Critical Thinking- Presidential Polls

Description

Study Room Forum Week 7 POST 1= 1 PAGE  POST 2= 1 PAGE Polls in the presidential election in 2016 were overall shown to highly inaccurate. This week we will undertake the ambitious task of analyzing a couple of 2020 presidential polls. Our goal is to learn about the analyze and evaluate polls–how to ask questions that may lead to an evaluation of the accuracy of the poll. These polls present statistical arguments. So, in the process of this exercise, we are analyzing and evaluating a statistical argument. This is a learning experience–try to help one another–we do not have all the information we need to do a complete analysis–but let’s have some fun and give it a try: First, let consider the following taken by the Trafalgar Group: https://drive.google.com/file/d/12_IApGhCQY27OukwWXGnk0MFgfPctmhQ/view Then, let’s consider the following poll taken by CNBC for the same state: https://changeresearch.com/wp-content/uploads/2020/08/CNBC-CR_Battleground-Toplines_Wave-12_August-21-23-Wave-12_-8_21-8_23.pdf There is a lot of information in both these reports, but there is also a lot of important missing information. Let’s first note some puzzling, key information: Both polls ask about voting preference for president in the state of Michigan. The times of both polls are very close–but not exactly the same. The results of both polls are very different.  The Trafalgar poll has the following results: Trump 46.6% Biden 45.2% +/- 2.98% Here is some information on the general methodology of the Trafalgar Group (but please note we do not know precisely what methodology was used for the above poll): https://www.thetrafalgargroup.org/polling-methodology/ And the CNBC poll has the following results: Biden 49% Trump 46% +/- 1.4% FIRST POST As noted above, there is a lot we do not know about the methodologies of these particular polls from the reports. But, try to think of some questions related to the samples and the (target) population that might be important in trying to figure out why these polls have such different results. And, if you would like to do a little research (not required) please do. Here are some key questions: What is the population in each poll? Are they the same? What are the samples in each poll? How were the samples obtained? What do we know about the methodologies? What do we not know?  Try to come up with some hypotheses for why these polls have such different results–and do your best to base these hypotheses on evidence. Here is some information on the general methodology of the Trafalgar Group: https://www.thetrafalgargroup.org/polling-methodology/ SECOND POST Dialogue with students about these poll–offer comment on another student’s post (standards of civility and decorum apply), further elaborate, introduce new information.  Students may also explore other polls at the Web Site RealClearPolitics, and offer comments–again, try to give arguments, that is, support your claims with evidence. https://www.realclearpolitics.com/epolls/latest_polls/ This is a very informative Web Site. Students can click on any one of these polls and drill down to a report.