3. CROWDLAW | 10 RECOMMENDATIONS FOR DESIGNING BETTER CROWDLAW INITIATIVES
In this section we summarize ten recommendations for the thoughtful design of crowdlaw initiatives organized around: clarifying the demand for participation, increasing the supply of public participants and information, and experimenting with and improving on initiatives. The recommendations are relevant across all stages of the legislative process.
This is a draft version of the report (dated October 12, 2017) and will be updated in November.
What distinguishes successful crowdlaw processes is not the choice of technology platform, but whether the process is well-integrated into the mechanics of lawmaking. Does participation go beyond mere suggestions to yield constructive participation? Is participation curated to foster meaningful engagement? Is work done by the people integrated into the workflow of professional staff? These recommendations summarize our key learnings about how best to design initiatives such that crowdlaw enhances the legitimacy and effectiveness of lawmaking.
Clarify the demand for participation
1. Optimize for institutional as well as public engagement. Crowdlaw designers concentrate on making public input easy for individuals, but to be successful any process also needs to make input useful to institutions. Therefore, consider the needs of the government and public servants and create platforms and processes that account for obligatory and acculturated processes and the staff’s capabilities. This principle does not rule out changes to legal procedures or the hiring of staff with relevant skills to enable public participation, but there must be a workflow that makes the inputs from participation usable. For example, of the 2 million petitions submitted on the White House’s We the People e-petition platform, not one can be directly tied to a government action, arguably because a petition with no supporting documentation creates, rather than alleviates, work for public officials. Such a platform is not well-designed to enhance decision-making, nor does it create an established process for channeling the right information to the right policymakers.¹
2. Design to achieve the desired goal. Public engagement has the potential to foster democratic legitimacy, build social cohesion, increase government accountability, and improve the quality of legislation. An engagement process that asks people how they feel about a draft bill accomplishes a different purpose from a process that asks them, for example, to supply data to inform the crafting of the bill, or that invites them to use cellphones to monitor its implementation. The choice of goal will dictate what constitutes a successful system and the information or action that is sought from the public. Thus, if the goal is to obtain implementable proposals, do not ask merely for ideas. Ask for ideas supported by evidence. In the case of Peer to Patent, which was engagement by an administrative agency not the legislative branch, the United States Patent Office asked participants to supply information that would help a patent examiner determine whether an invention met the criteria for a patent. They did not, however, ask for people’s opinion about the patent, since that would have been legally irrelevant to the decisionmaking process.
3. Identify who will manage the process. One finding that consistently emerged from our analysis of global cases was that moderating the discussion during engagement is crucially important. Poor moderation can quickly derail the process and leave participants confused and frustrated. Although peer-to-peer community moderation can help to distribute the work by inviting the public to moderate one another, such as by upvoting and downvoting and flagging contributions as spam or abusive. The plan must also include a professional team to respond and explain how input will be used. This is an important part of the bridge connecting the public and government participants. For instance, LabHacker/E-Democracia in Brazil uses 200 volunteer legislative consultants to serve as “technical translators” between citizens and representatives and help to ensure that input meets legal requirements.² Similarly, the process on Parlement & Citoyens is facilitated by volunteers. The same directive applies to offline engagement: citizen assemblies coordinated by Ireland’s We the Citizens ran successful regional meetings largely because of the role of skilled moderators.
4. Plan for use, not only solicitation. Public engagement without an institutional learning mechanism for taking outside contributions onboard and integrating them is frustrating for all involved. By analogy, government can open and publish its public procurement data, but such transparency does not in itself reduce corruption. Rather, public institutions have to learn how to use such data to change how they buy goods and services. Similarly, capacity must be built within legislatures to curate and use public input. This might require changes to the current processes by which legislation is proposed, drafted, negotiated, and implemented. As important as soliciting public input is, there must be a corresponding learning mechanism for redesigning how the parliament operates to make beneficial use of engagement.
Increase the supply of public participants and information
5. Focus on incentives. Ask: “why should a member of the public participate?” and get the answer by talking to and surveying potential users. Crowdsourcing literature indicates that perceived meaningfulness and fairness are critical to the quality of contributions and the viability of crowdsourcing platforms.³ The onus is on managers not only to design a process that can have meaningful impact on government, but to articulate for the public their potential for impact, while making it easy for them to do so. In other words, make the rationale for participation explicit and “sell” the reasons to participate through both good design and clear explanation. On Decide Madrid, a platform launched by the Madrid City Council for public participation in decision-making, the section where users can make proposals is much more popular than the discussion section because proposals are binding and have the potential to create change, whereas discussions are simply fora for more discussion. A survey of 482 users who had not registered for Decide Madrid found that 11% said participation was pointless, and 27% said they lacked time to participate — — the most common reason cited for non-participation.⁴ If an individual cannot quickly engage on a platform, it will be very difficult to overcome that reluctance through other incentives.
6. Explain clearly how to participate. The legislative process is complex, with many more bills proposed than ever become law. Therefore, a successful public engagement must explain the process and what is being asked of the participant, including setting out thresholds for action, such as the number of signatures required or what it takes for a comment to be considered. Crowdsourcing literature indicates that when “average participants” are “asked to perform technical tasks with specific instructions and detailed job classifications, their performance is equal to or better than the performance of experts.”⁵ For instance, in the annual “Help Cut Red Tape” reports of British Columbia’s GovTogetherBC, clearly explains what an engagement is about, how input will matter, and when to participate. The reports details popular ideas for streamlining government, statistics about the participation process, ideas submitted, and the government action taken on the issue (Figure 1). Such information helps participants understand in real time how their participation is transforming into government action. The Lisbon Participatory Budgeting process drove votes from 2,800 in 2008 to 29,000 in 2012 by increasing the presence and clarity of the process, such as adding a feature allowing citizens to track the state of implementation of successful proposals, setting up mobile participation booths, and even touring the city with a “Participatory Budgeting Bus.”⁶
7. Respect privacy and authenticate users when needed. Although it is technically possible to certify residency or identity, decide whether and when such hurdles are necessary. For example, if the goal is to get the best ideas to solve a problem, does it matter where they come from? In order to direct opportunities to participate to people based on their interests, a voluntary request for information might be welcome where involuntary data collection on people’s preferences may not. As an example of participant vetting Reykjavik’s City Council is obliged to consider the 12–15 most popular proposals on the Better Reykjavik/Better Neighborhoods platform each month, so it authenticates participants using an electronic ID or password delivered through the citizen’s online bank to ensure one-citizen-one-vote. As an example of more complex authentication, Decide Madrid has a three-tiered system that determines the actions a member of the public can take consisting of:
Unregistered users may browse site content.
Basic verified users — verified through residence data and a mobile phone number — can post in discussions as well as create and support proposals.
Completely verified users — verified in-person or via mail — can do all of those actions plus vote on proposals.
8. Communicate the outcome of final decisions. Public officials should respond to contributions and endeavor to communicate regularly about outcomes. Even if the public is invited only to participate in making proposals at the outset, create a mechanism to share final outcomes. For instance, participants of vTaiwan engage in on-going deliberations with each other and with representatives of relevant government ministries. Participants know that if consensus is reached, the Taiwanese government must either adopt the idea or provide a response as to why the idea is not feasible. GovTogetherBC publishes the results of every engagement. On the other hand, the Irekia system in Spain’s Basque region lacks thresholds for when citizens’ proposals receive a government response or are deemed actionable, creating ambiguity around what it takes for government to actually engage with a citizen proposal.
9. Diversify engagement opportunities and diversify who participates. Empirical research suggests that participation opportunities may be failing to attract diverse participation. Ensuring participation by diverse members of the public is hard work, including investment in campaigns to recruit and give voice to the voiceless. A study of the representativeness of 186 of the participants who contributed ideas to improving an off-road traffic law found that they were overwhelmingly male (86%), had formal education, and were between 35–54 years old (46%). They also had previous civic experience: 72% of participants had written on an online forum prior, 41% had contacted a representative, and 33% had written an op-ed before.⁷ Attracting diverse participation was a common challenge for all case studies (detailed in the Discussion section). Causes of demographic imbalance include citizens’ time or ability to the contribute, their awareness of the platform, or their facility with platform technology. Our draft law outlines specific opportunities for groups to participate in lawmaking on issues of special significance (e.g. targeting specific economic, social, environmental, cultural, gender, or territorial issues) as a critical step in actively bringing underrepresented populations into the participation process.⁸ Thus, in addition to an online web portal that allows the public to submit proposals, participate in dialogue, and engage in participatory budgeting activities, the creators of Decide Madrid established 26 “Citizen Service Offices.” These offices are dispersed throughout the city and allow residents the opportunity to voice their opinions in person, if they so choose, in addition or in place of engaging online. The redrafting of the South African Constitution included an extensive communication strategy that distributed four million copies of the draft constitution in the drafting phase, and seven million copies of the final document, including illustrated guides for non-literate portions of the population. (An early survey of areas disconnected from the redrafting campaign helped to find areas that needed such communiqués).⁹
Experiment and improve
10. Test what works and iterate. Crowdlaw is a new phenomenon. In order to accelerate adoption, more research is needed, necessitating that practitioners and researchers collaborate to design experiments. Research can involving natural experiments to observe how the platform works, who participates, and how. Simple analytical software can generate data that platform owners and others can use to study a crowdlaw initiative. Always ensure that such administrative data is open and available. In addition, consider running simple controlled trials by dividing participants into two groups and presenting them with alternative experiences, comparable to the A/B testing. An example of such testing is to try different ways of explaining how to participate or testing participation’s relevance at different points in the legislative process. Also critical are surveys to gather information that can help in improving the effectiveness of the platform and process. When redrafting its constitution, the South African government surveyed citizens in areas not being reached by the campaign. By learning the needs and motivations of excluded South Africans, the campaign was able to drive meaningful participation opportunities: for many South Africans, “it was the first time they were able to interact directly with their elected representatives. It elicited nearly 1.7 million submissions […] and meetings reach[ed] approximately 95,000 people.”
- Gabriella Capone & Beth Noveck
Next posts in the series:
- #4. CrowdLaw | Framework for Institutionalizing Public Engagement in Lawmaking
- #5. CrowdLaw | The political context in Spain and elsewhere
- #6. CrowdLaw | Why public engagement? Normative goals
Or go to the first post in the series and table of contents here.
Footnotes
¹ Beth Simone Noveck, “Smart Citizens, Smarter State: The Technologies of Expertise and the Future of Governing” (Boston, MA: Harvard University Press, 2015): 75–76.
² Julie Simon, Theo Bass, Victoria Boelman, and Geoff Mulgan, “Digital Democracy: The tools transforming political engagement,” Nesta, February 2017, accessed June 26, 2017, http://www.nesta.org.uk/sites/default/files/digital_democracy.pdf
³ Helen K. Liu, “Crowdsourcing Government from Multiple Disciplines,” Theory to Practice (2017). See also: Dana Chandler and Adam Kapelner, “Breaking Monotony with Meaning: Motivation in Crowdsourcing Markets,” Journal of Economic Behavior & Organization 90 (2013): 123–33. Nikolaus Franke, Peter Keinz and Katharina Klausberger, “Does This Sound Like a Fair Deal? Antecedents and Consequences of Fairness Expectations in the Individual’s Decision to Participate in Firm Innovation,” Organization Science 24, no. 5 (2013): 1500.
⁴ Investigación, marketing y opinión, “VALORACIÓN DE LA ACCIÓN DE GOBIERNO EN EL AYUNTAMIENTO DE MADRID,” June 2016, accessed July 24, 2017, https://ahoramadrid.org//wp-content/uploads/2016/06/Info-Ahora-Madrid.pdf
⁵ Helen K. Liu, “Crowdsourcing Government from Multiple Disciplines,” Theory to Practice (2017). See also: Tara S. Behrend, David J. Sharek, Adam W. Meade, and Eric N. Wiebe, “The Viability of Crowdsourcing for Survey Research,” Behavior Research Methods 43, no. 3 (2011). Alexis Comber Linda, et al., “Comparing the Quality of Crowdsourced Data Contributed by Expert and Non-Experts,” PLOS ONE 8, no. 7 (2013).
⁶ Giovanni Allegretti and Sofia Antunes, “The Lisbon Participatory Budget: results and perspectives on an experience in slow but continuous transformation,” The journal of field actions: Field Actions Science Reports Special Issue 11 (2014), accessed June 23, 2017, https://factsreports.revues.org/3363#ftn1
⁷ Tanja Aitamurto, Hélène Landemore and Jorge Saldivar Galli, “Unmasking the crowd: participants’ motivation factors, expectations, and profile in a crowdsourced law reform,” Information, Communication & Society 20, no. 8 (2017): 1239–1255. See also Huang, S.W., Suh, M.M., Hill, B.M., Hsieh, G., “How activists are both born and made: An analysis of users on change.org,” Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (2015): 211–220.
⁸ This is the case for the popular initiatives mechanism. Consult Appendix III for an overview of the client’s draft participation law.
⁹ Catherine Barnes and Eldred De Klerk, “South Africa’s multi-party constitutional negotiation process,” Conciliation Resources (2002): 33. Available at: http://www.c-r.org/accord/public-participation/south-africa-s-multi-party-constitutional-negotiation-process
