New York is not an easy place for filming. The city is busy, restless, and pricey. So, what keeps filmmakers coming back? Because the city radiates unique energy: the rhythm, the curbside drama, the feeling that something entirely unexpected could be around the corner. Consider this visual a fan mail to the highly dedicated location squad, production staff, and other executives who bring it to life.
For this project, I want to develop a visualization that looks at the 65,112 filming permits issued by the Mayor’s Office of Media and Entertainment in New York City during the last ten years. Informing visitors on the volume of permits issued, which are arranged by season and classified into five boroughs. It covers permit issues dating back to January 2012. This should provide a decent comprehensive evaluation for each year, including season averages.
Inspiration & References
The image beneath is from my final assignment for my Programming User Interfaces class, in which my classmate and I worked together to code a website with an interactive map of filming locations in New York City. This was my major motivation for continuing to analyze this subject. There was no distinct dataset with the movie names, therefore we decided to create one on our own and looked through multiple sources, and compiled it into one dataset. Like many large, anticipated movies and sequels that use secret codenames or working titles to fly under the radar, we couldn’t find any dataset with a list of all movie names.
I sought inspiration from visualizations that differed from the theme and subject I was working on since I wanted to explore other possibilities and believed that only viewing visuals for the topic I was working on would be restrictive. The visuals below inspired me to build something unique. Because I wanted to organize the data by season, I came up with the concept of stacking bar graphs in a circular arrangement. The graphics are simple (added below), but they provide useful information. Despite the fact that my database is enormous, this visualization was an excellent example of displaying substantial information. Finally, I didn’t want my representation to appear complex while emphasizing seasonal trends.
The visualizations I discovered were a fascinating example for me to investigate graphs and bars, and the ones used here are suited to the material and do not mislead the visitors. It prompted me to incorporate graphs and bars into my visualization to offer more precise information about the data. The hierarchy is also clear, and the eye movement is quite natural. These visualizations are a good starting point, and they helped me understand how and what I wanted to achieve with my data visualization.
The data for this study was gathered from Filming Permits granted across the city’s five boroughs, which are given anytime a team wishes to film in a specific spot or have access to the streets without automobiles or other interferences, and this data comes from the New York City Open Data initiative. Tableau Public, a data processing application, provides an easy-to-use drag-and-drop method for producing charts and data visualizations. The graphics are then extracted from the local program and saved on Tableau Public’s site, where they may be viewed and shared with others.
I needed to start with a data set, like with any data visualization project. I checked over the Data and discovered a few potential problems. One advantage was that the specific areas I planned to research were in understandable formats. For example, I had planned to utilize the street closure data to pinpoint specific places in the city where filming took place. However, because the data in this column was recorded as street names and cross junctions in numerous patterns, the accuracy was adjusted to use zip codes.
Some of the data also contained many points, not only filming permissions. Multiple zipcodes separated by commas were provided in the field for some entries. Because this would result in inaccuracies chart layout, I divided these values by comma and only included the first value as the primary filming location. Now that the data was cleaned and ready to render, I created various worksheets in Tableau to understand different aspects of the data.
Visualization design explorations
The design objective for the project was to achieve the following goals:
– Visually explore relationships between the season and the filming requirements.
– Trend in preference of borough for the requirement of filming permits.
– Comparison over the years through seasons.
I couldn’t go straight from drawings and spreadsheets into Figma; I needed to compute a graph to accurately and efficiently smooth the visualization methodology. I wanted to evaluate the trends, possibilities, and overall data in Tableau. I also decided to investigate the filter choices, which led me to the conclusion that it is not feasible to examine the permits divided by seasons since the option to divide the months into quarters was confined to the automated grouping of the months based on their order in the calendar. I quickly realized that doing this for the next 10 years of data would be quite daunting, and I was unsure how this might be grouped by seasons. That caused me to manually establish a group inside the same months that comprise a season, as well as to manually construct a stacked bar graph using Figma and representative colors were assigned to each borough.
Unfortunately, some preliminary assessments indicated both challenges and possibilities. When I started structuring the stacked chart into a circular pattern, it became less legible and useful. It was more difficult to comprehend and comprehend the patterns. It didn’t appear possible to proceed with this design without leveling the space between the vertically stacked bars on the circular path. Considering I did not want to compromise on presenting information and data readability, I chose to arrange the bars in a straight line.
Even though I still had a long way to go, the initial experiment was instantly insightful and engaging. The Tableau spreadsheets were set up to filter all data by the borough where the filming took place, and for the month breakdown, an additional filter reveals the months to enable clear comparisons of many months. It was easier to identify and compare boroughs with more filming. Here, I attempted to investigate the color to correspond with an aggregate summary of that particular borough. I also decided to position the bars vertically, with the borough with the most permits aligned at the bottom. There are now more apparent approaches to understanding the information within the visualization.
UX Study findings
During the class discussion, I received a lot of questions and suggestions and found them very valuable. In order to ensure that my visualizations were usable and easy to understand, I did some user testing. When I planned for user experience research, I wanted to see if users were able to understand what the visualizations were communicating. This helps me better understand if the user thinks anything is unclear or confusing. The dashboard that I used for my test sessions is the static image provided below.
The study sought to determine how effectively visualizations engage viewers. The goal was also to put the visualization’s language, visual aesthetics, and content to the test. Participants for this visualization came from a variety of backgrounds, which helped me comprehend the visualization from several angles.
It was challenging for users to make an exact comparison between the years. They believed they needed to count each bar to approximate where certain years are on the graph. They wanted more information on how to understand the data would be and said that it would be helpful and more informative. The users suggested adding labels on the color scales. There was some initial confusion over the relationship between the size of the block on the bar graph. They didn’t comprehend the color tones used to signify the specific boroughs.
Based on meaningful comments from the UX study, I sought to try and spread out the seasons in the visualizations more. I attempted to build a flow for how the information would be read. I also adjusted the color scale to line with the Y-axis, and the hierarchy was directly tied to the way the bars are organized on the graph; I believed this would be very helpful and efficient in understanding how color intensity is related. Following that modification, I relocated the material from the header explaining the color intensity under the scale. The area highlighted the most permits, which was also color-coded according to the name of the borough. The completed dashboard is shown below.
Final custom visualization
Overall, the visualization was able to tell me and the users a story about the link between seasons, shooting permissions granted throughout time, and borough. During Fall, Manhattan had the highest rate of permits issued. When comparing seasons, Fall had the most permits out of all of them. Because I am still relatively new to this nation, I was somewhat surprised to learn that Winter had the least filming, despite the fact that there are numerous Christmas movies focusing primarily on New York. However, having lived through that weather firsthand, I realize that it is not favorable for film production. I wish the data for the filming street locations were more data-oriented so that retrieving the geocoordinates was easier and more dependable.
The viewer notices the custom visualization first when they watch this visualization. The idea was to capture the viewer’s attention and retain it long enough for them to read the entire visualization. As soon as the user sees the personalized representation, they get more interested in learning more about it. My UX research findings were incredibly helpful in designing my visualizations. As I worked on this visualization, the theme evolved and grew along the way. It also helped me realize how I can efficiently show different types of information. In this manner, I experimented with many options for differentiating the data by season until I settled on the final one. The final visualization experiment uses customized visuals. The plan was to turn this data visualization into a poster. Something that viewers may keep as a poster.
There was a lot of learning for me in this final visualization. I was able to create a unique visualization without using any data vis software. I now appreciate and understand how I can use software to interpret data by using various filters and settings, but it is achievable to make the visualization more user appealing by leveraging the results to design in different software. This final assignment was an essential skill for me to evaluate what information may be important for the viewers and how to avoid information overload. At times, I felt like I had explored a lot of colors and was becoming overwhelmed with determining the best strategy for properly expressing the information. Now I can say that I absolutely loved experimenting with different color combinations and believe that this is the foundation of the design that was developed.
It was necessary to take a step back to evaluate the visualization because when I become too familiar with the data, certain essential aspects may be missing without my awareness. While the data visualization is made publicly available, in order to successfully communicate the message, we should design with the purpose of allowing someone with no prior understanding of the issue to interpret the data. As a result, I saw the value of user experience research in providing important suggestions for improved design. For example, based on user testing, I discovered that not only are color, size, font, layout, and so on vital but how we use words to describe the material is also significant in influencing the audience’s understanding of the message we provide. Overall, I was able to answer the questions I raised at the beginning of the project with the available sources.