Friday ~ 18 November 2016
What did we do?
Today’s studio focused on visualization. For this activity, we were required to use Tableau, a software application that allows organizations, companies, and others to create a visual of their data, from line graphs to pie charts to maps. In our case, we created visuals for a data set of 911 calls and responses in Seattle. Our task was to create useful visuals that will help us analyze when and where it would be safe to plan a parade around U-District to celebrate the UW Huskies winning, specifically what day of the week and what time of the day, catering to both the safety concerns of the parade facilitators and the peak attendance of UW students (when are they most likely to be on campus).
The very first thing we did was to get familiar with the particular data we were working with. Here, we concluded that the data contains the 911 calls during 2014 and 2015 grouped and sub-grouped into categories, such as traffic related violations, disturbances, property damage, etc. All calls were labeled according to district and contained dates and times of each call. We first created a symbol map that visualized the location in the Seattle area that the 911 calls occurred and the frequency of these 911 calls. We narrowed down the 911 calls to “traffic related violations” in order to analyze where pedestrians are least likely to be harmed by a car.
Next, we created three different visualizations to help plan what month, day, and time the parade should be on. For these visualizations, we set restrictions to what set of data would be included. For example, because the facilitators wanted to plan a parade on a day and at a time where more UW students would likely be on campus, we restricted our data to include only weekdays (Monday-Friday) from 10am-5pm. Additionally, we decided to only include data from September to December, during which college football season is occurring. Below are the graphs we made, each visualizing the frequency of traffic related violation 911 calls based on hour, weekday, and month, respectively.
After studio, we were assigned a deliverable where we were to focus on a user, also known as a persona, of our choosing and create a research question that we assume they may want to consider. In my deliverable, I chose to focus on the Cascade Bicycle Club. In this situation, they wanted to organize a bike ride for all bicycle enthusiasts and wanted to know the best route to plan around U District to ensure safety and limit as much potential dangers to participants. I created three visualizations that compliment each other very well, working together to answer the main research question. Similarly, working with Tableau is still a bit unfamiliar and became a challenge initially. As time went on, the buttons and features of Tableau gradually became a bit more familiar.
Link to Tableau Public: https://public.tableau.com/profile/jamie.luz.villanueva#!/
Reflection on Experience
- What surprised me? It was a bit surprising to me that the restrictions and filters you place upon a data can affect the visualization of that data. For example, my partner and I initially created a tree map for traffic related violation calls according to weekdays that included all data from all twelve months and from 2014 and 2015. When we did this, the map told us that Tuesday is generally the safest, with the least amount of traffic related 911 calls. However, when we edited the filters to be more specific and only include data from September to December, quite the opposite was shown, with Tuesday having the largest amount of traffic related 911 calls.
- What challenged you? It was definitely challenging working with Tableau. It is a software application that I am unfamiliar with, therefore, it was a bit difficult learning how to maneuver through it.
- What would you like to explore? I would definitely like to learn more about Tableau and visualization in general, because I believe that it is a useful way to delve in deeper in what we are trying to research.
- One question that I would like to explore is: What consequences arise when we fail to reveal data at various levels? When we fail to reveal all data at various levels, we tend to distort the meaning of the data. It can cause the outcome of the data to reverse or differ from one another. The issue of respect and beneficence comes into play. Is this going against the ethical principle that we must disclose information about and the (potential) risks of the research to participants, citizens, and the like? The best option is to make sure there is a variety of visualizations used, from general overview to specific details.
What did you enjoy?
I enjoyed the final product. Having to go through various challenges in creating visualizations allowed me to be excited and relieved once we succeeded in creating those visualizations of data. It was also interesting to look at data in various ways.
As previously mentioned before, I believe Tableau and visualization in general is important in understanding what we are trying to research. It allows researchers to visualize not only the bigger picture, but also the details within a data. This is useful for future research in business, architecture, engineering, and many more.