« Class StandardsThursday -- Polling I: Sample Error »

Monday

  12/08/18 11:55, by John

Comment if you will be attending any of the suggested sections (besides the Basic) so I can prepare accordingly.

BASIC SECTION: Optional Class meeting to decide which students will not receive credit for this class this semester -- I won't be attending.

Here is a link to the Project Deadline page -- every student (except Dylan) in the class has been tasked to record specific deadlines for their specific projects for the rest of the semester as comments to this post before Monday's class. Those class members (except Dylan) who do not complete this task will risk losing access to gaining a credit in this class. (I, literally, have no idea the number of students who think they are earning credits in this class)

Those who have completed projects -- bring an example(s) of your completed projects as a demonstration of what a completed projects in this class will look like.

The class will decide how to deal with the extremely low (non-existent?) standards this class allows for itself which continue to fall with each class session. There is no need to look further than Monday's class when 5 people attended the class sections or Thursday's class where only three people took the time and spent the energy to read the handout on sampling error (enough to get 3 out of five questions on a comprehension quiz). Yet, if "the class" is not embarrassed by these last "classes" who am I to complain?

The class will take 15 minutes (or longer a necessary) to decide what will be done about the grading in this class -- what will constitute a passing evaluation for a student's work in this class and, what will be grounds for no credit.  These standards must be specific and objective, recorded in writing, and signed by all members of the class present and posted as a comment to this page so I have a methodology moving forward.

SECTION ONE: Polling at Work

PREPARATION:

1. Here is a link to a grading of pollsters. Check out why, who, and how this grading process works.

2. Here is a link to a top rated poll from a top rated polling site.

a. Find some interesting tidbits from the data.

b. Check out the "methodology" link.  Bring something of interest to the discussion from this description.

Unfortunately (learning from experience) if those who plan on attending this section DO NOT post some (or all) of their preparation, I will need to administer another room clearing comprehension quiz.

3. Optional -- play around and post links of other interesting polls and polling sites -- particularly, look for statements of methodology of how polls do (or do not) look to eliminate errors in their sampled data. (here's a survey of Kansas voters before 2016 presidential and here's Minnesota; here's a fun site YouGov with hundreds of polls)like this one on this year's holiday spending or even funnier here's one on embarrassing parents.

SECTION TWO: Excel Practice

Here's another Google Sheet with an exercise on in -- make a copy of the sheet and visit the three pages on this sheet -- the step by step exercise is on sheet "exercise" -- tell me if you have problems and I will fix these.

 SECTION THREE: Basic correlation for basic predictions....

Simple correlation to create a simpler algorithm for predicting hockey games (why not) -- I'll create a correlation of statistics too wins and you chose how to use this information to have a machine predict the outcome of hockey teams.

Here's a Google Sheet with this organized data -- look at the top of the chart on DATA SORT page for correlations .

Preparation: Examine the correlated data and write down a brief outline of how you would use this information to say which hockey team will do better in the next few weeks.

SECTION FOUR: 

Five Platonic solids -- Origami!

Preparation: Create this tetrahedron....

or a modular model for more shape building flexibility (another module tutorialand best this one) -- the modules help if we want to build more of the solids....

19 comments

Comment from: [Member]

Question: Till when do you think we will be discussing Excel? I want to change my “Understanding Excel” project deadline accordingly.

12/08/18 @ 12:52
Comment from: [Member]

Tangwyn – EXCEL is the only section that has sparked interest in the class (4 - 5 people) so I will continue Excel sections throughout the remaining semester.

12/08/18 @ 13:03
Comment from: [Member]

Ok great! Thanks John!

12/08/18 @ 14:03
Comment from: [Member]

I will be in the Basic section tomorrow.

I will also be in section Four. I am working on making a Tetrahedral Skeleton and will bring it to class tomorrow.

12/09/18 @ 10:50
Comment from: [Member]

I will be in the basic section as well as section two. I will complete the Excel exercise as well as a can.

12/09/18 @ 17:00
Comment from: [Member]

Thanks John! I will post my project deadline right now. I am struggling with the motivation for the projects and the knowledge of how or what to do. I like the idea of learning excel but I’m not sure what I need to do to have it count as a project. John: could we talk tomorrow about my struggles with the projects and could you give me some direct directions of how to do them and what to do?

I will be attending the basic section. The remaining time I will be reading wisdoms of the crowds as one of my projects.

12/09/18 @ 17:54
Comment from: [Member]

Thanks John! I will be attending the basic section of class, as well as section 1 and section 3. I did not understand the Pollster Ratings very well, but I did look at the survey they did about the Mexican caravan.

12/09/18 @ 19:18
Comment from: [Member]

I will be at the basic section as well as section one.

Grading of pollsters: The internet plays a big part in this. What does ‘live’ mean? Live on TV, live on the internet or both?

Top rated poll: It was interesting how for the first question the top answer had mostly over 50% but a lot of the questions the first answer was under 50%. why was that? Also, I notice that most of the questions are pretty touchy subjects.

methodology: I had to read this multiple times to understand some of it. One part that stood out to me was: ‘ Re-attempts are made to busy signals, no-answers, and answering machines’ Does that mean that if you don’t pick up they will keep bothering you? What if you pick up hear it a survey and hang up. Will they keep trying till you do it?

12/09/18 @ 19:30
Comment from: [Member]

I have just posted my deadlines for all of my projects.

I will be attending the basic section and Section 1.

So I looked at all of the graphs and polls and particularly 2. the interesting part that I found while looking at the different polls was that there were some strong differences between age, party, and location of living. I noticed specifically that the drastic age gap of 18-34 and 65+ had the most clashing and different beliefs. For example on immigration.

With the methodology one I didn’t even know that surveys were even done anymore by calling someone’s telephone. Wow things have changed a lot for me to say that.

I am still just interested in how to get data that doesn’t have some sort of bias, because that seems like it would be hard to do.

When I am not in class I will be finishing up my calculations for the last project I did.

12/09/18 @ 19:51
Comment from: [Member]

Thank you John. I will be in class as well as attending the excel portion.

12/09/18 @ 20:37
Comment from: [Member]

Thanks John! i will be attending the basic section of class tomorrow. And possibly the excel section, depending on how my working on it independently goes…

12/09/18 @ 21:17
Comment from: [Member]

Thank you John, I will be attending the basic part of class along with maybe the excel portion.

12/09/18 @ 21:29
Comment from: [Member]

I will be in the basic section, section 1 and section 4.

My very first question in reading this was how do polling firms make money. Turns out the political campaigns that the polls are about hire pollsters.

I also looked up TCJ Research, 538’s worst rated polling firm. TCJ Research’s rating of +2.8 is poor partly due to the fact that they only use IVR (Interactive Voice Response) for their polling. I also visited TCJ’s site and their most recent update is a poll report from 2012 showing Mitt Romney ahead of Barack Obama at 51%. Oof. Also found a funny article on The New Yorker’s site that mentions TCJ Research.

https://www.newyorker.com/humor/daily-shouts/welcome-to-nate-silvers-haunted-house

I also found it interesting that SurveyUSA differentiates between respondents who have already voted and those who are likely to vote but have not yet done so.

12/09/18 @ 21:31
Comment from: [Member]

I will be attending section four.

I made the first model…the second is in crumpled up mess…45 minutes of my time down the trash…

12/09/18 @ 21:44
Comment from: [Member]

I will be attending section one, two, and four in addition to the basic section.

Preparation for section 1:

I am not quite sure how to analyse the data in the polling sites, but I am exited to lean more about this on Monday.

a. I think that the “percentage of races called correctly” is a cool and helpful piece of information from the pollster rating site.

b. Something interesting about the methodology statement is the inaccuracy introduced into polling by asking for certain household members.

_____________________________________

Preparation for section 2:

I checked out the exercise sheet and I can see some patterns. Is it related to Galton board?

_____________________________________

Preparation for section 4:

I slaved over the construction of the tetrahedron skeleton for three hours with underwhelming results… :-( But I have a completed tetrahedron skeleton, it’s just not quite as sharp as I had hoped.

During the time of section three, I will be working on one of my projects.

12/09/18 @ 22:01
Comment from: [Member]

I will be in the basic section tomorrow as well as class one and two. I have been having a lot of fun comparing the categorized responses to the different questions (44% males and 46% females have been following news stories about the ‘migrant caravan arriving in Tijuana with the intention of seeking asylum at the US-Mexico border in San Diego’!!) I also find it interesting that the youngest male is requested on about 30% of calls to home phones from Survey USA.

Thanks John!

12/09/18 @ 22:06
Comment from: [Member]
Cayleigh

I will be in section one.

Top rated:
I found it interesting how on none of the polls listed did they have sufficient data from the black population. I’m wondering why that is.
Also on the polls regarding immigrants, the categories education and income were very similar throughout. The party and ideology was where there was a huge split between answers. The latter is not surprising, but I would have guessed education/income would have more of an impact on how people felt about the question.

Methodology:
Order bias is a new concept to me. It makes a lot of sense why the order of questions is influential, but it never occurred to me before.

12/09/18 @ 22:16
Comment from: [Member]

I guess I will jsut be attending the basic class and also maybe the excel class….

12/09/18 @ 22:35
Comment from: [Member]

My new major project is going to be the Fibonacci sequence.
I’m going to spend a lot of time researching it and then make some kind of presentation or project out of it.

12/10/18 @ 10:38


Form is loading...

December 2019
Sun Mon Tue Wed Thu Fri Sat
 << <   > >>
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31        

Search

Random photo

Second Guesses

  XML Feeds

Content Mangement System

©2016 Laurel Charter School