Interview with Andrew Durso from the Snake ID Challenge
“Do you know this snake?” That was the question we asked a vivid community of snake lovers out there in our Snake ID Challenge in late February. We were overwhelmed by the level of participation from all over the world.
Venomous snakebite is responsible for over 100,000 deaths and many permanent disabilities every year, mostly affecting poor and rural communities in the parts of the world where snake diversity is the highest. Communities and healthcare providers working in these areas are often limited in their herpetological expertise.
The ultimate goal of the snake ID challenge was to create digital tools that anyone can use to identify snakes in order to help clinicians better treat snakebite cases, and improve snake conservation through educating people and communities. Read more about the background of the challenge here.
Andrew is a herpetologist and postdoctoral researcher at the Institute of Global Health at the University of Geneva and a member of the team behind the Snake ID challenge. He was born in New York and grew up catching snakes in North Carolina. He also writes a blog about snakes called ‘Life is Short, but Snakes are Long‘.
Hi Andrew, how many people took part in the challenge and where were they from?
Just over 1,000 people took part in the challenge, from 48 countries. In total they contributed over 115,000 image identifications.
The goal of that first challenge was to measure the performance of the crowd. So how well did they perform?
Overall, 42% of the answers were correct at the species level, a further 11% at the genus level, and yet a further 19% at the family level. About 13% of the answers were incorrect at the family level. Finally, 14% answered “I don’t know”.
What feedback did you get?
We got very positive feedback! Most users found the challenge fun. Many users were proud of how well they did, especially at identifying snakes from regions with which they were unfamiliar, and many people said that participating in this challenge was a great learning opportunity for them.
What happens with the results now?
We’ll analyze the results in greater detail soon, including a breakdown by species, geographic region, and higher taxonomic group (e.g. vipers, elapids). Preliminary analysis suggests that the most important factor influencing how correct people were was the species of snake—some groups are simply more diverse and difficult to identify than others. This seemed to have a far greater effect than global region or the quality of the image.
What are the next steps?
We’ll do a second challenge where we ask people to suggest IDs with and without geographic information, to test more explicitly how much knowing the location of a snake influences someone’s ability to correctly identify it. We also want to move forward soon with asking participants to help us tag images that we don’t already have IDs for, to grow the overall size of our dataset & formalize a process for identifying snakes from images
Were there some funny or special interactions happening?
Several people told me that they worked on this challenge while they were at work or while they were supposed to be doing other things, suggesting that we did a good job making it fun! Some found it quite addictive, staying up late or neglecting other important tasks to complete it (we did not explicitly encourage this behavior)! A few people recognized their own images in the challenge & one person said that if they had not identified that images correctly, his privileges as a group admin for a prominent snake ID group on Facebook should be revoked!
The challenge was developed and implemented with the Citizen Science Center Zürich. How did you find the collaboration with the Center and what was the most useful support in your opinion?
Working with the Center was fantastic! Everyone at the Center was very helpful & enthusiastic about the challenge. In my experience most people don’t get particularly excited about snakes, at least not in a positive way, but the Citizen Science Center Zürich fully took on this challenge. Their technical support was rapid & accommodating and their help with community management was invaluable.
Thank you very much for your positive feedback and the interview! We also enjoyed collaborating with you a lot and are very much looking forward to the next steps together in this unique and important Citizen Science project.
Visit snakes.citizenscience.ch for more information.
One thought on “Citizen Science for Global Health”