Beijing's air was notoriously bad a few years ago. How is it doing now?

[ksliney]

Pitch

Are local government efforts to Improve Beijing's air quality successful? Let's compare 2008 through 2017.

Summary

I want to plot the air quality in Beijing, China from 2008-2017. (Hint: It used to be bad! I want to know if it still is.) My data came from the US Embassy.

Details

Data set(s): http://www.stateair.net/web/historical/1/1.html Code repositoryhttps://github.com/ksliney/data-studio/tree/master/code

Possible problems/fears/questions: I want to better contextualize what these air polution numbers mean, in terms of impact to health. Data for 2008 and 2017 is incomplete, so the chart is not as up-to-date as I would like.

Work so far

UPDATE: I reworked the data, and found the average air quality reading per day, and then sorted each day by its level of small particle pollution and danger to public health.

You can see the interactivity here:

I found the average monthly air quality reading for the year 2008, and then did this for every year, though 2017.

Older work:

UPLOAD SCREENSHOTS

Checklist

This checklist must be completed before you submit your draft.

• [ ] I have already spent time with my data set, opening it, exploring it, etc
• [ ] I have created a "DIARY.md" file to save links and list all of the terrible, no good problems I come across
• [ ] My issue issue has [Project] in the title
• [ ] My issue links to my data set(s)
• [ ] My issue links to my code repository
• [ ] My issue explains what I'd like to explore in the data set
• [ ] My issue includes images - either inspiration or what I've done so far
• [ ] I have received two comments of peer feedback on my pitch issue

[ksliney]

UPDATE: Is Beijing's Air Quality Really Improving?

Any changes in direction or topic?

I re-worked the data to average AQI readings for each day (rather than monthly), for 2008 through 2017. I made the chart using google charts, which is interactive, though the static screenshot (hopefully) tells the same story just as well.

Problems/Questions

*When cleaning the data, I dropped any values in the dataset where the AQI detector errored and returned a negative reading. I noticed that many of the AQI reading errors I dropped were sampled during the winter months, when coal consumption is high and pollution is typically at its worst in Beijing. So the updated chart may misrepresent pollution levels due to lack of data and reliance on one detector for AQI values.

*If I were to re-work the data again, I would plot the AQI at a certain time of day (say, 8:30 AM, or 5:30PM) instead of plotting the daily average.

Checklist

• [X] I have included my visuals
• [X ] I have filled out the sections above
• [?] I have been updating my DIARY.md with details about my process
• [X] I have uploaded/updated any Jupyter Notebooks or other datasets into my code repository

[xeophin]

• I like the colour scheme of the chart, obviously a lot of thought has gone in there.
• The interactivity might not be necessary, as it does not give that much more information (and: in case it would give additional important information, it should not be hidden in a tooltip)
• It might be worth reconsidering the missing data. Does the missing data skew the trend in some way? Can the missing data be interpolated from the neighbouring readings? Do these errors happen under certain conditions? It might already be interesting to see them as part of the chart, in order to help know "what is not known".

[jsoma]

• I'd ditch the lines in between each section to simplify things, probably linewidth=0 or edgewidth=0 or someting like that
• Is moderate bad? If it isn't too bad, I'd also make it blue-ish
• Instead of 'days' as a label on the left, I'd add it with 365 days (...or maybe make that "1 year"?)
• You could also pretend months are all 30 ish days, and then mark "x months" on the left-hand side? But I guess it isn't technically correct. Just trying to think of a way to communicate time a little more clearly

[playfairbot]

Greetings! I'm a little robot, beep beep boop boop.

You need some feedback, let me summon @jessimckenzi, @Yuanqi-Hong, @Katerinavts for you

It looks like we need to fix up your your update a little bit! Edit it by clicking the pencil in the top right-hand corner. It requires:

• New, updated content for your project
• The checklist (be sure you're using the template!)

Maybe you just didn't use the template? If not, edit your comment, cut and paste the template in, and then fill it out.

[jessimckenzi]

This is fascinating! I do think the use of blue/grey is a very effective use of color for storytelling, but agree with Soma that the lines between the colors should go. I also find the grid lines distracting, and would prefer the graph without.

But the content is really interesting! Would love to know why the info about 2017 isn't available—is it just too soon to have been released? Last thought: If I'm not a very data literate reader, I might not know what to make of the graph, and you don't explain. Maybe the title should explain the graph: "Beijing's air quality improved between 2014 and 2016" or similar

[Katerinavts]

Good work! My only suggestion would be to narrow down the legend to 3 or 4 categories of air quality.

[Yuanqi-Hong]

I'm honestly interested in this pitch. I lived in Beijing (2014 - 2018), and the data for 2017 is really counterintuitive for me! People (classmates and colleagues) in Beijing were talking about how 2017 was a better year in terms of air quality compared to previous ones. I googled the topic and articles like this BBC story came up. I'm curious to know why the discrepancy.

[ksliney]

UPDATE 2: https://ksliney.github.io/airpolution/

UPDATE: Lots of time spent researching solutions to recommendations, and a few changes made

Here's an updated link to my notebook and data.

Any changes in direction or topic?

It turns out, I was interpreting the data incorrectly. The story actually is a bit bleaker than I realized for small particle pollution only. Because of this, you'll notice the number of blue "good" days has gone down dramatically in this later draft.

Problems/Questions

Unfortunately I am not yet able to answer my question regarding 2017. The US embassy's data for 2017 is (1) simply not present for all months (i.e., those months are not included in the data set), and (2) for the months that are present, there are days within those months that returned error readings that I had to drop from the data. So there's too much missing from 2017 to draw any big conclusions.

I hear all of you on there being too many color categories. These correspond to the EPA's own categories, but I realize that doesn't mean terribly much to a reader. I'm trying to figure out how much to convey, without simplifying too much or recreating the same table the EPA puts out. So that will be next.

Checklist

• [X] I have included my visuals
• [X ] I have filled out the sections above
• [?] I have been updating my DIARY.md with details about my process
• [X] I have uploaded/updated any Jupyter Notebooks or other datasets into my code repository

[collleenwang]

Excellent work! To make the y axis to see what happened in 365 days is very good idea! I think the blue part can be more brighter, so that we can see more comparable results.

[jlstro]

Hey, I know that all our instructors say that we shouldn't rely on readers interacting much, but I still like it a lot when something moves and changes and acts like it was interactive :-) Coll that you got it to work. There are a few minor things I'd do differently with regard to your final page (even though it is too late now, I guess?): It looks like the diagram does not align correctly, so it looks a bit weird postion-wise. Also, I have no idea what PM 2.5 pollution levels according to EPA standards means. I assume that each segment of each year is a sum of the days that fall into the respective category, but other than that, a subheadline explanation would help a lot. Your color scheme is cool, it really reminds me of dirty, dirty air.