3) A study in the Journal of Fake Research reported that individuals who earn higher incomes are more likely to be divorced than individuals who earn lower incomes. This finding was obtained using basic cross-sectional regression of individuals' divorce status on income, and included no other control/explanatory variables. The authors conclude that earning more money causes divorce. a. What is one alternative explanation for this finding? That is, why might this finding suffer from bias? Would the bias lead the researchers to over-estimate or under-estimate the causal relationship? b. Suppose you have data on a very large sample of individuals in the U.S. These track each individual over many years, including information such as occupation, employer, zip code of residence, zip code of employer, marital status (single, married, divorced, widowed, etc.), number of children, demographic characteristics, and more. How might you use these data, perhaps along with information on relevant policy changes, to credibly test the hypothesis that earning more money causes divorce?