Task 1: Specification of the Dependent Variable (1 points)

You are given the absolute counts of votes for Trump (TRUMPVOT16), Clinton (CLINTONVOT) and others[1] (OTHERVOT16), as well as the number of persons 18 years and older[2] (POP18PLUS), number of registered voters[3] (REGVOT16) and the turnout rate[4] (TURNOUT16).

[a] Calculate the percentage of voters who voted for either candidate. Be careful to select the proper reference population in the denominator. Justify your calculation.

[b] Evaluate the distribution of both percentages and chose that dependent variable, those percentage distribution is easier to transform to symmetry. Map the percentage of voters of your candidate and interpret its spatial distribution.

[c] Can all 254 counties be used in the analysis or few counties have a too small denominator, thus leading to instable percentage estimates.

Note: The mapping function in the library TexMix uses quantiles; therefore, your map pattern will look slightly different from that shown in the back of your handout, which uses fixed intervals in 10% increments.

[1] Besides the two main candidates, the electorate also has had a choice to vote for independent candidates and Libertarians. Only a very small number of voters in each county has chosen these alternatives.

[2] Note that not all persons 18 years and older qualify to vote; for instance, because some are not U.S. citizens.

[3] In Texas, voters need to register in order to be eligible to vote. This does not imply that all registered voters will participate in an election.

[4] The turnout percentage is that proportion of registered voters who participate in an election.