A Citizen Science Training Day

At the end of March, the Into the Night early career researchers – Adele Powell, Emma Garnett, Pen-Yuan Hsing, Valentine Seymour, Gianfranco Gliozzo and myself – sat down together to admire what we’d achieved over the last three months. We all felt a bit breathless and surprised, as we’d undertaken five separate events – the first of which I was supposed to have blogged a long time ago, the Citizen Science Training Day on 2nd February in Oxford. Sorry for the delay – to say the least, we were quite busy!

Project proposals compete with each other and thus promise lots of project outputs, such as training or public engagement. So we invited up to 60 early career researchers – especially PhD students in ecology or environmental science – to come and spend a day hearing talks and doing interactive exercises learning how to incorporate citizen science into their work.

There are big tasks that go with this, such as inviting speakers and finding a venue. Then there are all the little tasks: the signs you want printed, the phones to bring to act as extra wi-fi spots, the leaflets to write, the catering and finding out everyone’s allergies to arrange.

We were very lucky with the venue: we got Oxford Town Hall, a vast room full of old paintings, grand round tables, plenty of space and an excellent projector – but not, as we later discovered, such good sound quality. Note for future event planners: see if they’ll let you visit and test things out in advance! We also got an impressive line-up of speakers: Muki, of course; Michael Pocock, ecologist at the Centre for Hydrology and Ecology who leads several citizen science projects and is full of interesting stories; Ian Thornhill, Adele’s colleague from Earthwatch; and Grant Miller from the Zooniverse. Grant, I found out later, was a lucky coincidence: just a couple of weeks previously I’d brought Adele, Pen and Valentine to a “zoominar” I’d given at Chris Lintott‘s invitation, talking about the early days of Galaxy Zoo. Grant had been there and had shown the others around and how to use Zooniverse Project Builder – for if you ever want to create your very own citizen science project. Oh yes – and me.

Valentine and I wrote an invitation which we put out on Eventbrite, and we began publicising it to universities where we had connections and also generally on social media, for these things travel fast. “Have you ever wondered how citizen science projects are created? Do you want to learn how to engage the UK public with issues of environmental concern?” we asked. “The training will take participants through techniques and goals of developing a citizen science project, including project design, recruitment and retention of volunteers, reporting, monitoring and evaluation, costing as well as issues to consider. The context for these sessions will be related to environment, ecology, light pollution and human wellbeing. There will also be dedicated time for Q&A and networking.” We also took along leaflets about DITOs, a video camera to interview people afterwards, and evaluation forms. We encountered our first problem here: as it was a DITOS event, we had to bring the DITOS evaluation form – these cannot be altered, or it would compromise DITOS’s evaluation process – but this did not ask all the questions we wanted to ask. But, as one of us pointed out, “it’s hard enough getting people to fill out one evaluation form, let alone two!” However, we had no choice but to make the two as different and interesting as possible, and simply explain the situation to the participants. (Being mostly PhD students, they were bound to be sympathetic. They generally were – we had some very thorough feedback.)

The day also brought us the companionship and silliness of studenthood: checking into an AirB&B, assuring each other that it would be OK to wake anyone up who snored, hauling a ton of bizarre equipment such as a giant tripod up five floors of stairs and upon discovering that it had a fantastic kitchen and we all liked aubergines, going to four different shops looking for aubergines in vain! (We had pasta with pesto and various vegetables in the end.) Adele caught us up on the morning of 2nd February, traipsing through the wet streets of Oxford with laptops and tripods on our backs.

Because we wanted our participants to be able to share information with each other and of course with everyone else (if this doesn’t happen, what’s the point of science, especially citizen science?), I prepared this slide for everyone to see as they came in. The hashtag is still full of interesting tweets and feedback:

We then quickly laid out a table of books, leaflets and registration forms – welcoming people in, inviting them to help themselves to a coffee, asking if they were OK with being photographed for blog posts like this. Only one person wasn’t, and I don’t think we photographed you but if you see yourself let me know and I’ll take that picture out! Things got off to a good start:

(It’s a good book. I recommend it.)

Muki’s talk then began with an introduction to citizen science, and you can see all the slides on his SlideShare:

Muki began with describing the beginnings of the Into the Night project: a history of Earthwatch and UCL ExCiteS working and generating ideas together, and more people meeting to germinate ideas; then, a narrowing down of ideas as the initial plan to study both light and noise pollution was dropped, and to focus just on excessive artificial light. Because such issues affect so many people, and because many different types of organisation have various types of expertise, it was cocreated by several. He then went on to describe many different approaches, disciplines and technologies: for example, does a scientist think up a project and ask for help, or do the volunteers have a large part to play in the project design, too? Or do non-scientists work completely independently? Is the project to solve a scientific or social problem? Does it have other beneficial spin-offs, such as increasing STEM education?

The next talk was by Ian Thornhill, and we had an excellent time using Kahoot to almost conduct the whole talk as a conversation with Ian through voting. For instance, why were we interested in citizen science? – which itself produced a wide variety of results (doubtless a good thing). His work has been on using citizen science to assess the visible water quality, one of the Sustainable Development Goals; but, as we saw, people will see the same river or pond and give different answers:

When we ask citizens to monitor their environment and they take the trouble to examine a place and submit data, they will want to know immediately: has their result gone through, was it correct, what does it mean? In the longer term, they will wonder: is anyone listening, and what can be done to address problems? It is vital to have answers available to these questions; the first can be done with technology, while the longer-term will require action and the writing of open access reports. “Ask not what the citizen can do for you, but what you can do for them,” Ian told us. He also set us a problem that occurs in real life: what do we do with unexpected or worrying results – especially before speaking to an environmental regulator?

Grant Miller took us through the story of the Zooniverse, including how it made media headlines – and also the importance of community:

A community can talk amongst themselves, ask their own questions, teach and support each other and spur itself onto greater achievement than if all citizen scientists act alone. Also, they can compare their results – for instance, “Yes, I’ve been seeing galaxies like that too, but I wasn’t sure if they were not just a glitch/if I was going mad.” Multiple pairs of eyes (whether working as a community or not) will be able to filter out what is common or uncommon, average out results to reduce the impact of mistakes, and to spot things that researchers hadn’t even thought of asking: “unknown unknowns”.

We went from galaxies to penguins – “We all knew we were going to get to penguins at some point” – and learned a surprising result from projects involving animal camera traps – which of course do not trap animals but photograph anything that moves. Now, this can include things such as blowing grass, which initially projects such as Snapshot Serengeti filtered out, for the Zooniverse has a policy never to waste anybody’s time. But, to their surprise, this actually decreased engagement. Not having an animal in every picture made each picture that did contain an animal more special.

After this, everyone had a go at building a zoo – but unfortunately, despite our best efforts, the wifi made this a painfully slow process.

After lunch, Muki was back to the stage with technology for citizen science:

If we use technology in citizen science, such as sensors, do we want citizens to use their own computers and smartphones, or provide them with something more specialist? There are pros and cons of each. You can also use existing social media platforms, for example submitting photos. We went through a lot of different technologies – the important thing is to keep monitoring to see if it’s doing what you need, and make sure everything remains open source so that anyone can access and use it.

My presentation was next: a story of working with citizen scientists. Although it was an ecology-based day, my story came from Galaxy Zoo and one or two people’s evaluation forms showed dissatisfaction with this. Nevertheless, hopefully they got a useful perspective from a volunteer this time: someone running the community rather than the science, who spent many years finding out what people need and how to provide it. One thing that has come to annoy me at conferences is the endless focus on citizen scientists’ motivations. To me this seems rather limited – treating them as donors instead of colleagues. Anyway, to me, a citizen scientist needs three things: firstly, a task to do (you’d be surprised how many projects are extremely vague about this, with “engagement” and “outreach” as their main goal); secondly, some data to find or play with; thirdly, a place to talk. This last was the main focus of my talk – “the community” in this case being Galaxy Zoo’s first discussion forum.

I spoke about difficulties volunteers have had, not just at Galaxy Zoo – for example, the interface being hard to see or fiddly; or demands being incompatible with people’s ability – constraints on time, money, dexterity, language … This is especially important when considering, for example, a project on age-related diseases whose participants might be elderly and experiencing the above problems more than researchers. Therefore, we need to beta-test not just with colleagues, but with the public:

Often, I explained, the community itself will solve problems. People will volunteer to translate for each other (I enjoyed, for example, putting Spanish speakers in touch with each other) and answer each other’s problems and questions. To build up a supportive community is hard work, but it will be of incredible value!

My interactive task was three “case studies” of various problems that might occur within the community, or between the community and researchers. (I learned that it’s far better to present a variety of small case studies than a few large ones – and tried this out at the next event, which worked much better!) A few people seemed rather bewildered and intimidated at how complex it can be to take care of all these issues. But hopefully, they were left with lots of ideas too.

We did not just learn about the speakers’ projects – some of the audience had their own, too:

The final talk was Michael Pocock’s: Citizen Science Project Design. Our participants felt that this talk was especially useful:

First, Michael debunked the idea that citizen science lies somewhere between “real science” and “engagement”; the two are not mutually exclusive but common goals which make the project possible and valuable in different ways. We were then introduced to Conker Tree Science, which you can read more about here; and an interesting snippet was that Michael found about short and long term motivations (them again: the difference between recruitment and retention!). Conker Tree Science’s website screams: “Our conker trees are under attack by ‘alien’ invaders!” Although it defines “alien” as “a species not naturally found in an area or habitat”, this appeals to people’s sense of threat. While it provided short-term engagement, it did not keep people there. What provides longer-term engagement was a shared enjoyment of nature.

We learned of the huge variety of what “citizen science” can mean, the sizes of projects, the different measures of success – and when it might and might not be a good idea to recruit volunteers (for example, something short-term with little available data, with a complex protocol, few resources and a vague reason for participation is not suitable). Like the Zooniverse, Michael found that the primary aim of volunteers is to contribute to science. Clear, precise questions need to be asked – and keep figuring out where you can make them better and simpler. He asked how many of us had tried other citizen science projects:

You can watch a bit of Michael’s talk here:

Indeed, one or two evaluation forms remarked that we should have had Michael’s talk first because it was the first one to (in their opinion) state exactly what citizen science is. (The unlucky sound quality may have obscured Muki giving the explanation they needed that morning.) This taught us two things: firstly, that signing up to something doesn’t mean someone knows what it means; secondly, that more planning time would have really helped! We only just got everybody’s slides in time – we had had only about three weeks to prepare from scratch. I suggested that for future events we could draw up a list of what points need covering and in what order, and ask speakers to tick off any that they will mention and put people’s talks in order based on this, which we nicknamed a “wedding registry”. If anyone else has tried this approach for a training day, I’d love to hear how it went.

Overall, people’s least favourite things were the sound, the fact that it was a long tiring day and the slow wi-fi. A few also suggested that they should have been asked to mingle more – people remained at the same chairs all day. However, we had a lot of positive feedback too – several people came up to me to say, “Thank you, that was such a useful talk, I hadn’t thought of any of that”. A few participants put, for the question “What was the most useful part of the day?”, “All of it”. Several said that the number of real-life examples and interactive exercises had been the best, and that they had been introduced to many new platforms. The different perspectives – different organisations, researcher versus volunteer – were also reported as very useful. If you were at that event and have more comments, please add them here. I really hope to organise another training day in the future.