc. Ensure it is neatly and appropriately presented. Interpret the intercept and slope coefficient and comment on whether either are statistically significant. [4 marks] ii. Use the regression to predict what value X would have to be to give a poverty rate of 9%. Comment on your answer. [2 marks] Interpret the goodness of fit for your regression, and use it to test if the model has any explanatory power. [4 marks]