Working With Data

For M 5/11

Professor Shortell — Mon, 11 May 2020 00:38:21 +0000

Let's take a look at this: https://medium.com/nightingale/covid-19-data-literacy-is-for-everyone-46...

We'll discuss data narratives. Each group will take a look at one narrative online. Answer the following questions about the narrative: (i) What is the story?; (ii) How compelling is the visual presentation of the story?; (iii) What is left out that might make the story more compelling?

A) https://www.pewforum.org/2019/11/15/americans-have-positive-views-about-...
B) https://www.pewforum.org/2019/12/12/religion-and-living-arrangements-aro...
C) https://www.pewforum.org/2016/12/13/how-religious-groups-differ-in-educa...
D) https://www.pewforum.org/2019/10/17/in-u-s-decline-of-christianity-conti...

What we'll discuss together is how to shift the story from the text to the visuals to make a more compelling dashboard. Our aim is a three page report that tells the data story clearly and persuasively.

Assignment 9

Professor Shortell — Mon, 04 May 2020 19:46:31 +0000

Make the model predict better! You can find additional reviews to better train the model, or you can extract additional features to include.

Write the necessary code. Explain your approach.

What can we learn from this classification task?

For M 5/4

Professor Shortell — Mon, 04 May 2020 17:16:43 +0000

We'll take one last look at our machine learning classification project and then discuss data visualization narratives.

Colab: https://colab.research.google.com/drive/1ChvenFc80B6ZqcLOp3dDtsbcszXSTthY

Activity
Each group will look at one visualization and discuss the strengths and weaknesses of the data story it tells.

1. https://atlas.cid.harvard.edu/countries/138
2. https://sportengland-production-files.s3.eu-west-2.amazonaws.com/s3fs-pu...
3. https://ourworldindata.org/trade-and-globalization
4. https://trendeconomy.com/data/commodity_h2

For M 4/27

Professor Shortell — Mon, 27 Apr 2020 14:42:00 +0000

Read: https://www.nytimes.com/interactive/2020/04/26/us/politics/trump-coronav...

Colab: https://colab.research.google.com/drive/1ZNWNWddXfW2Ie3xN4vh1cqO3ewSSohKu

For M 4/20

Professor Shortell — Fri, 10 Apr 2020 20:12:27 +0000

Read: http://www.bbc.com/culture/story/20180525-every-story-in-the-world-has-o...

Colab: https://colab.research.google.com/drive/1h9Io90PffOUHq3fvfawjnvhs2iKnVA_K

Activity
1. Parse a news feed.

2. Code for some content, including themes.

3. Visualize the results.

For M 4/6

Professor Shortell — Mon, 06 Apr 2020 21:09:14 +0000

Here is the notebook we'll discuss.

Which social media should we discuss?

Professor Shortell — Mon, 30 Mar 2020 20:02:13 +0000

Facebook

11% (2 votes)

Instagram

50% (9 votes)

22% (4 votes)

GitHub

11% (2 votes)

Web pages

6% (1 vote)

Total votes: 18

(M 3/30) Activity: DataFrames, Part V

Professor Shortell — Wed, 25 Mar 2020 13:58:46 +0000

Another notebook on merging data sources:
https://colab.research.google.com/drive/1ZoX3saX78GXNVUvGa2I__1vxfwVXt3C0

Then, we'll focus on analyzing Twitter data:
https://colab.research.google.com/drive/16Td0kwHPd4UPyEADcUZu0-EP40DxWo-W

For more information about text analysis in Python, I recommend:
https://www.google.com/books/edition/Natural_Language_Processing_with_Py...
https://www.google.com/books/edition/Applied_Text_Analysis_with_Python/l...

Activity
Create a DataFrame using a keyword or hashtag and visualize some aspect of the tweet data.

(M 3/23) Activity: DataFrames, Part IV

Professor Shortell — Sun, 15 Mar 2020 20:19:20 +0000

In this activity, we'll work with combining multiple data sources.

Here's the Colab link: https://colab.research.google.com/drive/1YMK75wWOLruhRGQVucGH1bsphhNoNnae

If there is time, we'll also look at scraping Twitter data.

Colab: https://colab.research.google.com/drive/16Td0kwHPd4UPyEADcUZu0-EP40DxWo-W

Exercise
Create a DataFrame with at least two of the parts of the global social indicators referenced here.

Create a linear model to explain variation on some dimension of gender inequality.

Visualize the relationship.

Write a one paragraph explanation of the insights from your model and visualization.

Add a comment with your notebook or link to your Colab notebook.

Change of Format

Professor Shortell — Wed, 11 Mar 2020 18:10:07 +0000

The university has announced that there will be no classes from 3/12 to 3/19, and beginning on the 19th to the end of the semester, all instruction will be online. So we're moving our work to a fully online format during this time of public health challenges.

I think this is a prudent course of action and I hope that all of us (and our loved ones) stay healthy during this unusual time for the city and the university.

We'll use the course site to interact in lieu of in-class meetings. Check the course site for more information. I'll create pages in which we can discuss the assigned readings. I would ask that each week, each of you answer three of the questions from the list. If you are responding to a question someone else has already responded to, try to add some additional insight in your comment.

I will also create programming exercises to do like the activities we've done in class. I would ask that you share your Colab notebook or upload your Jupyter notebook as a comment to the page in order to participate in those activities. I'll comment on your code and I encourage you to comment on other students' code as well.

In addition, I'll continue to post the weekly "outside of class" assignments. Please email your notebook or link to me directly, as we've done.

Finally, I will set up occasional Zoom meetings which you can join with audio (computer or phone) or a/v so that we can continue to interact as a group. I'll post the meeting information on the course site a week in advance so that you can plan your calendar. I have also set up and invited each of you to a private Slack channel, shortellnyc.slack.com, in order to provide for us another medium for interaction in the absence of face-to-face meetings. (We'll use the private channel workingwithdata on the Slack site.)

I will extend virtual office hours on the course site to cover the in-class meeting times: Mondays, 4:15 to 8:30.

We're in this together. Let's do our best to make it work!