Greater Seattle Shoplifting Visualization
Foundation of the Case Study
On the second week of May, I reached out to the local “CanYouID” organization for all of Washington State where the community helps catch shoplifters through identifying their pictures taken from CCTV, security camera and phone cameras. The purpose of this connection was to research what is their purpose and to offer them my experiences to narrow down the areas where most of the shoplifting occurs in the hopes of increasing the security and the response time for the police to arrive. By collecting “Shoplifting” data from the 911 Incident Response Dataset it would help them identify certain “hotspots” for these crimes and potentially reduce the amount of shoplifting in those areas by catching person in interest, and potential suspects.
I’ve contacted the organization to find out what their cause is and research questions as well. The communication was a 30 minutes’ phone call and the research questions were the following: “What districts and area experiences the most shoplifting? How frequently these shopliftings occur and if so is there a pattern they follow? When does the average shoplifting occurs (hour of day) in the most shoplifting area throughout Seattle and its neighboring villages? Last but not least, they were also curios about how the previous year data (2014) compares to the data we were provided with (2015). This was an extra step that they assigned but it’s a bit outside of the scope for the assignment, nonetheless I still provided the data in Tableau.
Throughout the visualization process I put my Tableau skills into use which allowed me to carefully extract data from the data.seattle.gov website specifically from the City of Seattle’s 911 Incident Response database from 2014 and 2015. There were over 450,000 variety of data points where through several filtering I was able to condense it to only the ones that involves shoplifting. This really gave me the ability to build up several visualizations that not only helps the can local CanYouID organization but it also enhanced my data wrangling expertise. Tableau was quite a bit complex but that’s exactly what I liked about this challange, thus the tutorial that was provided with the case study helped me orientate through the obstacles and generate my data into a helpful concept that hopefully saves our community millions of dollars worth of stolen products.
Stepping Stones for Sketching
As with Usability Testing and Prototyping, Visualization is just as important to show an illustrated data to the public that they can conceptualize and understand without ever thinking about the vast amounts of data and numbers. By making data work, in other words, converting data into sensitive visual information it helps with complex situations that could prevent further damage to our surrounding communities. Of course this can be quite tricky because rarely the data could be either biased or compromised which would not be an approrpiate step if an organization or a comapny wants to receive accurate feedback and statistics. One example would be of why I have choosen shoplifting. In the past I worked as a barista and I observed hundreds of shoplifter cases per year which influenced my curiousness to help my community, not only by preventing further loss of profit (U.S retailers lose +$60 billion due to shoplifting), but by also making it more safer and trusted. Thus by using my past experiences and the datapoints that are available to me by the public I could change the future simply by creating my visualization and contribute it back to the public which communicates that certain areas needs more security at specific times of the day. Therefore I will continue to put emphasis on visualization in my future case studies due to the weight and influence it can deliver to the public and possibly even change the way they think about certain events or even change their point of view.
Tableau Public for the Greater Seattle Shoplifting Visualization I’ve created: https://public.tableau.com/shared/CXKWFXFN9?:display_count=yes