Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added 02_activities/assignments/JASP_Boxplots.pdf
Binary file not shown.
Binary file added 02_activities/assignments/JASP_workbook.jasp
Binary file not shown.
Binary file added 02_activities/assignments/Python_Line_Graph.pdf
Binary file not shown.
19 changes: 5 additions & 14 deletions 02_activities/assignments/assignment_2.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,24 +10,15 @@
- For each visualization (good and bad):
- Explain (with reference to material covered up to date, along with readings and other scholarly sources, as needed) why you classified that visualization the way you did.
```
Your answer...






Bad visualization: https://www.flickr.com/photos/laurenmanning/5658951917/
For an example of a bad visualization, I picked a bar chart that is supposed to show how much French fries people consume for each month of the year 2010. The bars are represented as fries. The x-axis are the months and the y-axis is the consumption amount. Although it is creative to use fries as bars which adds visual appeal, this chart has several problems. First of all, it is unclear where this data comes from. There are no sources provided and it is not clear at all what the data is supposed to even represent. As a result, we cannot tell if the visualization accurately represents data. Secondly, the y-axis doesn't have any label. It is unclear what the numbers are supposed to represent. Do they represent the raw number or percentage of French fry packs? How was this calculated? The visualization raises more questions than answers. Thirdly, the y-axis numbers are so sparse that it is hard to see which number the bars are at. Another issue is that the bars which are shown as french fries are of irregular shape, making it difficult to see which number they represent. Overall, even though this visualization is somewhat visually appealing and quite simple, it is difficult to understand.

```
- How could this data visualization have been improved?
This visualization could be substantially improved by adding appropriate label to the y-axis, making the y-axis less sparse, improving the appearance of bars (make the fries more straightened and less irregular in shape), and add a source.
```
Your answer...






This is an example of a good visualization: https://www.columnfivemedia.com/work/ncsa-how-far-do-college-athletes-travel-to-play/.
For an example of a good visualization, I picked a bar chart that shows average miles traveled by sport. This visualization is good because it has a source that provides some information about the data, which makes it easier to verify if the data is accurately represented or not. Unlike the previous bar chart, it is clear what numbers the bars represent and what these numbers mean (miles) because the exact numbers sit on top of each graph and the title specifies that they are measuring miles. The bars are also color-coded: purple represents males and orange represents females. This way the average mile traveled is shown separately between males and females for each sport. Some sports are dominated by women and other sports by men. This allows for a more accurate and fair assessment. Overall, this visualization is not only visually appealing but also easy to understand. It has all the information that a viewer needs without looking too busy.

```
- Word count should not exceed (as a maximum) 500 words for each visualization (i.e.
Expand Down
56 changes: 40 additions & 16 deletions 02_activities/assignments/assignment_3.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,23 +7,47 @@
- Choose a dataset of interest from the [City of Toronto’s Open Data Portal](https://www.toronto.ca/city-government/data-research-maps/open-data/) or [Ontario’s Open Data Catalogue](https://data.ontario.ca/).
- Using Python and one other data visualization software (Excel or free alternative, Tableau Public, any other tool you prefer), create two distinct visualizations from your dataset of choice.
- For each visualization, describe and justify:
> What software did you use to create your data visualization?

> Who is your intended audience?

> What information or message are you trying to convey with your visualization?

> What aspects of design did you consider when making your visualization? How did you apply them? With what elements of your plots?

> How did you ensure that your data visualizations are reproducible? If the tool you used to make your data visualization is not reproducible, how will this impact your data visualization?

> How did you ensure that your data visualization is accessible?

> Who are the individuals and communities who might be impacted by your visualization?

> How did you choose which features of your chosen dataset to include or exclude from your visualization?

> What ‘underwater labour’ contributed to your final data visualization product?
Visualization #1
> What software did you use to create your data visualization?
I used VS Code to create this visualization (the file name is "Python_Line_Graph").
> Who is your intended audience?
My intended audience is people in healthcare including government workers.
> What information or message are you trying to convey with your visualization?
My visualization shows how many people died in shelters across years separated by sex.
Specifically, we can see from the graph that the number of decedents for both males and females decreased again after the pandemic years. We can see that there were many deaths between 2019-2022, especially among males, in homeless shelters. Another interesting observation is that males died substantially more in pandemic years than females.
> What aspects of design did you consider when making your visualization? How did you apply them? With what elements of your plots?
I wanted to ensure that we can see the number of deaths across all years and how it was changing (when the highest and lowest number of deaths were), so a line plot would be ideal in this case. I also wanted to show the male and female data separately so I plotted two lines of different colors (pink for female and blue for male). The chosen colors would also make it easier to understand and remember the graph since pink is associated with females and blue with males. The shade around the line is standard error of the mean. I also didn't use any strong language or add any arrows/pointers anywhere to ensure the visualization looks as neutral as possible.
> How did you ensure that your data visualizations are reproducible? If the tool you used to make your data visualization is not reproducible, how will this impact your data visualization?
My data visualization is reproducible because the raw data and code are easily accessible and I also commented my code. If people access the code and view the comments, which would help them understand how the data was analyzed and what each step means, then they should be able to reproduce it.
> How did you ensure that your data visualization is accessible?
The code and the data are both accessible to public. They can be found on GitHub.
> Who are the individuals and communities who might be impacted by your visualization?
My visualization indirectly points that the government support for people in need is susceptible to situations like lockdowns (although the deaths may also be linked to covid itself). Health care individuals are also impacted by this.
> How did you choose which features of your chosen dataset to include or exclude from your visualization?
I didn't include the months and individual data. This data isn't very informative when it comes to understanding how people in shelters are doing over the years and whether there is any progress.
> What ‘underwater labour’ contributed to your final data visualization product?
Initially my data was in wide format (there were separate columns for males and females) so I had to manually change it to long format. I created a separate column called "Sex". With the wide format, I wouldn't be able to run my code the way I did.

Visualization #2
> What software did you use to create your data visualization?
I used JASP to create this visualization (the file name is "JASP_Boxplots").
> Who is your intended audience?
My intended audience is people in healthcare including government workers.
> What information or message are you trying to convey with your visualization?
My visualization shows how many males and females die in shelters in the last 3 years (2024-2026). The main focus is the difference between males and females in number of deaths in shelters after the pandemic. Males in shelters are more likely to die than females even after the pandemic.
> What aspects of design did you consider when making your visualization? How did you apply them? With what elements of your plots?
It is hard to design plots on JASP as there is no option to code. However, I tried to create a boxplot which each box representing sex. I also overlaid it with a violin plot to show where the individual data points mostly lie for each sex. The plot is color-coded for an easier visualization. I didn't use any strong language to ensure the visualization looks as neutral as possible.
> How did you ensure that your data visualizations are reproducible? If the tool you used to make your data visualization is not reproducible, how will this impact your data visualization?
The visualization is reproducable because I saved the JASP workbook with all the raw data and analyses I ran and made it accessible on GitHub.
> How did you ensure that your data visualization is accessible?
The raw data as well as the JASP workbook are publically avaialble on GitHub.
> Who are the individuals and communities who might be impacted by your visualization?
In this case also it would be the healthcare workers especially those within the government sector. We want to understand why males have such high mortality rates: are they overlooked in some way? This is an issue even in the recent years.
> How did you choose which features of your chosen dataset to include or exclude from your visualization?
I excluded the pandemic years and the years before that because I wanted to visualize how the data looks in the most recent years under normal conditions (without the influence of other external factors like covid and lockdown). If the data looks concerning even under normal conditions, then there are serious systematic issues that must be addressed. I also excluded the months because it's not important for the question I am asking.
> What ‘underwater labour’ contributed to your final data visualization product?
On top of converting the data from wide format to long format and creating a "Sex" column, I had to also delete the entries corresponding to the years before 2024.

- This assignment is intentionally open-ended - you are free to create static or dynamic data visualizations, maps, or whatever form of data visualization you think best communicates your information to your audience of choice!
- Total word count should not exceed **(as a maximum) 1000 words**
Expand Down
117 changes: 117 additions & 0 deletions 02_activities/assignments/assignment_3_code.ipynb

Large diffs are not rendered by default.

23 changes: 23 additions & 0 deletions 02_activities/assignments/lecture2_notes.ipynb
Original file line number Diff line number Diff line change
@@ -0,0 +1,23 @@
{
"cells": [
{
"cell_type": "markdown",
"id": "81aa2bc8",
"metadata": {},
"source": [
"Lecture 2 activity notes. Reviewing different types of visualizations. \n",
"\n",
"-Visualization #1: From this visualization, we can learn by what age people were killed by gun violenece and how many years were stolen from them as a result. We can also learn how each of them was killed by clicking on individual line as the visualization is interactive. But this visualization is not neutral or objective as they use strong words like \"years stolen\" when we don't even know how long each of those poeple would live. It also provides a reason for killings to influence on people's emotions even more. \n",
"\n",
"-Visualization #2: From this visualization, we can learn how many active shooters there were in each year. This visualization is more neutral than the previous one but it's unclear what they mean by \"active shooters\". Some people may interpret \"active shooters\" differently, which is dishonest. Also, we don't know which state or city this visualization represents."
]
}
],
"metadata": {
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
233 changes: 233 additions & 0 deletions 02_activities/assignments/visualization_lec1_code.ipynb

Large diffs are not rendered by default.

471 changes: 471 additions & 0 deletions 02_activities/assignments/visualization_lec2_code.ipynb

Large diffs are not rendered by default.

281 changes: 281 additions & 0 deletions 02_activities/assignments/visualization_lec3.ipynb

Large diffs are not rendered by default.

1,182 changes: 1,182 additions & 0 deletions 02_activities/assignments/visualization_lec4.ipynb

Large diffs are not rendered by default.

Loading