Who are we reaching, and how do we reach further and have greater impact? This was what the Royal Institution wanted to know about their new ExpeRimental project: a series of free online films that aimed to encourage and support parents (and also teachers) to do science activities with their children.
The first ExpeRimental film published
They commissioned me to put together a report that would do this, evaluating the audience and their reaction to the films, but also looking at what they could do to better reach and engage with an audience that they strongly suspected was not watching the films: those were less confident or knowledgeable about science. This followed an initial piece of research I carried out for the Ri that evaluated the pilot (which was used to help develop the full series of films, a very useful process in a project of this nature).
In this post I’ll share the process and some of the key findings, which I think could be very relevant to anyone embarking on a science communication project, or perhaps even any subject-specific digital engagement work. Spoilers: the non-science savvy are probably not likely to seek out your online content, no matter how good it is, and to reach them you are going to have to work a bit harder. If you do only one thing, it should be to try and reach parents via their children’s schools.
The first task was to look at the current audience, and try to find out more about who they are, and what they think of the films. To do this, I looked at analytics from YouTube and the Ri’s website, and put together an audience survey that asked for opinions of the films, as well as questions about their expertise and confidence in discussing science subjects.
One thing to note about the research: there were over 20 films in the series, but most had already gone live. The survey was linked to from most of the films, but the vast majority of results will come from the last three that went live (during Science Week) after the survey was published, skewing the results to those, and to people who came to them and the survey via subscriptions and social media (ie probably those who were already “fans” in some way of the Ri and who wanted to be helpful). Whilst we had over a hundred responses this does mean the results can only be indicative rather than a fully significant and complete picture of the audience. The lesson there is to capture viewer feedback from the very beginning.
That said, some of the results were pretty striking. First, the good news. The audience was overwhelmingly positive about the films themselves and fulsome in their praise for them. When asked what could be improved, the most common answer was “nothing”, closely followed by “have more”. A few wanted to see more “further reading” links, such as related videos or ways to extend the activities, but for the rest there were only a few minor criticisms or suggestions.
The audience loved that the presenters and settings were relatable (they showed “real people” and “real family homes”), and that they depicted kids actually doing the activities. They said that the videos were clear, and the activities were easy to do.
A pleasing number had also picked up on the fact that the activities were less about imparting factual information but were instead aiming to depict a more questioning approach, that it was asking the questions that was the most important thing. Perhaps the most encouraging stat was that 60% of parents had already done the activity with their kids or students, and of the rest, only one person said that they weren’t planning on it.
However, the less good news was that the audience appeared to be very much from within a particular science “bubble” – they were very science savvy. Not a single respondent said they were “not at all confident” discussing science with their kids or students, over 70% said they were “very confident” or “quite confident” doing so (compare this with the results from the Phase 2 recruitment survey where only a little over 30% from a more general audience were “very confident” or “quite confident” and about 20% were “not at all confident”).
Respondents were asked about their level of science studied, and over 55% had a degree or post-graduate qualification in science (again the phase 2 recruitment survey would show that less than 20% had a degree in science, over 50% not going beyond GCSE science). So, as suspected, there were clear indications that the current/existing ExpeRimental audience was somewhat unusual, and that there could be a huge untapped audience of less science-confident parents out there. Which leads us to the next phase of research.
The next question was clear: how does the Ri reach this audience? Their promotional plan so far had included use of social media, distribution via third parties such as AOL, national media coverage including TV appearances such as a slot on Channel 4’s Sunday Brunch and a piece on ITV London news, a Huffington Post blog series, hospital newsletters, promotion via Brownie groups and parenting blogs, and collaborations with organisations such as a link to British Science Week via the British Science Association and training sessions for play-workers via London Play. Whilst this had resulted in some strong numbers for a few of the videos, it apparently was still preaching to the converted, for the most part.
Looking at the analytics, discovery type varied between videos, some on well-known subjects (making playdoh, or bubbles) were being found by search on YouTube and others were mostly being found via YouTube recommendations or subscriptions. One or two had seen a lot of traffic from online articles such as on the Huffington Post and BuzzFeed. On the website people were arriving via various searches along the lines of “fun activities for kids” or “science for preschoolers” (most science related queries appearing to be school related, including “ks2” or “kindergarten”).
Being found via search was clearly very important to the success of some of the videos, but this only works if it is something that people might actually be searching for (almost nobody is going to be searching for “balancing structures”, the subject of one of the videos, for example). If it isn’t on a popular subject, then it is competing with all the other results for “fun activities for kids” that might come up when a parent gets desperate on a rainy bank holiday. Which is an awful lot of competition.
Social media is obviously likely to result in getting a message out to the like-minded (existing Facebook or Twitter fans and followers, and their similar connections). You’d expect that more general press and especially an appearance on Sunday Brunch might reach a more general audience, but it hadn’t translated into lots of views. A new strategy was needed.
In my experience, getting to the bottom of questions like this requires a more in-depth, qualitative approach. Surveys will only get you so far, and can be a bit self-selecting. I needed to find people in this target audience, and have a long chat with them. To do this, I posted a recruitment survey to Mumsnet and the East Dulwich Forum (a very active local forum), that was looking for parents who were not at all confident discussion science subjects with their kids.
I would note at this point that the two forums I posted the survey on are known for having a fairly middle-class demographic, and they are naturally going to be an audience that are very active and engaged internet users. We had discussed reaching an even broader audience with the research, but it would have taken quite a lot more resources and we decided that it was beyond the scope and means of this evaluation, at least for the moment. Also, the Mumsnet/East Dulwich Forum audience is likely to be a good fit for the videos, so it made sense to focus on them. As it turns out, I think the strategy suggested by this research will also help reach a wider audience anyway.
I found eight parents who fit the criteria (“not at all confident” and mostly educated to GCSE science level or below) and who were willing to be interviewed, in return for a £15 Amazon voucher. The discussions I had with them were extremely informative, and whilst some of the insight may seem obvious in hindsight, it certainly didn’t beforehand. It’s also useful to just have some suspicions confirmed, and evidence to bolster the case for a particular strategy.
Firstly, when these parents did watch the videos (after our interview), they really liked them. They all planned to do the activities with their children afterwards. So there was nothing about the actual content of the video that was putting off this less science-savvy audience. That was a good start.
However, most of our interviewees just would never have considered doing science experiments with their children at home. If it had ever crossed their mind, they wouldn’t have known where to start looking, and would be nervous about “getting it wrong” and misleading their children. Instead, they tended to focus on doing more arts and crafts activities, general play, or sports and outdoorsy pursuits, because this was what they were comfortable and familiar with. It was clear that this audience were never going to find the videos by search, or by seeing them shared on social media, it was going to take something more direct, and something that addressed their concerns.
When I asked what they thought would be a good way to reach them, every single parent said something along the lines of “get a note into the kid’s school book bags”. The schools were the key, they said, they respected information that came via that route, and they paid attention to it. Whilst a mailshot to every school with leaflets for every child would require a pretty huge marketing effort and budget, the basic message of “go via the schools” suggests other possible avenues (via school newsletters, targeting teachers as ambassadors, or encouraging schools to send out emails or texts, perhaps).
Other possibilities mentioned included holding community events or targeting community centres, trying to reach after-school science clubs, tapping into childminder networks on Facebook and elsewhere, getting mentioned on general parenting blogs and forums, and via museums (the Science Museum and Natural History Museum were mentioned often as popular destinations for family visits). Nobody felt that Facebook or Twitter would be particularly good ways to reach them directly as parents.
The other important aspect of this is that if you did manage to reach this audience with a promotional message, it had to feature some key information. It had to be clear that the activity would be easy to do, use things that you would already have at home, or be straightforward and cheap to buy, that it wouldn’t take a lot of time to prepare (and then take only 5 minutes to do), and that they were be supported in answer the kid’s questions. One of my recommendations from this was to test future marketing messages directly with this audience, to make sure it does all of these things.
The parents particularly wanted to know the context of the science activity. What subject did it fit into, especially in terms of the school curriculum? What was it related to, and why was it relevant to their lives (i.e. is this really about electricity, or sound waves)? They also wanted to encourage their kids to spend time outdoors, so activities that could be done in the park or garden would be seen as a good thing.
What I love about doing research of this sort is that moment where someone says something that suddenly seems so obvious, but that somehow hadn’t been before. It’s usually a good sign that your findings are plausible. So, it now seems very obvious that you have to be pretty confident and aware on science subjects to be seeking out experiments for your kids to do at home. It now also seems very obvious that schools are the best route to the audience that isn’t. Somehow, it wasn’t obvious at all when I started on this, but hopefully it will be useful for future ExpeRimental films and perhaps other science communication projects too.
Click here to browse the full set of ExpeRimental films.