Virginia Mathematics Teacher Spring 2017

Trump:

cause confusion about the recent polling results among the voting age population for the 2016 Presidential election cycle. 1. Clinton had an overall lead over Trump claiming that her gains over the past day or two helped her, she was unable to move far enough ahead to remove the overlap between their two intervals. 2. The number of undecided and third-party voters is much higher than in more recent elections. 3. Clinton’s coalition relies increasingly on college-educated whites and Hispanics, which is somewhat inefficiently configured for the Electoral College, because these voters are less likely to live in swing states. If the popular vote turns out to be a few percentage points closer than the polls projected then, Clinton will be an Electoral College underdog. The polling data does not reflect the actual election process. The United States voting system uses the Electoral College process to select the winner. The pollsters use raw numbers and percentages to project the winners, which is consistent with the plurality voting system, which is, the person with the most votes wins. The Electoral College system uses electors. The electors are proportionate to the population for each state. In addition, all of the states, except for Maine and Nebraska, have a winner takes all policy, which means, the person with the most popular votes captures all of the electors from that state. Maine and Nebraska have a proportion policy in which each candidate captures a proportion of the electors that aligns with the proportion of votes they received. The number of electors needed to win the election is 270. If the polling organizations wished to report a more accurate prediction, they would need to replicate the Electoral College voting structure using the polling data. This would require polling each state separately, assign the electoral votes to the candidates and sum up the electoral votes for each candidate. Prior to the election, there were some swing states that conducted their own state level polls that is

46.3% 1.9% 44.4% 48.2% p E to r r

Again, confidence intervals were calculated for each of the candidates. The polling results show that 48.2% of the people polled claimed they would vote for Clinton with a margin of error of 1.9%. This means that it is expected that Clinton would receive from 46.3% to 50.1 % of the popular vote. Likewise, the polling results show that 44.4% of the people polled claimed they would vote for Trump with a margin of error of 1.9%. This means, that it is expected that Trump would receive between 44.4% and 48.2% of the popular vote. Comparing these two intervals, we again observe an overlap (see Figure 2). This time, the overlap is a larger even though the intervals are smaller. Almost half of each candidate’s confidence interval overlaps with each other. Evaluating the polling data statistically, using confidence intervals, suggests there is no clear leader emerging from the polling data. Many times when news organizations, political parties, polling organizations, or other outlets report their polling results, they exclude an important aspect of the election process, the Electoral College. Statistical literacy is crucial to understand the polling results, but being political literate is also necessary. Political literacy, in the traditional sense, refers to an educational goal that is, preparing students to become knowledgeable and engaged citizens (Cassel & Lo, 1997). Denver and Hands (1990) define political literacy similarly, in which people have the knowledge and understanding about the political process and political issues so they can effectively fulfil their roles as engaged citizens. We draw on the definition supplied by Denver and Hands (1990), in which people need to know and understand the political process, in this case, the political election process. It appears that some organizations play on people’s weak political literacy skills in an attempt to provoke confusion by expounding on extraneous issues surrounding the polling data. For example, an article published by Silver (2016, November 8) in Fivethirtyeight.com cites several reasons that

Virginia Mathematics Teacher vol. 43, no. 2

44