程序代写代做代考 go graph Problem Set 2 􏰁 􏰂Learning to crawl􏰃 􏰄15􏰅􏰆􏰇 due midnight􏰈 7 October 2020

Problem Set 2 􏰁 􏰂Learning to crawl􏰃 􏰄15􏰅􏰆􏰇 due midnight􏰈 7 October 2020
􏰊 Working in teams of one to four people􏰈 please consider this scenario􏰇 􏰂You are employed as a junior statistician at Petit Poll 􏰁 a Canadian polling company􏰋 Petit Poll has a contract with a Canadian political party to provide them with monthly polling updates􏰋􏰃
􏰊 Please pick a political party that you are 􏰂working for􏰃􏰈 and pick a geographic focus􏰇 1􏰆 the overall election􏰈 2􏰆 a particular province􏰈 or 3􏰆 a specific riding􏰋
􏰊 Then decide on a survey methodology 􏰄hint􏰇 p􏰋 13 of Wu 􏰌 Thompson provides a handy checklist􏰆􏰋 Some questions you should address include􏰇
○ What is the population􏰈 frame􏰈 and sample􏰍
○ What sampling methods will you use and why 􏰄e􏰋g􏰋 you could
choose SRSWOR􏰈 stratified􏰈 etc􏰆􏰋 What are some of the statistical properties that the method brings to the table 􏰄e􏰋g􏰋 for SRSWOR you could discuss Wu 􏰌 Thompson􏰈 Theorem 2􏰋2􏰈 etc􏰈 as appropriate􏰆􏰍
○ How are you going to reach your desired respondents􏰍
○ How much do you estimate this will cost􏰍
○ What steps will you take to deal with non􏰁response and how will
non􏰁response affect your survey􏰍
○ How are you going to protect respondent privacy􏰍
Remember to consider all of this in the context of your 􏰂client􏰃 􏰁 for instance􏰈 who would be more interested in Alberta ridings􏰇 Bloc Québécois or the Conservatives􏰍 Who likely has more money to spend 􏰁 the Liberals or the Greens􏰍
􏰊 Develop a survey on a platform that was introduced in class􏰋 Be sure to test it yourselves􏰋 You will want to test this as much as possible􏰈 maybe even swap informally with another group􏰍
􏰊 Now release the surveys into the 􏰄simulated􏰆 􏰂field􏰃􏰋 Please do this by simulating an appropriate number of responses to your survey in R􏰋 Don􏰃t forget to simulate in relation to the survey methodology that you proposed􏰋 Show the results and discuss your 􏰂findings􏰃􏰋 Everything must be entirely reproducible􏰋 You may like to consider linking your survey 􏰂responses􏰃 with other data such as the census or GSS􏰋
􏰊 Use R Markdown to write a PDF report about all of this􏰋 Discuss your results and findings􏰈 your survey design and motivations􏰈 etc 􏰁 all of it􏰋 You are writing a report that will eventually go to the 􏰂client􏰃􏰈 so you must set the scene􏰈 and use language that demonstrates your command of statistical concepts but brings the reader along with you􏰋 Be sure to include graphs and tables and reference them in your discussion􏰋 Be sure
13

to be clear about weaknesses and biases􏰈 and opportunities for future
work􏰋
􏰊 Your report must be well written􏰋 You are allowed to􏰈 and should􏰈 use
mathematical notation􏰈 but you must explain all of it in plain english􏰋 Similarly􏰈 you can􏰈 and should􏰈 use surveys􏰎sampling􏰎observational data terminology􏰈 but again􏰈 you need to explain it􏰋
􏰊 Your report must include at least the following aspects􏰇 title􏰈 date􏰈 authorship􏰈 non􏰁technical executive summary􏰈 introduction􏰈 survey methodology􏰈 results􏰈 discussion􏰈 appendices that detail the survey􏰈 and references􏰋 Your 􏰂client􏰃 has stats graduates working for it who need to be impressed by the main content of the report􏰈 but also has people who barely know what an average is and these people need to be impressed also􏰋 This is why your report should include a non􏰁technical executive summary􏰋 In terms of length􏰈 this would typically be roughly 10 per cent of the report􏰋 It would be more detailed than an introduction􏰈 but still at a high level􏰋
􏰊 Your graphs must be of an extremely high standard􏰋
􏰊 Check that you have referenced everything􏰋 Strong submissions will draw
on related literature in the discussion and would be sure to also reference
those􏰋 The style of references does not matter􏰈 provided it is consistent􏰋
􏰊 Via Quercus􏰈 submit a link to your PDF report which is hosted on GitHub􏰋 At some point in the introduction to your report􏰈 you must provide a link to the GitHub repo where the code that you used for this assignment lives
􏰄Hint􏰇 Comment􏰋 Your􏰋 Code􏰋􏰆􏰋 Your entire workflow must be entirely
reproducible􏰋
􏰊 Please work in public where it is at all possible􏰋 If you have a personal
situation for not wanting your name to be public􏰈 then please email me a request for this 􏰁 please do NOT reveal your personal situation to me 􏰁 just make the request and suggest a work􏰁around􏰋
􏰊 Please be sure to include a link to your survey in your report and screenshots of the survey in the appendix of your report􏰋
􏰊 There should be no evidence that this is a class assignment􏰋
14