Data Viz Thoughts .|: Reflecting on my most viewed and controversial viz + Tushar More has our #VOTW
A Tableau-centric weekly blog about the viz making process, #datafam member interviews, #DataVizThoughts Viz of the Week + entertainment for introverts (consisting of a music morsel & a binge bite).
This week we will share thoughts on one of my vizzes + Tushar More (@tusharz_more) Donations Accepted by Political Parties in the UK is our Viz of the Week (#VoTW)!
In late June 2018, the City of Madison, Wisconsin opened Tree Lane Family Apartments (TLFA) using a Chicago management company (and property owner). Its intent was to build a property to house families that were homeless or very close to living on the streets; most families had multiple children at elementary level or younger. The thought is dandy and we should fully address these issues, but the problem is when you concentrate poverty without proper support (1) — news flash… it resulted in a terrible situation for the families, schools and neighbors.
Right after it opened, there was a lot of crime in a neighborhood that had experienced very little crime leading up to the move in (45 mostly non-violent offenses in a 0.25 mile radius from the property between June 2017-May 2018, per LexisNexis). Even more, the new crimes were violent, with multiple 30+ people fights, gun/weapon issues and much more that resulted in a severely heightened police presence. This became increasingly worse until a woman was shot through her door by a person that shouldn’t have been on the property. After this and a couple of public meetings in December addressing areas of difficulty, the city finally put pressure on the management company to act on these intense issues. Since then, the incidents have decreased in seriousness and volume.
What does this all have to do with the viz?
I live very close to the property. We had a public meeting shortly after it opened (2) because of the crime concerns and an executive from the Chicago company glibly shared that crime always starts off high, but then decreases once residents settle in (without any time frame or whether ‘peak’ crime had already arrived). I asked for supporting data and he could not come up with it, so that is what started me on the year long viz creation path to seek an answer to the question — will crime just decrease just because families settle in (as claimed)?
Besides my question, there was a bit of hysteria from the unknown because specifics were light on crime, the large police presence, the city and company’s plan to address the issues and etc. I wanted to separate fact from fiction — the local media generally reported on calls for service [all reactive and pro-active calls to the Madison Police Department (MPD) and provide a little detail about most of the calls and their nature]. To break it down, I needed to make (monthly) open data requests. Like many open data records requests, it came with a bucketful of complications.
In order for me to grab the data, I needed to request calls for service (of which approx. 30–40% are criminal and 60% are usually either parking issues, 911 hang ups or foot patrols) and type them in a spreadsheet — you read that right.
The MPD had no way to share a spreadsheet because of their sadly antiquated system. Even worse, the data is an image .pdf and not something that could be converted to text (e.g. readable in Tableau). Another consideration was how can crime be measured; since there is a lack of information to determine who did what crime, crime specifics besides general ‘type’ unless it was widely reported or crime committed by residents (or guests). Via the powers of deduction and the grayskull (3)… the only reasonable method to gain insight is looking at crime specific to the property.
I also wanted other ways to that add context besides month over month comparisons, so I created calculations for time of day and weekend with a full report on the bottom to share all data.
The viz was originally shared on NextDoor, the Madison Police Department and city officials (monthly from August 2018). At 1st, it was fully supported by the NextDoor ‘community’… but after a couple months, some well-meaning advocates (many not very close to the property) were upset because I was using the specific address & that this was my way to contribute rather than visit, donate and whatnot. (4)
To them, I explained the purpose of the measurement and the complexity of typing in and reviewing each item from a .pdf to add to the sheet — I even went as far to request people grab the data, asked for them to contribute to the data collection & I would be happy to map out, but predictably nobody wanted to assist. Regardless, I continued to post to monitor crime activity. After a couple months of back and forth and making it clear this was only looking at the data and the management’s claim and measurements help immediate neighbors, the MPD and city officials best understand what was going on crime wise (and whether it was improving), the aggression toward me ceased (especially when there were improvements with crime and its display on the viz).
What about the viz?
My most viewed viz is a bit of an eyesore. Back then, I was more focused with simple layouts with interactivity rather than the attractiveness of the design. The color scheme was intentional to highlight the specific types of crime (and the color mess was purposeful to show how eclectic the crime was (I KNOW it’s not best… but I’m a little bit of a maverick who was making a point).
The data showed that the most common type of crime was a disturbance [which could mean anything from a physical altercation (e.g. the 30+ person fight), loud argument and anything in between]. 2/3rds of the crimes occurred between 12PM and 12AM and a similar amount during weekdays vs. weekends. For the entire year, there was a crime every 2.01 days on the property vs. every 8.11 days in the 0.25 radius the year before — it dropped to every 3.58 days in May and June (and tilted more to ‘petty’ crime).
This improvement was due to the city adding a lot of resources to support the property and making the property owner/manager accountable. (5) They had a deputy mayor (one of five deputy mayors in a city of around 250,000…) working directly at the TLFM, changed the supportive care (added a significant budget), organized a detailed plan with the Chicago management company and held weekly meetings with the company & filed a nuisance abatement action against the property manager/owner (for its negligent handling of these issues). Several months of focused effort and accountability reduced crime and my question was answered.
Creating a housing project that concentrates poverty is a significant challenge with huge impacts on the residents, neighbors and surrounding business — engaging the neighborhood (which the City of Madison admittedly failed to do), proper planning including supportive care funds (which the city also admitted was an oversight), using a property manager that has business in the city rather than a company that came from out-of-state can help reduce the negative impact and garner widespread support — the city can better protect their residents from such crises in the 1st place by only providing tax credits to builders who build actual affordable housing or require all new developments to be mixed income complexes. In addition, they can significantly expand transit services to outlying areas that have more affordable rent (and to reduce the burden on an already over-congested belt-line). (6)
Viz of the Week
Tushar More’s MakeOver Monday viz is complex, but chosen for its elegant design and beautiful flow of information in a buoyant bite-sized layout.
Adam Mico (AM): When you received the data and reviewed it, how did this design come into being? Was it something that you wanted to do for a while, was it mapped out before you created the viz or was it inspired while working for with the viz?
Tushar More (TM): After reviewing the data, I thought to create a viz that fans from summary to detail. I tried a couple of designs and was getting so frustrated I almost reached the point to not to participate this week, as I was not happy with the (previous) design. Then, this design came to my mind and it was something I wanted do for a while.
AM: You referenced Klaus Schulte as your inspiration for the viz. What viz inspired that and how did you apply it to this viz?
TM: I am a big fan of the Klaus’ designs and always love his work. He created UK pet population viz for one of the MakeOver Monday challenges last year. I really like the way it flows through the details. Therefore, I took the flow part of it in my design and added few other elements such as filled bar graph and line chart.
AM: Did anything surprise you with this data set as you were working with and how did you deal with it?
TM: One thing that surprised me was my values were not matching initially with the original viz. It was quite a wide data set, so I somehow missed one column that I needed to filter on. I dug in to some other community vizzes to get to the root cause.
AM: What was the most challenging aspect of this viz and how did you overcome it?
TM: The most challenging part of the design was creating those curvy lines. I knew it required some data preparation and complex calculations. Again, Klaus came to the rescue. I downloaded viz & it help me to understand what he did. I also referred his blog on creating tournament flow chart. These two things bits gave me the tools I needed to create my version of the curvy lines.
This week shares an aggressive live rendition of a pop-rock classic from Fleetwood Mac. Stevie Nicks wrote this song while she and Lindsay were floundering with their previous band, Buckingham Nicks. Unlike the radio version (and fueled by more than coffee), she progressively warbles the words into a manic froth that intensifies into a cathartic chant and rocker — the band ebbs and flows with bombastic grooviness.
I’ve binged on a number of great series spanning countless genres, but there is no series that captures me like The Americans. Its basis has some truth of the KGB sending spies to the USA to pose as ‘normal’ Americans in order to provide data back to the Soviets. Adroitly applying creative license to this scenario, the show covers an ‘American’ family involved in all sorts of operations aided by incredible acting, primo writing, an intriguing backdrop and the perfect soundtrack. The entire series is free on Amazon Prime (NOT kid friendly — intended for mature audiences).
1) For months had little to no furniture — the neighbors had no idea until the MPD posted a request for donations in December and helped provide them with additional necessities.
2) The meetings before it opened were very dismissive of neighbors concerns of a small living space, potential crime, lack of activities for children, lessening of green area and potential damages to home values
4) I generally do not engage with people on a IRL/social level. Other people are better equipped and benefit from that; these situations usually just offer anxiety and it’s not a treat for my audience either. Donating is something I do privately and not because it’s shoved down my throat.
5) The residents aren’t to blame; from what was shared at numerous meetings, many of the residents of the apartment were looking for a fresh start and lived in fear because the rampant crime at TLFA. Also, it was clear the management company and city did not do a lot of vetting to look at people that would excel in this environment or have a plan to address issues (e.g. evictions) when the non-vetted family members and guests were causing repeated harsh crimes.
6) This article goes into Madison’s housing crisis and plans to correct it.