The accompanying scatterplot gives data from a random sample of 27 countries where the response variable is infant mortality (deaths per 1,000 live births) and the explanatory variable is the gross domestic product (GDP) per capita (in thousands of dollars). The gross domestic product per capita is a representation of a country€™s standard of living.

a. Describe the relationship between infant mortality and GDP per capita. Is this relationship linear?

b. Th e regression line is also included in the scatterplot. Will this regression line be a good prediction of infant mortality given GDP per capita? Why or why not? Sometimes we can transform data that don€™t fit a linear pattern in such a way that they do fit a linear pattern. We calculated the logarithm of all the values of both the infant mortality and GDP per capita and plotted the results in the scatterplot below.

c. Describe the relationship between logarithm of infant mortality and logarithm of GDP per capita. Is this relationship linear?
d. The regression line is also included in the scatterplot. Will this regression line be a good prediction of the logarithm of infant mortality given the logarithm of GDP per capita? Why or why not?

100 80 40 20 40 10 20 30 50 GDP per capita Infant mortality 2.8 2.1 1.4 0.7 0. - 0.7 0.7 1.4 Log (GDP per capita) Log (infant mortality)