UFO Sightings

Report

Introduction

The project motivation was rooted in the curiosity by the team members to explore UFO sightings and excitement to learn more about potential patterns, frequency, and significant sighting locations. The data was sourced from the National UFO Reporting Center, which contained information such as city, location, date, and more about UFO’s. 

Our main objective was to promote scientific literacy and critical analysis amongst the general public by providing tools that allow them to explore unexplained sightings for themselves. A lot of people may be curious about UFOs, but may not know where to start. Since most of our team were international students, we all held the sentiment that space exploration was a much bigger deal in the United States, and the concept of UFOs appeared to be bigger in Western media. We wanted to explore ways to make this concept readily available to everyone.  Our project aims to make information about UFOs  accessible to anyone, whether they are a UFO fan or not, and empowers them by giving them the opportunity to track these sightings themselves. 

The main conclusions we found were that the top cities with the most UFO sightings were all from the United States, which was aligned with our initial hypothesis. We also found through the creation of our app and user testing that people engaged with the information more when they could relate to it, ie. when they could see UFO sightings close to them and explore themselves.

Justification of approach

Our main deliverable was an interactive shiny app which acts as an educational tool to create awareness about UFOs and allows users to explore UFO sightings. The intended audience can be divided into two groups: 

  1. People who are interested in UFOs and want to know more about them 

  2. People who do not have previous knowledge about UFOs but are interested in learning more

One of the key concepts of our approach was to make the experience as engaging and interesting as possible for all intended audiences. We wanted to stray away from more scientific approaches such as reports or posters, which is why we chose an interactive web app. We believe this was the best approach for our audience because it is very low stakes, so removes the daunting pressure of learning something new, and also has a fun element as it allows users to explore and play around on their own.

Data description

What are the observations (rows) and the attributes (columns)?

We have 111,582 rows and 17 variables.

Each row represents one specific UFO sighting, along with other characteristics about the sighting, such as the date, time, duration, summary, location, etc. The columns we will be focusing on are:

year: The year that the specific UFO sighting occurred.

month: The month in which the specific UFO sighting occurred.

day: The day on which the specific UFO sighting occurred. 

hour: The hour of the day at which the specific UFO sighting occurred.

minute: The minute of the day at which the specific UFO sighting occurred.

city: The US city in which the specific UFO sighting occurred.

state: The US state in which the specific UFO sighting occurred.

shape: The shape that the UFO appeared to be in to the person it was seen by.

duration: The duration of time that the UFO was seen for. 

summary: The summary of the UFO sighting provided by the person who saw it. 

event_url: A link to the posting about the UFO sighting.

images: An indication of whether the specific UFO sighting has an associated or not. The values are either ‘yes’, ‘no’ or NA.

img_link: A link to the image associated with the specific UFO sighting. 

lat: The exact latitude of where the UFO sighting occurred.

lng: The exact longitude of where the UFO sighting occurred. 

population: The population of the city in which the UFO sighting occurred.

date: The full date of the UFO sighting.

Why was this dataset created?

The National UFO Reporting Center established this dataset to collect, verify, and archive accounts of potential UFO sightings from diverse sources. The creation of this dataset promotes transparency by making UFO-related data publicly accessible and encourages further research and understanding of UFO phenomena.

Who funded the creation of the dataset?

The National UFO Reporting Center was founded in 1974 by Robert J. Gribble, a UFO investigator. 

What processes might have influenced what data was observed and recorded and what was not?

The main means to report sightings since 1994 has been the website. However, before reports were made through phone calls and mail. The type of reporting mechanisms available, and who has access to them, could have influenced who was able to report the sighting and what details they could include. Furthermore, the extent of public awareness and belief in UFO sightings could have played a role too; periods of higher public interest or notable events might have led to more reporting, or certain regions where people are more likely to believe in UFOs could have reported more sightings. Lastly, because the Center ensures anonymity of the person reporting, it is possible that more people felt comfortable reporting. 

What preprocessing was done, and how did the data come to be in the form that you are using?

We got both our datasets from data world (linked below under sources), and they got the data from the National UFO Reporting Center. During the preprocessing, the authors added geocoding data such as latitude, longitude and population of city to the ufo2 dataset. The time variable was also broken into hours and minutes for the ufo1 dataset.

If people are involved, were they aware of the data collection and if so, what purpose did they expect the data to be used for?

Given that the primary data comes from individuals who reported their sightings to the National UFO Reporting Center, we can assume that these individuals were aware that their reports were being collected as data. The center also has a policy of guaranteed anonymity, which suggests that informants were aware of their part in the data collection process. The Center also seems to be very open about their policy surrounding collected data, so it is likely that individuals expected their reports to be used in research and public knowledge. 

Data Limitations

  • All of the data we have is self-reported, which means that there are several inconsistencies within the data, especially in text columns such as duration and summary. This means that it is difficult to standardize the data in these columns.

  • Another potential problem arises from joining the two datasets that we have. Some columns that we will be using, such as population, are only present in one dataset, which led to a lot of NA values after we performed our join.

  • One of the major limitations was the geographical data associated with the dataset. Due to reportings being digital, reports that occurred more than 7 years ago did not have specific longitudinal and latitudinal data, rather only provided city names. 

  • The main topic surrounding the dataset is UFOs, which can be seen by many as superstitious. Only people who believe in UFOs would have reported to the website, even if others had seen similar sightings. This means that we have an unrepresentative and biased dataset.

Due to the time constraints we faced during the semester, we cut down our dataset to only contain the rows that had the information we needed, and also mainly focused on locational data. However, if we had more time, we would have liked to explore other variables such as shape or weather.

Design process

Our design process for the web app followed a user-centric approach:

  1. We started by identifying our core user needs and defining the core functionalities of our web app. This included making sure the website following an intuitive flow, making sure we had state-specific search queries, interactive maps and a user-friendly interface for reporting sightings.
  2. Before we began developing our app, we spent a few hours creating wireframes and sketching out what a potential user interface would look like. This allowed us to visualize the app’s flow and layout.
  3. Once the design was solidified, we moved onto development—translating our designs into a functioning application.
  4. We did some user testing with a few friends to refine our app and make sure it was easy to use. We got some feedback on adding more instructional sentences to guide the users through using the app.

Some challenges we faced were:

  1. Deciding whether we should allow users to input data through a search bar or present them with dropdowns that they can select. Since users can make mistakes when spelling out their states, our search query may not be able to recognize the input. On the other hand, dropdowns can be frustrating for the user as they have to scroll through to look for their state. We ended up choosing the dropdown just because we could ensure correctness and efficiency.
  2. We tried to cut down our findings from the exploratory phase to figure out which ones would be most useful for users to know. We thought the location would be most relatable, and this was confirmed after a short round of user testing we did.

Limitations

While we acknowledge the universal curiosity surrounding UFO sightings worldwide, the app’s search functionality is presently confined to United States, indicating a geographical limitation. The app also restricts users from utilizing zip codes to narrow down their search for UFO sightings.

Acknowledgments

National UFO Reporting Center. NUFORC. (2023, October 28). https://nuforc.org/

UFO data NUFORC - dataset by CK30. data.world. (2021, August 23). https://data.world/ck30/ufo-data-nuforc

National UFO Reporting Center Reports - dataset by khturner. data.world. (2017, April 25). https://data.world/khturner/national-ufo-reporting-center-reports

https://www.secretsdeclassified.af.mil/Top-Flight-Documents/Unidentified-Flying-Objects/