Black Lives Matter Platform: A Visual Companion

[ghost]

For this pitch, I am collaborating with @gcgruen and @HarshaDevulapalli.

Pitch

Possible headlines:

-A Visual Companion to the BLM Platform

I would like to begin to make visual supplementary material for the Black Lives Matter platform.

This week I will focus on pre-school disciplinary statistics, which are discussed here in what I feel is a very static way. I'll try to make the visuals as compelling as possible. I will also implement the strategies for statistical analysis that we have touched on in the Algorithms course so far.

Data

I'll use the data they reference.

I've already had a few issues with the data, upon first glance:

The shape of the data frame is challenging. There are 1929 rows describing the different attributes in the “contents” .csv file, and subsequently there are 1929 columns in the actual data. 95,507 is how many schools participated in the data, one row for each school. Each row has 1929 columns. This is kind of a nightmare. It's difficult to work on without first defining your scope. So, this week I am only working on pre-school disciplinary topics!

df.shape output: (95507, 1929)

Claims:

-Black public preschool children are suspended from school at high rates: Black preschool children are 3.6 times as likely to receive one or more out-of-school suspensions as white preschool children.

-Black children represent 19% of preschool enrollment, but 47% of preschool children receiving one or more out-of-school suspensions; in comparison, white children represent 41% of preschool enrollment, but 28% of preschool children receiving one or more out-of-school suspensions.

-Black boys represent 19% of male preschool enrollment, but 45% of male preschool children receiving one or more out-of-school suspensions.

-Black girls represent 20% of female preschool enrollment, but 54% of female preschool children receiving one or more out-of-school suspensions.

(This is analysis from the US Department of Education Office of Civil Rights, not my own)

Clip search:

Black students are less likely to attend schools that offer advanced coursework, less likely to be placed in gifted and talented programs, more likely to attend schools with less qualified educators, and employ law enforcement officers but no counselors. "

Black students more likely to be suspended

https://www.theguardian.com/education/2016/jun/08/us-education-survey-race-student-suspensions-absenteeism

Story issue checklist

My pitch was (use the number): #220

• [ ] My pitch has been approved (see PITCHING.md)
• [x] My story issue links to my pitch issue
• [x] I link to my finalized (ish) data source(s)
• [x] I've included a brief summary of my story
• [x] I've included some possible headlines or findings
• [x] I've included some links or images as inspiration (if you have any)
• [x] I have received two comments of peer feedback
• [x] I've included an update of my visualization/story in a comment
• [x] I have received two comments of peer feedback after posting an update
• [ ] I have received editorial feedback

[playfairbot]

Hi there, I'm the Playfair Bot!

Thanks for posting your story issue, but would you mind adding editing the original issue to add the first draft of your image? You have my sincere apologies, but it's easier for dumb robots like me when the comments are only used for updates.

Thanks! :pray:

[ghost]

I calculated the percentage of disciplined white children and black children in preschool and made a histogram comparing the two. Values are normalized by enrollment percentages.

[gcgruen]

I think for these graphics is crucial that you or we come up with example sentences how to read them! But I think this way we could make people understand also more complex graphic examples.

To show the rates of students being diciplined I would suggest sth like boiling them down to 100 students. At my former magazine we did a piece "If the world were 100 people" which was out most successful piece ever. We could do the same here (maybe with 10). Of 10 students, 2 are black, 4 are white, 4 are other (all same shape, different stroke colour). Out of these students 1 black is disciplined, 1 white is (fill them, leave the undiciplined unfilled). You could put each ethinicity in a separate row. That's not really a dataviz but more an infographic, but maybe a nicht (and easy) addon? I could take care of it, if you like. In case you would like to do it yourself but don't know where to get shapes from: I highly recommend The Noun Project

[ghost]

...I re-did the stacked bar graphs for black preschool children:

and for black female preschool children:

I'm experimenting with a rainbow palette to remind viewers that these are 4 year-olds...

[gcgruen]

I think the stacked bar chart was a good graphic choice over the slope graph! I also like the bright colours of the later version more than the one in the previous stacked charts which are more translucent. I would suggest a tiny tweak though (since we are mainly interested and otherwise only focusing on this): make only three groups: black + white + other The other ones are tiny and unreadable anyway, you really don't want to focus on Hispanics in this projects because it would weaken the message of BLM if you have to add (Hispanics, too!!!).

I hope giving your histograms the same bin size works out, because I really think that it is crucial. Also, I would make it more reasonable splits, maybe only full 1%, 2% than what you have now (0.25, 0.5, 0.75, 1 etc - I believe)

But #yay that this is finally coming to life and in an enjoyable shape! :)

[ghost]

In making my histograms, I decided to only look at schools with more than .05% black enrollment, and more than .05% total discipline rate.

This highlights a clear point: that preschools with high discipline rates are much more likely to be punishing black children, and especially black female children, at disproportionately high rates.

[gcgruen]

Yay, d3! I look forward to see this transitioning :) And I kind of start liking your rainbow colors ;)

[mercye]

the colors struck me as very bright and cheerful, which on the one hand is cool to me because its about preschoolers. i am not sure if they're TOO bright, maybe muted versions of these colors? idk.

[ghost]

Thanks @mercye and @gcgruen for the feedback, changing the colours a bit to make them less garish and more intuitive? but sticking with rainbow, for now!

[ghost]

New graphics: Looking at how demographics change when preschools have higher overall punishment rates. I know the histograms are hard to understand, anyone have suggestions for how to make them less opaque?

[ghost]

Taking a different approach, zooming in on the parts of the disproportionality in population histogram that I want to focus on. I think it is clearer, but I'm not sure if it is misconstruing?

[mercye]

I am having a similar issue with scale/whether its misleading. i think that maybe visually emphasizing the y axis so that people pay attention to the number of schools in each group since the number of heavy handed schools are much less than the number of lenient schools -- it could be clearer.

[ghost]

@mercye I think this is better/clearer/less misleading!! PS I tried to make the colours less rainbow, and I just couldn't...

[ghost]

ps, horizontal gridlines, or no horizontal gridlines?

[ghost]

[playfairbot]

Closing since pull request #266 has been accepted