A Citizen Science Research Platform - Transforming A Concept Into Something Practical?

127 views
Skip to first unread message

Steve Whitbread

unread,
Feb 12, 2014, 9:23:01 PM2/12/14
to nfbr-...@googlegroups.com



Summary: It is suggested that developing a suitable Toolbox or Platform to provide a collection of customisable toolkits/resource sets would offer the means of simplifying the setup, coordination and monitoring of Citizen Science projects, whilst embedding consistency, credibility & satisfaction for all involved.

Developing, testing and demonstrating particular tools via a set of exemplar projects - addressing a range of key themes or issues - would enable a possible consortium approach to developing such a platform for Citizen Science in the UK and elsewhere.

In relation to the deliberately bare bones outline below:

  • Is this concept worth developing?
  • What might the exemplar projects be?
  • Who might be the partners in any consortium? 
  • What would be the best way to get the idea across to NERC and other funders?
  • How might we encourage more people to comment on the idea and the resulting suggestions for what could be turned into the first crowd-sourced NERC funding proposal?

…................

There are many examples of excellent Citizen Science projects of local-international scale. These cover a range of observational/experimental studies, from long-running amateur naturalist schemes to major, institution-led collaborations, e.g. the OPAL Project.

Big Society, Big Data, current issues and modern technology each give emphasis to the value of and the need for Citizen Science - and for mechanisms to enable, support and get the greatest returns from projects, including for the participants, as well as ancillary benefits, e.g. in education, biodiversity conservation.

Look in detail at past projects and what Citizen Science might address in future, and you could quickly identify a number of common & recurring elements, e.g. tasks, audiences, awareness raising, engagement, capacity building, skills testing, training, planning, coordination, sampling, monitoring, analysis, reporting, learning, sharing etc. Characterising the tools/procedures required to address these elements would highlight the suite of tools most likely to be needed for future projects of particular types.

Providing customisable toolkits to enable the right tools to be made quickly available to meet the various needs of a particular project would seem a useful approach, especially if those tools conformed to required standards. An excellent existing example is the Indicia toolkit (developed as part of OPAL) which, complies with the National Biodiversity Network data exchange standard and enables rapid construction of highly tailored recording websites for species and/or environmental data. It has already been used in a wide range of recording initiatives as well as the development of iRecord.

In contrast, the BiodivERsA stakeholder consultation toolkit is intended to address a more general administrative element of project management/delivery.  I’m sure you can suggest lots of toolkit examples of which I’m unaware. However, it is unlikely that there are toolkits to cover all the elements identified above, not least in the coordination of projects and people, and how many these actually work well together?

The suggestion is that the Citizen Science Community and future projects would each benefit from the existence of a platform or toolbox of different, interoperable tools and toolkits which (where relevant) adhere to standards that projects will inherit, because:

  1. It would make it quicker, easier and cheaper to initiate and deliver projects;
  2. Standards-based tools would introduce the necessary degree of rigour to projects – instilling confidence in their outputs for all concerned;
    1. Helping to guide citizen-led CS projects, and
    2. For scientist-led schemes - provide academics with confidence that a CS project/component would be more likely to succeed (and be green-lighted by a funding body), as well as
    3. Adding credibility to the resulting evidence.
  3. It would give funding bodies of all sorts the incentive to back CS projects that used the toolbox because
    1. There would be added guarantee of success;
    2. They would be cost effective (inputs to one project more likely to benefit other schemes)
    3. Each project would contribute to the ongoing development/extension and updating of the toolkit
  1. More could be achieved within the same level of resources;
  2. Wheel re-invention would be minimised.

Now, if you’re in broad agreement with 1-5, then it may be argued that the current lack of such a CS Platform is wasting resources and holding back the development of Citizen Science.

So the reason for posting this here is that, rather than putting a vague concept in NERC’s ideas box, it would seem much better for the CS community to

  • highlight what they think it could/should/must include
  • identify reasons for why the toolkit/box approach would be worth pursuing and suggestions/examples of how
  • catalogue existing tools/kits that could be incorporated/adapted or learned from and where the main gaps in  current provision are, and
  • go on to identify a suite of exemplar CS projects that - in process of addressing interesting and important issues in themselves – would help develop, test and refine the toolkit.

 

There would then be potential in submitting a proposal to NERC based on that suite of projects.

As the interest in last November’s BES SIG workshop shows, there are a wide range of organisations interested in or involved with different aspects of CS, as well as researchers keen to use Big Data. Citizen Science will become increasingly important in future and investing some resources and effort in facilitating future initiatives would appear to be strategically and financially sound.

All comments welcome


....


The concept is am interweaving of various CS suggestions resulting from the recent BES workshop on influencing NERC funding, chaired by Bill Sutherland.

….


Paula Lightfoot

unread,
Feb 14, 2014, 11:57:27 AM2/14/14
to nfbr-...@googlegroups.com
I've just posted this reply in the BES Citizen Science LinkedIn Group but I'm posting it here in case anyone is interested and is not a member of that group (but if you're interested in citizen science please do sign up!)

Thanks Steve – I think there’s a lot of potential in this idea and the BES Citizen Science SIG seems like the right group to refine it and take it forward. 
Some comments from me: 

• If this is put forward to NERC or other funding bodies, it should refer to the findings and recommendations of the CEH/UK-EOF report on citizen science. 
• I agree with Richard about the need for standards of data exchange, and this needs to be transparent and well understood so that volunteers know that their data isn’t just going into a ‘black hole’. I think the CEH platform that Richard refers to is iRecord and the underlying Indicia warehouse, in which case I agree that the Indicia toolkit is a very successful way of enabling data from (many!) apps and websites tailored to different audiences to be collected in one place in a standard format. 
• The NBN is already working on a way of ‘branding’ the standards and tools for the capture, management (including verification) and dissemination of biodiversity data and derived information products, to help users understand how things join up…especially those suffering from ‘app overload’! 
• This proposal is about more than just data technology, it’s about standards and tools for all aspects of citizen science projects, e.g. audience selection, engagement, impact evaluation, feedback etc, so I think that provided it refers to existing tools and standards where applicable, then it will complement and not duplicate. 
• I think it’s important to define the scope of this proposal – is it about citizen science projects related to biological recording (species and habitats) or the wider natural environment, or other branches of science too, including social science? There are successful citizen science projects in other areas, e.g. astronomy, so would this project seek to set standards for those areas, or rather learn from them to set standards in a more focussed (natural history) area? 
• The proposal mentions ‘added guarantee of success’ (point 3.a). If this is put to NERC it needs to recognise that there are different criteria for ‘success’ of a citizen science project, e.g. data to answer a scientific question; data to support a campaign; public engagement; education; impact on individuals (e.g. connectedness to nature); profile raising of an NGO (membership, donations, media coverage); profile raising of an academic institution (peer reviewed publication). Also ‘engagement’ could be targeted at a specific audience – a project might engage a large number of participants but still not be deemed a success if it failed to engage the particular demographic it was targeting. 
Hope that's helpful, I look forward to hearing what others think!

Jonathan Silvertown

unread,
Feb 14, 2014, 4:06:27 PM2/14/14
to nfbr-...@googlegroups.com
I think I agree with Paula that it is standards and not new tools that are needed. The tools are developing so fast and there are so many possibilities that what is needed is a way of finding and keeping up with these, not adding to them. I also don't like the top-down implications of the NERC proposal either. CS is a heterogeneous set of activities that cannot be directed and arguably needs no stimulus either. Its hyperactive already!

Paula Lightfoot

unread,
Feb 14, 2014, 4:22:52 PM2/14/14
to nfbr-...@googlegroups.com
I'm not sure I agree that citizen science 'cannot be directed'.  Citizen science projects are funded...by NERC, Defra, HLF, EU sources etc...if the funders accept the standards and set them as criteria when funding CS projects then surely citizen science CAN be directed?
The important thing is that the standards should be agreed by the community (hence the original post being made in these two groups - to encourage democracy not a top-down approach) and should not stifle creativity.  
Message has been deleted

Steve Whitbread

unread,
Feb 14, 2014, 7:00:17 PM2/14/14
to nfbr-...@googlegroups.com
I also agree standards - together with interoperability are vital - Jonathan. The proposal is in no way to simply introduce a new set of tools but to take what's working already (and fill in any gaps that are left) and make these as fully customisable and interoperable as possible to provide the suggested toolbox/platform [My fault for not being clear].

Essentially, you'd then have a Rapid Project Development platform that would help people achieve what they want without them having to develop tools or other resources from scratch each time (so avoiding the introduction of yet further new set of standards/formats that don't actually work with anything else - and benefiting no one). And far from suggesting a top down approach, the idea is that the CS Community (whatever that is) identifies what it needs, what it can use or adapt that already exists and thinks about how best to provide the missing pieces. That's one reason why the concept is so bare bones. And its also the reason for suggesting that the idea could be developed via a series of projects.

Hopefully you'll now allow that the concept isn't about direction, control or a need for stimulus but all to do with enabling, empowerment and needful consistency/comparability. So far as NERC and other funders are concerned it's partly about building out perceived risk so that proposals that are wholly or partly CS based won't be at a disadvantage. For those who want to undertake a praticular project, the intention would be to make it easier to achieve their objectives and to get the funding to do so - whilst ensuring the participants get what they wan t from their contributions as ell.

The thing is intended to let projects fly not to be a strait jacket. Each individual project would contribute to the projects evolution [More detail in reply to Paula tomorrow]. Looked at from a different direction, the UK-EOF undertook a review of 234 CS projects. Of those, how many contributed tools or collected real data about participant behaviour to inform other projects? Some, but not a very high proportion.

The hyperactivity is great. Let it make the most of itself.

Paula Lightfoot

unread,
Feb 15, 2014, 5:30:40 AM2/15/14
to nfbr-...@googlegroups.com

I do think there is some value in the idea of having a set of standards around citizen science projects (not just data standards) - what do others think?  Is such a thing already in development somewhere?

On the BES LinkedIn group someone referred to 'app overload' - the plethora of biological recording software, apps and portals is certainly confusing and it's not always clear if they do link up and share data.

However I'm much more confused by the plethora of organisations involved in citizen science!  There's the BES Citizen Science SIG, the European Citizen Science Association, UK EOF's Citizen Science Working Group, and initiatives with a more specific focus like BNHC's forum for sharing best practice on Bioblitzes.

How do these organisations interoperate and share information?

The UK-EOF guide (Tweddle et al 2012) is really helpful but how will it be kept updated, are there gaps in it, and is it just 'good advice' or could it be developed into a set of standards accepted by funders as criteria for funding - and if so could that be done without stifling innovation?  

Steve Whitbread

unread,
Feb 15, 2014, 2:27:57 PM2/15/14
to nfbr-...@googlegroups.com
I've responded to comments so far in the BES SIG group on linked in as follows:

Many thanks, Richard, Paula and Jonathan.

 [Reviewing my original post & comments so far, I could have been clearer about the tools/resources that might be included in any new toolbox/platform. The concept isn’t about simply introducing yet another set of tools(!) but by questioning - What do CS projects of different sorts need (now and in future)? What’s already available or could be adapted? What standards do these tools work to? What gaps in tool/resource provision are there? – to create something that would assist project development and delivery.

 The resulting answers ought to let us identify (1) a series of interoperable standards that (where relevant) any tools need to comply with (2) Existing tools that could be put straight into the toolbox (3) Tools that require some adaptation first (e.g. something created for a specific purpose that could be made customisable) (4) Tools that have yet to be created (but which may be made to work well with existing resources).

 I have not attempted to define categories of ‘tools’ or ‘resources’ so assume that they range from computer tools to sampling protocols or to style guides, rather than excluding anything at this stage.]

 Response to Paula's points

 • If this is put forward to NERC or other funding bodies, it should refer to the findings and recommendations of the CEH/UK-EOF report on citizen science.

 The UK-EOF Report and Guide would provide much of the foundation and framework for any platform that might be developed (YOU might like to suggest other info/advice). In part, the concept is about matching tools/guidance/protocols to the various tasks involved in the project phases as identified by the UK-EOF Guide. [Additionally, this would benefit community projects that are otherwise lacking in scientist input]

 

• I agree with Richard about the need for standards of data exchange, and this needs to be transparent and well understood so that volunteers know that their data isn’t just going into a ‘black hole’ [etc.]

• This proposal is about more than just data technology, it’s about standards and tools for all aspects of citizen science projects, e.g. audience selection, engagement, impact evaluation, feedback etc., so I think that provided it refers to existing tools and standards where applicable, then it will complement and not duplicate.

Absolutely, on both points. Consequently, it is about their interoperability too. [And, extending Richard’s point, if there’s an excellent tool already available for task X or something that could be quickly customised to let you achieve it, which you know others have used successfully wouldn’t you also like that tool available in a form you can use, knowing that it is standards compliant and got a really good reception from participants in previous projects? And if it doesn’t fit your needs it might well be a good starting point for developing something that does, which others might then use in future]. One obstacle is that probably nobody in the SIG or elsewhere knows about every one of those existing tools/standards, let alone how they fit together (or not) or where the main gaps lie.

Although I kept my original outline +/- as brief as possible, I included mention of Indicia because I’m not aware of any better example of a ‘design once, tailor many times’ tool – and of course it was designed to be (NBN) standards-based from the ground up. So I agree with you both!

Managing CS projects so that they match well with participant motivation, maintain (and preferably build) interest and don’t disappoint expectations is one of the elements crucial to project success – as emphasised in the UK-EOF report.


• I think it’s important to define the scope of this proposal – is it about citizen science projects related to biological recording (species and habitats) or the wider natural environment, or other branches of science too, including social science? There are successful citizen science projects in other areas, e.g. astronomy, so would this project seek to set standards for those areas, or rather learn from them to set standards in a more focussed (natural history) area?

I agree but I also wanted the concept to be open to everyone’s ideas as possible, rather than imposing my blinkered views. Given that this is presented to the BES CS SIG with the possibility of developing a proposal for NERC you can assume a scope somewhere twixt the two, whilst remaining open to opportunities to learn from what’s been done in areas beyond - as well as within - the CSSIG/NERCsphere.

Whereas a CS project looking at the success of polecat conservation efforts in Wales, has a specific focus, it probably has elements in common with other mammal projects, conservation projects and projects in Wales etc. Some project elements e.g. mapping, education, will also be found in a very large number of schemes, whilst things like project planning, budget management will also be pretty much the shame in other (CS) projects, including those involved with astronomy.

Should the scope for what the platform might include encompass administrative and promotional elements, the people engagement/involvement/training elements or just the tools that the participants/coordinator(s) might use to obtain and report on the project results? [At the moment, I’m happy to ‘include everything in’

As a starting point, one approach would be to use the UK-EOF Guide to create a generalised project template broken down by major phases (inception, preparation, design & testing, implementation, analysis & reporting etc.) and the component parts and the sub-tasks of each of these. This would then give you the means to deconstruct any project, and to identify the elements common to and overall profile of, e.g. a recording project. [I wonder how easy it might be to do this with the UK-EOF case studies] and perhaps the tools they used or could have].  This approach might be the one to take if the primary concern were to develop the platform first and harmonise existing tools first.

An alternative route would be to consider the goals related to different spheres of interest conservation, education, awareness raising, biodiversity monitoring etc. and to identify the tools that are needed/already exist to address these goals. This approach would fit in with letting the exemplar projects, addressing specific goals, provide the means for a more distributed approach to begin creating the platform.

Getting the scope right and getting the concept across to others would probably require some amalgam of the two.

 

• The NBN is already working on a way of ‘branding’ the standards and tools for the capture, management (including verification) and dissemination of biodiversity data and derived information products, to help users understand how things join up…especially those suffering from ‘app overload’!

It is really good to learn this. [Derived information (indices/assessments/graphs!) has been lacking for far too long – not least for giving useful feedback to project participants. The R module to work with NBN data will be great as well – and should really open up the database to academic use (and encourage data flow with the institutions as well).]


• The proposal mentions ‘added guarantee of success’ (point 3.a). If this is put to NERC it needs to recognise that there are different criteria for ‘success’ of a citizen science project [etc.]

I absolutely agree and it is unlikely that any project funded by a research council, the Lottery or one of the Environmental Funders Network would be judge on a single criterion. The point is perhaps that a project which uses well-tested tools, tailored to a projects’ goals and compliant with the appropriate standards will be more likely to e.g. produce more, better quality, statistically significant records, have an impact on conservation policy, raise public awareness provide the participants with a satisfying experience that leaves them happy to put their increased skills to use in future projects. If the platform would help to ensure this, whilst making projects easier to set up and implement, it would seem worth considering.


At this point the question is really whether this suggested platform would be a way of encouraging and enabling more Citizen Science Projects to be delivered more effectively from all viewpoints.

  • Does the idea have potential? 
  • Do you perceive obvious drawbacks?

I would very much like to see the potential pros and cons thrashed out here so that we can determine whether there is merit in the Citizen Science Community developing a more fully-fledged proposal.



Charles Roper

unread,
Feb 16, 2014, 9:21:29 AM2/16/14
to nfbr-...@googlegroups.com

Good discussion. I have some points to add. Will try asap, but coming week is a bit full on. Give me a poke if you don't hear anything.

Charles

Charles Roper

--
--
You received this message because you are subscribed to the Google
Groups "National Federation for Biological Recording" group.
To post to this group, send email to nfbr-...@googlegroups.com
To unsubscribe from this group, send email to
nfbr-group+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/nfbr-group?hl=en?hl=en
 
NFBR website: http://www.nfbr.org.uk/wiki/index.php5?title=Main_Page
---
You received this message because you are subscribed to the Google Groups "National Forum for Biological Recording" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nfbr-group+...@googlegroups.com.
Visit this group at http://groups.google.com/group/nfbr-group.
To view this discussion on the web, visit https://groups.google.com/d/msgid/nfbr-group/98fae981-617f-4d8f-b388-f826e227f74c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
Reply all
Reply to author
Forward
0 new messages