Ideation in an online participatory platform: Towards a conceptual framework 1

Abstract

With the development of online participatory platforms, many governments are extending consultation processes beyond traditional face-to-face meetings. These online spaces are used by government agencies to collect ideas from the public and to aggregate preferences through non-binding voting procedures. Many of these digital platforms allow residents to post comments, creating potential for ideation processes. Examining data from an unmoderated online platform initiated by a municipal government, this paper explores whether ideation emerged organically in the deliberation process, and if so, how this occurred. After analyzing over 800 comments in the 20 most highly discussed ideas debated in the forum, the study confirmed that ideation took place and identified five main activities in the process: idea proposal, idea reception, idea development, idea closure, and idea implementation. It also distinguished eleven different roles performed by participants throughout the ideation process: initiating, supporting, disagreeing, clarifying, informing, arguing, commentary, expanding, wrapping, mobilizing, and implementing. The study also found that ideation patterns differed by idea theme and by the level of controversy of particular issues. The paper concludes with ideas for further research and recommendations for the design and implementation of online public participation processes.

Keywords

Online community engagement ideation e-government civic participation

1. Introduction

Throughout history, governments around the world have engaged residents in consultations to gather their input on a variety of issues. In the last two decades, with the development of information and communication technologies (ICTs), these consultations have also taken place online. In some cases, government agencies not only ask for feedback about specific policies, plans, programs and activities, but also invite residents to propose and discuss ideas through online participatory platforms. These online ‘invited spaces’ are institutional arrangements created by government agencies that provide opportunities for communication, public deliberation and eventually co-governance. These opportunities range from finding solutions to community problems to drafting municipal and national constitutions [1].

To promote online public engagement, governments use social media (e.g. Twitter, Facebook, YouTube) and a great variety of participatory platforms (e.g. MindMixer, Peak Democracy, UserVoice, Change.org). These digital platforms help local government consult, collaborate with, and even empower citizens to make decisions with lower time and travel costs and more access to information than face-to-face participatory meetings [1, 2]. With access to a computer or mobile device, anyone interested in selected issues can potentially participate in these spaces regardless of their age, race, socio-economic status, citizenship, or location. In this regard, it has been argued that, despite the digital divide, online participation is an effective avenue to expand access because many people face lower social barriers in online spaces than in face-to-face meetings [2, 3]. Over time, these democratic innovations are expected to make governments more transparent [4, 5], accountable [6, 7], and effective [8].

It is pertinent to clarify that not all online discussions are collective or collaborative in nature. Indeed, some online forums are characterized by uncivil exchanges and uninformed statements [9]. In these forums, participants are often opinionated and post their ideas without carefully reading previous comments. Nevertheless, due to their communicative nature [10], online invited spaces can be places for “ideation”, which in the context of our study refers to the collaborative process of generating and developing ideas [11].

Although online invited spaces have proliferated in the last decade there is still little research on the ideation process in these online participatory forums. In this study, we were precisely interested in exploring to what extent ideation occurs in an online invited space, particularly when ideation is not planned or even intended by government officials and when there is an absence of facilitation.

By examining participants’ comments in an unmoderated online forum managed by a U.S. city, we explore the main features of the ideation process. Our research contributes to the fields of public management and civic engagement by providing a framework for understanding ideation processes and a typology of roles played by participants. We also expect that our ideation framework will help public managers in the design and management of participatory processes.

2. Literature review and conceptual discussion

2.1 Online public participation and deliberation

When governments seek to increase online public participation in their decision-making processes, it is important they understand the nature of online participation. Since online participation spaces lessen some of the limitations of traditional participatory forums [12], it has been argued that they can provide a better environment for communication and deliberation than traditional face-to-face settings [12]. As there are fewer constraints of time or distance, discussions could be more interactive, inclusive, and deliberative [15, 16], allow the “ongoing co-creation of new ideas” [17] and encourage the wisdom of crowds [18].

Having said that, taking into account that an ideal deliberative procedure requires public argument and reasoning among equal citizens [19], it is pertinent to recognize that online forums are characterized by participation inequalities, with an over-representation of young, male, and white users [20]. There tends to be a strong concentration of contributions from a small group of very active users [21]. Moreover, it has been found that idea proposers actively participate in the discussion of their own ideas [25].

In order to improve both the quantity and quality of online public participation, government responsiveness (such as taking responsibility for the participatory process and providing feedback) is crucial [26]. Other factors that increase the likelihood of success are reward systems, awareness of the participation opportunities [26], accessibility to the online platform [27], transparency of the decision-making process [28], and integration of online and offline processes [29, 30].

2.2 Ideation

In this study, ideation refers to the collective process of generating, developing, and implementing ideas [11]. Throughout the process of ideation, participants develop a better idea than the original one by building on others people’s contributions. Ideation, then, can be understood as processes of innovation and collective creation that are premised on sharing ideas and mutual collaboration [31].

Ideation includes several stages starting with initial conception and ending with eventual realization [31]. In the literature, it is possible to identify seven distinct stages. The first stage is “innovation” based on knowledge accumulation [34]. In other words, an idea has to be conceived by someone based on his or her own knowledge. Second, the new idea needs to be registered or incubated [31, 34]. Unless the idea is presented and registered, it remains as implicit thought. Third, the idea is recognized by someone [31]. Fourth, the idea is developed and improved [31]. Fifth, the idea is perceived as “implementable” [31]. Sixth, the idea reaches key actors who are interested in its implementation and in amassing the necessary material and human resources to make it a reality [31, 34]. The final stage is the eventual implementation of the idea [31, 34]. Although not every idea makes it through to actual realization – there is always a possibility in each stage that an idea fails to proceed to the next step – the life cycle of ideation can be summarized in three main stages: 1) innovation, 2) development, and 3) actualization.

Although the academic literature on public deliberation seldom uses the term ‘ideation’, it alludes to some of its elements. For instance, in a case study of crowdsourced legislative process, discussions moved towards collaborative problem-solving [35]. In another study in the context of the crowdsourcing model, scholars found that creative solutions are developed through harnessing collective intellect [17]. Other studies evaluated ideas as most innovative if they received many comments [25] or if the quality of interaction was high [36]. In other words, peers facilitated discussions by evaluating, criticizing or challenging ideas with their comments and therefore developed the idea together [25]. It is also pertinent to note that people have different interests and therefore play different roles in the ideation process [17].

2.3 Participants’ roles and contributions in online forums

Generally, the roles of participants in online participatory forums are not given or firmly determined in advance. Most participants join online participatory forums without a given role assigned by someone else because the nature of participation is voluntary. Sometimes incentives or rewards are provided for participation, but the decision to participate or not is solely on the participant’s will.

That said, participants do take on different roles. Depending on the topic and the flow of discussions, a participant might play several different roles even within one idea. Some online forums explicitly assign moderation and facilitation roles to specific participants. Prior studies have found that moderation and facilitation in online deliberation tend to improve the quality of the discussions [14, 26, 35, 37]. For instance, moderators are expected to step in when certain challenges to deliberation appear, such as low participation among or domination of particular individuals or groups, inappropriate, disrespectful, or highly emotional behavior, excessive off-topic conversations, and excessive (dis)agreement or politeness [38]. Indeed, a facilitator can include more people into the discussion, inform participants so they understand the issues better [15], encourage deliberation and consensus building [35], and promote a democratic process [14]. Barber [37] argues that it is not a matter of whether or not to facilitate or mediate, but rather in which way a forum should be facilitated or mediated.

When an online participatory forum is initiated and managed by a government agency, the role of managers is also important. Since citizens’ inputs can help initiate new ideas, managers need to consider how best to maximize these inputs [17]. Government managers are expected to design the forum to be user-friendly [15], be responsive and committed to the participatory process, and try to integrate ideas gathered into policy decisions [14]. Managers can have different levels of intervention, from regular monitoring of the conversation to active participation. The decision on what roles managers play in an online participatory forum is often the result of a discussion within the government agency before the forum is open for public participation.

3. The case study

The online forum studied in this research was part of a city’s process to update its general plan for the city’s long-range growth and development as required by state law.2

²
Even though the comments posted on the online forum are open to the public and anyone can access if they open an account, we keep the site anonymous in this paper to reduce the traceability of participants.

In this process, city staff met with over 11,000 residents through more than 125 community meetings in order to listen to residents’ ideas. To enrich outreach efforts, the city expanded the process to a web-based discussion forum. Both in the face-to-face meetings and the online forum, two overriding questions were asked: What do you love about our city? and What are your big ideas for our city’s future?

Participation in the online forum was free and open. Anyone who was interested could register and participate. The sign-up process was simple, requiring a name, email address, postal code, year of birth, and gender (optional) and was also connected to social network services such as Facebook, Linkedin, and Google $+$ . Participants could engage in four ways: proposing ideas, reading ideas, leaving comments, and voting for ideas. The city also designed a point system to reward active participants and facilitate online participation.

4. Data and methods

In order to analyze online deliberation in this forum, we selected ideas that generated the most comments, with a minimum threshold of over twenty comments. This was done to ensure a critical mass of interactions in order to examine the ideation process, as having a certain number of comments (therefore forming a discussion) is essential [23]. This resulted in a database encompassing 20 ideas and 836 comments. These comments were posted from August 2012 to December 2013. The number of comments in each discussion ranged from 21 to 113.

Table 1
Definitions of participant roles

Idea proposal
Ideation stages and roles	Definition
Initiating	Starting the discussion thread
Idea reception
Supporting	Simple expression of favor or agreement to the idea
Disagreeing	Mere expression of dislike or disagreement to the idea
Idea development I: Set-up
Informing	Providing facts and information directly related to the topic without somebody asking for clarification
Clarifying	Asking questions and providing answers; clarifying terms in order to keep everyone on track
Commentary	Adding interpretation or personal experience to the information
Idea development II: Improvement
Arguing	More detailed opinions, going beyond simply supporting and disagreeing comments
Expanding	Brainstorming, bringing in new perspectives, and operationalizing the idea
Idea closure
Wrapping	Summarizing and synthesizing the discussion
Mobilizing	Moving ideas to the next phase and encouraging to put them in action
Idea implementation
Implementing	Operationalizing the idea

Table 2

Number of comments in the most discussed ideas

Rank by number of comments	Number of comments	Focus of idea	Level of controversy
1	113	Transportation/roads (T/R)	3
2	89	T/R	2
3	86	T/R	2
4	60	Land use/neighborhood (L/N)	1
5	53	Tourism (T)	2
6	49	L/N	1
7	45	L/N	2
8	35	T/R	3
9	34	Environment/health (E/H)	1
10	33	L/N	2
11	29	E/H	2
12	29	T/R	1
13	25	T	1
14	25	E/H	2
15	24	L/N	2
16	22	T/R	1
17	22	E/H	1
18	21	T/R	2
19	21	T/R	3
20	21	T/R	2

Level of controversy index: 1 – overall balanced, 2 – high agreement, and 3 – high disagreement

Using a grounded research perspective, the research team identified themes related to ideation phases, roles, and types of ideas. From there, we analyzed who participated, the roles of participants, the degree of controversy of each idea, and how the process evolved in terms of idea development.

We followed several steps in our analysis. First, the authors reviewed the comments independently and selected themes related to the ideation processes and roles. In our analysis of the ideation process, we organized the posts into 11 categories that reflected the roles performed by participants: initiating, supporting, disagreeing, clarifying, informing, arguing, commentary, expanding, wrapping, mobilizing, and implementing. We also had an ‘other’ category (see Table 1). Subsequently, we had several meetings to discuss themes and develop the initial framework for the analysis of the data. One of us coded the comments according to the themes, and the other two authors reviewed the coding of the data and noted disagreements in coding. Our inter-coder reliability with Kalpha was 0.9687 [39].3

Kalpha $=$ 0.80 in the Krippendorff’s Alpha (Kalpha) reliability test is often considered the norm for a good reliability [39].

We discussed disagreements until we reached a consensus and then finalized the categories and terminology.

We also organized the topics of the 20 most discussed ideas into four main content categories: transportation/roads, land use/neighborhood issues, environment/health, and tourism (Table 2). The topic that garnered the most discussion was transportation/roads, followed by land use and neighborhood issues, and by topics related to environment and health. In addition, in order to determine the degree of controversy, ideas were coded in terms of the number of agreement and disagreement comments in each idea. This resulted in three categories: overall balanced, high agreement, and high disagreement.

5. Findings

5.1 Descriptive findings

A total of 127 people participated in the discussions. Interestingly, 15.7% of participants submitted almost 80% (78.8%) of the comments. This situation illustrates the Pareto principle (or the “law of the vital few”): small groups of people tend to make a strong concentration of contributions [21, 22, 23, 24]. Moreover, the five most active participants submitted half of all comments.

Some of the most active participants in the forum were those who proposed the original ideas. The 20 ideas analyzed were initiated by 15 people, and twelve of them were also among the 20 most active participants who contributed 78.8% of comments. Only three initiators did not participate at all or leave a comment on their own idea. This does not mean that people just commented on their own ideas without a meaningful discussion with other participants. Among the 836 comments, 661 comments had a dialogue (continued with at least one reply) and 175 comments were considered as a monologue (no reply made). Among the 661 dialogue comments, 132 were made by the idea proposer him/herself and 20 out of 175 monologue comments were made by the idea proposer. This means that even if the idea proposers commented in their own ideas, they replied to other participants’ comments which means that they “talked” with their peers.

In terms of gender, 52 (40.9%) of the participants in these 20 ideas were identified as female, 68 (53.5%) as male, and 7 (5.5%) not identified.4

⁴
We had to assume gender of participants based on their names and profile pictures. Seven people are undefined as they had unisex names and kept their profile private.

Although the forum had a slightly larger presence of males than females, there was a significant gender disparity in terms of active participation: males contributed 75% of the posts while females contributed only 24% (1% of posts were made by those who could not be identified by gender). Moreover, among the top 20 most active participants, only six were female (30%) and the five most active participants, who made almost half (49.3%) of the contributions, were all male.

In terms of spatial representation, most of the online participants lived in one particular geographic area. The city found this out when monitoring zip codes of participants, and then reached out to other areas in an attempt to balance the participation from all neighborhoods of the city. This made the general plan update process more hybrid, combining online and face-to-face meetings [30].

City managers (e.g., a project manager, village planners) and leadership committee members also joined the online forum occasionally. During the 15 months of our data collection period, eight managers left 54 comments, accounting for only 1.4% of the total comments of the forum overall. In the 20 most discussed ideas, only three public officials contributed, making a total of six comments. These comments aimed mainly at expressing appreciation for residents’ participation, noting their agreement and support of a particular idea, or providing specific information about a project related to an idea. Interestingly, none of them aimed at facilitating or moderating the process of ideation .

5.2 Ideation

5.2.1 Roles in ideation

The online forum was organized and managed to collect residents’ ideas for the city’s general plan update. The city did not expect the forum to be used as an ideation space. Their focus was on the ideas proposed by residents and on the aggregation of their preferences, not necessarily on comments to improve the idea. However, the participants took advantage of another available way of participation provided by the online platform, leaving comments and considering the ideas proposed with others. During the deliberation process, people contributed to the original idea in various ways, including its further development and refinement. In other words, the process of ideation evolved organically during the online deliberation process.

The initiators who participated in the discussions usually contributed comments that clarified their ideas (21.2%) and argued in favor of those ideas (20.4%). Other participants simply expressed their support for ideas (28.5%) or tried to expand the idea (17.6%) by bringing new perspectives or ways to operationalize it. These findings suggest that initiators played a significant role in igniting interest for an idea but a limited role in amplifying the discussion by actively participating in idea development compared to other participants.

In this forum, wrappers contributed to the discussion by summarizing and synthesizing the discussion, and mobilizers made concrete suggestions for the next steps for implementing actions. Interestingly, wrappers and mobilizers appeared in only about 2.5% of all 836 comments. In most discussions, the idea initiators were more likely to play the role of wrappers and mobilizers, probably because they became impatient with the length of the conversation and wanted to move their idea to fruition.

Figure 1.

Ideation framework: five components of the ideation process.

5.2.2 Levels of ideation

As noted above, the three authors coded all comments in the top 20 most discussed ideas in order to figure out what kind of contribution a comment represented in the discussion. From these categories, a framework of ideation was developed. The framework identifies five levels in the ideation process: idea proposal, idea reception, idea development, idea closure and idea implementation (Fig. 1). In each level of ideation, participants play particular roles. Strictly speaking, the first and last levels could be understood as input and output respectively, while the ideation process takes place in the three stages in between them.

5.2.2.1. Idea proposal

An idea needs to be initialized and shared by someone. This could take different expressions: a question, a paragraph, a short or full essay, an image, a video, or a combination of them. This stage provided the basic setting of the ideation by igniting the process and enabling other people to participate and start a discussion.

5.2.2.2. Idea reception

After an idea was proposed, the simplest reaction from the participants was either ‘supporting’ the idea or ‘disagreeing’ with the idea. We call this stage idea reception. Thus, idea reception includes two main options:

•
Supporting: Simple expression of favor or agreement to the idea. Supporting comments were simple expressions of agreement with the idea such as “I agree, agree, agree.” or “Wow, this is a great idea!”
•
Disagreeing: Mere expression of dislike or disagreement to the idea. Comments expressing disagreement consisted of statements rejecting the ideas proposed by other participants (e.g. “I am against [this idea]” or “DON’T BUILD THE HIGHWAY!!!!”).

In this particular forum, idea reception comments amounted to about one-fifth (19.4%) of all comments. Among them, most were supportive comments (15.9%) while only a minority (3.5%) rejected the idea.

5.2.2.3. Idea development

Original ideas were further developed in the online forum through five main actions: informing, clarifying, commentary, arguing, and expanding.

•
Informing: Providing facts and information directly related to the topic without somebody asking for clarification. “Informing” comments provided facts that were related to the topic, sometimes with specific information in the post, but most often referring participants to other sources of information such as websites and articles. For instance, “Here’s a good site with the scoop on all the bike lanes; what’s in place and what’s coming…”, or “A great article about bike infrastructure and how it improves the economy…”
•
Clarifying: Asking questions and providing answers, clarifying terms in order to keep everyone on track. The “clarifying” category included comments such as “what, exactly, is the cost of providing free street parking?” and “Yes, whenever the words ‘surface parking’ appear, remember that street parking is a subset of surface parking.”
•
Commentary: Postings labeled as “commentary” added interpretation or personal experience to the information. For example, “I also bicycle on sidewalks whenever possible – downtown the streets are old and lumpy – sidewalks are a smoother and safer surface to bicycle on down here.” and “At least the palm trees provide food for woodpeckers – I’ve seen them dining on the that infest the palms.”
•
Arguing: More detailed opinions, going beyond simply supporting and disagreeing comments. For instance, “Commuter rail doesn’t create sprawl. Sprawl is low density development, but train stations encourage walkability, and that means high density development.” Another example of a comment labeled as “arguing” can be found in this posting: “If it’s really worth that much to the hospital and businesses, then they should pay for it.”
•
Expanding: Brainstorming, bringing in new perspectives, and operationalizing the idea. Two illustrations of these comments are “I just wanted to expand on your point about realtors and developers…It should be required that this is disclosed to buyers and that’s not happening.” and “Try this: Google “[city name] canal drownings” and then read a few of the stories […] the major problem that seems to be […]. Moreover, in many cases, […]. Why? Because…”

Idea development comments represented about three quarters (75.6%) of all comments, with a relatively even distribution of the five actions: arguing (17.2%), clarifying (16.6%), commentary (13.6%), informing (14.2%), and expanding (13.9%). For the purpose of analysis, the idea development phase was further organized into two levels, taking into account the degree of contribution that each comment made to the refinement of the original idea:

•
Idea set-up: Informing, clarifying, and commentary (44.4%).
•
Idea improvement: Arguing and expanding (31.1%).

5.2.2.4. Idea closure

Some ideas under discussion went a little further when participants were able to synthesize the main points and move the idea to the next step, idea closure. This included two main actions:

•
Wrapping: Summarizing and synthesizing the discussion. For example, “This is encouraging, to say the least. Not only [do these] lend themselves to vibrant pedestrian-oriented development, but by making them more inviting and drawing focus to them would likely enhance surrounding neighborhood safety as well.”
•
Mobilizing: Moving ideas to the next phase and encouraging to put them in action. For example, “The next step would be to invite the grocery chains to locate potential markets for them. Then they could look for a lot. Next the grocery would contact the relevant council member.” and “If we want this issue to be in the budget, we must attend budget hearings and petition for it.”

Idea closure can occur formally when facilitators try to summarize and close the discussion. When a formal facilitator is not present in the forum, this could happen organically, when some participants spontaneously take the role of wrappers and mobilizers. Comments categorized as idea closure accounted for only a small portion (2.5%) of the total comments in the top 20 most discussed ideas. Of those comments, about 1.0% were categorized as wrapping and 1.56% as mobilizing.

5.2.2.5. Idea implementation

The last stage of the ideation process consists of idea implementation. The idea can be implemented by the participants themselves, or sometimes needs another implementer such as government agencies, private companies, non-profits, or other organizations. This stage was not presented in the online forum we analyzed, as it was an invited space for brainstorming style of idea gathering. However, it is an important stage in ideation.

5.2.2.6. Other

Although the vast majority of comments belonged to one of the five levels, we identified a few contributions made by participants that did not relate directly to the specific content of the ideas under discussion but to the ‘process’ itself, as well as expressions of appreciation or apologies.

Figure 2.
Idea controversy and ideation stages.

5.2.3 Ideation framework

Our conceptual framework of the levels of ideation was presented earlier in Fig. 1. Although for analytical purposes we described the levels of ideation between proposal and implementation (reception, development, and closure) in sequential order, in reality they did not necessarily appear in a sequence. After an idea was proposed, its reception, development, and closure sometimes appeared throughout the discussion in a scattered way.5

⁵
In some ideas, certain phases of the framework (especially the idea closure which includes wrapping and mobilization) did not appear at all.

One reason for this can be attributed to the nature of online deliberation in this particular platform. Participants were not involved all together at the same time. Rather, they joined the discussion anytime they felt free to visit the website, coming and going continuously. There were cases in which someone had tried to summarize the discussion and wrap it up, but then a new participant joined the discussion with a new expanding idea. A second reason could be found in the rules and functioning of the forum. There was no formal moderator or facilitator so regular interventions for efficient discussion did not occur. In addition, as there was no designated time frame for discussion and participants were not notified when discussions would end. Hence, few participants may have felt the need to contribute to idea closure, summarizing the discussion and mobilizing people to the next step of implementation.

The ideation process can be illustrated by looking at how one idea (about bicycle infrastructure) evolved in the forum. First, the idea initiator posted two paragraphs of text and a YouTube video about best bikeway designs. The initiator of the thread proposed that the city develop a separate bike road system for bike riders’ safety, and that such a system could take advantage of canal ways and residential streets. Then, other participants made comments expressing reception of the idea (either supporting or disagreeing), developing the idea (informing, clarifying, commentary, arguing, and expanding), and closing the idea (wrapping and mobilizing). They started to elaborate on the initial idea and introduced new but related proposals, provided more details, and operationalized some of the proposals. The discussion lasted for about a year and one month. Over time, there were at least 9 different ideas that emanated from the first one. Throughout the ideation process, the initial suggestion widened, and participants signaled other concerns and made additional proposals.

5.3 Controversy

We also looked at whether or not patterns of ideation would differ by the level of controversy (Fig. 2). There was not much difference in participation in the idea reception stage when comparing level of controversy. Not surprisingly, supporting comments appeared more frequently in less controversial ideas, whereas disagreeing comments were more prevalent in the more controversial issues. However, differences appeared in the idea development stage. While the high disagreement ideas had 51.5% of comments in the idea development set-up stage, other two groups had fewer comments in the same stage (41.6% in the high agreement ideas and 44.8% in the overall balanced ideas). Instead, the high agreement ideas were comparatively more present in idea development improvement stage (34.3%) than the other two groups (23.7% in the high disagreement ideas and 30.7% in the overall balanced ideas).

Table 3
Idea controversy, ideation stages, and type of contributions

	Idea controversy
	Overall balanced		High agreement		High disagreement
Ideation stages and type of contributions	#	%	#	%	#	%
Idea reception	47	19.50	82	19.25	33	19.53
Supporting	47	19.50	75	17.61	11	6.51
Disagreeing	0	0.00	7	1.64	22	13.02
Idea development I: Set-up	108	44.81	177	41.55	87	51.48
Informing	40	16.60	67	15.73	12	7.10
Clarifying	40	16.60	55	12.91	44	26.04
Commentary	28	11.62	55	12.91	31	18.34
Idea development II: Improvement	74	30.71	146	34.27	40	23.67
Arguing	27	11.20	85	19.95	32	18.93
Expanding	47	19.50	61	14.32	8	4.73
Idea closure	8	3.32	8	1.88	5	2.96
Wrapping	3	1.24	2	0.47	3	1.78
Mobilizing	5	2.07	6	1.41	2	1.18
Other	4	1.66	13	3.05	4	2.37
Process	4	1.66	9	2.11	3	1.78
Other	0	0.00	4	0.94	1	0.59
Total	241	100	426	100	169	100

When we looked for more details (Table 3), we found some differences by type of comment. In the idea development set-up level, while the high disagreement ideas included more clarifying and commentary comments than others, the high agreement ideas and the balanced ideas had more informing comments.

Table 4

Ideation stages and contributions by focus of ideas

	Focus of ideas
	Transportation/road		Land use/neighborhoods		Tourism		Environment/health
Ideation stages and type of contributions	#	%	#	%	#	%	#	%
Idea reception	47	15.8	44	20.9	31	24.8	40	19.7
Supporting	24	8.1	40	19.0	29	23.2	40	19.7
Disagreeing	23	7.7	4	1.9	2	1.6	0	0.0
Idea development I: Set-up	140	47.1	82	38.9	52	41.6	98	48.3
Informing	24	8.1	38	18.0	13	10.4	44	21.7
Clarifying	69	23.2	24	11.4	22	17.6	24	11.8
Commentary	47	15.8	20	9.5	17	13.6	30	14.8
Idea development II: Improvement	96	32.3	74	35.1	34	27.2	56	27.6
Arguing	70	23.6	26	12.3	20	16.0	28	13.8
Expanding	26	8.8	48	22.8	14	11.2	28	13.8
Idea closure	5	1.7	5	2.4	4	3.2	7	3.5
Wrapping	3	1.0	2	1.0	1	0.8	2	1.0
Mobilizing	2	0.7	3	1.4	3	2.4	5	2.5
Other	9	3.0	6	2.8	4	3.2	2	1.0
Process	7	2.4	5	2.4	2	1.6	2	1.0
Other	2	0.7	1	0.5	2	1.6	0	0.0
Total	297	100	211	100	125	100	203	100

In the idea development improvement level, expanding comments appeared more in the balanced idea group than other two groups. Although the number of comments in the idea closure stage was too small to compare, it is notable that wrapping and mobilizing comments appeared less in the high agreement ideas compared to the other two categories.

5.4 Focus of ideas

In addition, we looked at patterns of ideation by focus of ideas (Table 4). Interestingly, patterns of ideation appeared differently by idea themes. First, there was a higher proportion of clarifying (23.2%) and arguing (23.6%) comments in the ‘transportation and roads’ theme. This could be explained by looking at the idea reception phase. The idea reception phase in the ‘transportation and roads’ theme had about the same number of supporting (8.1%) and disagreeing (7.7%) comments, whereas in the other topics there were very little number of comments (or no comments at all) against the ideas proposed. The higher level of controversy found in the ‘transportation and roads’ theme may explain why there was a higher proportion of clarifying comments in this theme than in the other ones. This implies that the participants in this theme wanted to make sure that they were on the same page regarding the issues, and also actively argued their opinions compared to the participants who joined in other themes. Meanwhile, the participants seemed to mostly support the idea of promoting tourism, land use/neighborhoods, and environment/health related issues. In particular, there was no comment disagreeing with the ideas in the ‘environment and health’ theme.

Second, in the ‘land use and neighborhoods’ theme, the proportion of expanding comments was higher than other themes. In other words, the participants wanted to brainstorm, bring in new perspectives, and/or operationalize the ideas. One explanation for this may be that participants thought they had more direct impact on the issues related to ‘land use and neighborhoods’ compared to other themes. Third, the participants who commented in the ‘environment and health’ theme provided a higher proportion of informing comments (21.7%) than other themes. This may be because of the high level of general knowledge available on environmental and health issues.

The last point is that, although the difference is small, the ‘tourism’ and ‘environment and health’ themes had a higher proportion of the idea closure stage. These differences mostly came from the proportion of mobilizing comments (2.4% and 2.5% each). Since the participants mostly agreed on these ideas, they moved forward to find some ways to put the idea in action.

6. Summary and conclusions

This study examined the case of an online invited forum that gathered residents’ ideas on how to make their city a better place. The main goal of the study was to explore the ideation process in an unmoderated online participatory forum. There is limited research on this specific topic, particularly in the context of invited spaces at the municipal government level. Our contribution to the field is a proposed framework to study ideation processes and participant roles in unmoderated online invited spaces. The ideation levels framework emerged organically throughout the analysis. The framework includes five levels of ideation: idea proposal, reception, development, closure, and implementation. The comments analyzed in this study were located in the three middle levels: idea reception, idea development, and idea closure. In each level of ideation, people developed an idea by playing various different roles such as initiating, supporting, disagreeing, clarifying, informing, arguing, commentary, expanding, wrapping, mobilizing, and implementing.

The levels of ideation appeared in a slightly different pattern based on the level of controversy of the given issue. While idea reception stage appeared in a similar portion in each group, the high disagreement ideas were more active in idea development set-up. For high agreement ideas, idea development improvement was the most active. The high disagreement ideas had more clarifying and commentary comments than the other categories. Patterns of ideation also differed by idea theme.

Our study also confirms prior findings about inequality of participation in public forums [21, 22, 23, 24]. Consistent with the Pareto principle, approximately 20% of participants made about 80% of the contributions. Moreover, the top 5 participants made almost the half of all comments. This finding calls for strategies to motivate low-level participants and lurkers to take a more active role in the forum. It also recognizes that the online forum was not a solution to overcome limited participation in face-to-face meetings, but was just another tool that widened participation but still to a particular segment of the population.

7. Recommendations for future research and practice

We propose six recommendations for research, and three for practice. One interesting finding in this study was the manner in which ideas evolved. Although the city’s original intent was to provide a space to collect ideas and aggregate preferences, what also resulted was an organic emergence of ideation, as captured in our ideation framework. In the future it could also be interesting to undertake a comparative analysis of an unmoderated process like the one discussed in this paper, and a forum intentionally designed towards ideation that includes a facilitator. A facilitated version of the platform could make an explicit effort to draw voices that are silent, summarize discussions at certain points, encourage discussion by posing probing and reflective questions, and prompt participants to wrap up the discussion.

A second research recommendation concerns the roles of participants. Due to the limitation of the data, we could not explore whether the roles in ideation process were correlated with any participants’ background characteristics such as their civic organization affiliation or education level. It would be interesting and worthwhile to take a look at the roles in the ideation process with more data on participants’ characteristics. In addition, wrapping and mobilizing comments were almost absent, appearing in only about 2.5% of the whole discussion. Further research could examine the impact on ideation of deliberately including and supporting these roles in online deliberations.

A third research direction has to do with the rules of engagement. Forum participants had no clear information about timeline, idea selection, and implementation. Indeed, participants did not know how long the discussion would last (e.g. one month? six months? two years?). Further research could examine how ideation patterns might differ if a more structured timeline is put in place.

A fourth direction relates to different patterns of participation by gender. This study recognized different participation levels by gender, and further analysis could be done to look at ideation processes and roles through a gender lens.

A fifth research suggestion is about comparing this online participatory process to the face-to-face one. We could not compare the difference between the ideas and the ideation processes in online and face-to-face settings. It is possible that the scattered appearance of the ideation stages in this study comes from the settings of online discussion that people do not gather at the same time and place. It would be interesting in a future study to do a comparison.

Finally, the process did not provide sufficient information to participants on whether their ideas were chosen or not, or on whether the most popular ideas would be implemented in the future. Having the project managers and the village planners involved in the process was positive, as it constituted one way to show participants that their voices were being heard by the municipal government. It is plausible to assume that when residents note that city managers are paying attention to their ideas they are more likely actively participate in discussions. Moreover, the city managers could share valuable information that residents may have limited access to. Further research could examine if knowing that suggestions are being acted upon affects participation and quality of ideation [40].

From the findings of our study three recommendations for government officials arise. First, we recommend that city officials who plan consultations, whether they be online or face-to-face, consider this ideation framework in the early stages of design. As we noted earlier in the paper, the phases of this framework do not have to appear sequentially. Having awareness of the different roles involved in the ideation process can allow officials to monitor the degree of participation, encourage participation in roles that seem to be absent, and facilitate the process when needed.

Second, we would like to emphasize the importance of the idea closure stage, where wrapping and mobilizing occur. This implies establishing clear timelines for the discussion and prompting participants to summarize the ideas, express their preferences, and suggest actions to move forward. Since the roles of wrapping and mobilizing are less likely to emerge naturally in unmoderated online invited spaces, it is advisable to incorporate these phases in the platform design or to assign a facilitator. The idea closure stage could also provide participants an opportunity to reflect on the process and make suggestions for future processes.

Third, public officials should be wary that although online processes open up the arena to more groups, some voices may still not be heard. Hence, we recommend that they gather demographic information of participants to find out which groups are well represented, and which groups are underrepresented (or absent) at the table, and develop online and offline strategies to include them.

Footnotes

Acknowledgments

This material is based upon work supported by the National Science Foundation under grant no. #VOSS-1322296.

References

Cornwall

. Introduction: New democratic spaces? The politics and dynamics of institutionalised participation. IDS Bull. 2004; 35(2): 1-10.

Zhang

Kenny

. Learning in an online distance education course: Experiences of three international students. Int Rev Res Open Distrib Learn. 2010; 11(1): 1-10.

Bonsón

Torres

Royo

Flores

. Local e-government 2.0: Social media and corporate transparency in municipalities. Gov Inf Q. 2012; 29(2): 123-32.

Chun

Cho

. E-participation and transparent policy decision making. Inf Polity. 2012; 17(2): 129-45.

Tolbert

Mosserberger

. The effects of e-government on trust and confidence in government. Public Adm Rev. 2006; 66(3): 354-69.

Welch

Hinnant

Moon

. Linking citizen satisfaction with e-government and trust in government. J Public Adm Res Theory. 2005; 15: 371-91.

Wong

Welch

. Does e-government promote accountability? A comparative analysis of website openness and government accountability. Governance. 2004; 17(2): 275-97.

West

. E-government and the transformation of service delivery and citizen attitudes. Public Adm Rev. 2004 Jan; 64(1): 15-27.

Hwang

Kim

Huh

. Seeing is believing: Effects of uncivil online debate on political polarization and expectations of deliberation. J Broadcast Electron Media. 2014; 58(4): 621-33.

10.

Tracy

. Challenges of ordinary democracy: A case study in deliberation and dissent. University Park: Pennsylvania State University Press. 2010.

11.

Jonson

. Design ideation: The conceptual sketch in the digital age. Des Stud. 2005 Nov; 26(6): 613-24.

12.

Wright

Street

. Democracy, deliberation and design: The case of online discussion forums. New Media & Soc. 2007; 9(5): 849-69.

13.

Pieper

. Political participation via social media: A case study of deliberative quality in the public online budgeting process of Frankfurt/Main, Germany 2013. Univers Access Inf Soc. 2015; 14(4): 487-503.

14.

Coleman

Gøtze

. Two-way governance. Bowling together: Online public engagement in policy deliberation. London, UK: Hansard Society. 2001; 13-20.

15.

Coleman

Gøtze

. Bowling together: Online public engagement in policy deliberation. London: Hansard Society. 2001.

16.

Ostwald

. Virtual urban futures. The cybercultures reader. New York: Routledge. 2000; 658-75.

17.

Brabham

. Crowdsourcing the public participation process for planning projects. Plan Theory. 2009; 8(3): 242-62.

18.

Hong

. Citizen participation in budgeting: A trade-off between knowledge and inclusiveness? Public Adm Rev. 2015; 75(4): 572-82.

19.

Cohen

. Deliberation and democratic legitimacy. The good polity: Normative analysis of the state. Hamlin

Pettit

, editors. Oxford: Blackwell. 1989; 17-34.

20.

Baek

Wojcieszak

Delli Carpini

. Online versus face-to-face deliberation: Who? Why? What? With what effects? New Media Soc. 2012 May; 14(3): 363-83.

21.

Dahlberg

. Computer-mediated communication and the public sphere: A critical analysis. J Comput Commun. 2001; 7(1).

22.

Rojo

Ragsdale

. Participation in electronic forums: Implications for the design and implementation of collaborative distributed multimedia. Telemat Informatics. 1997 Feb; 14(1): 83-96.

23.

Swartz

. Who writes Wikipedia? [Internet] 2002 [cited 2017 Apr 20]. Available from: http://www.aaronsw.com/2002/whowriteswikipedia/.

24.

Liu

. Exploring online engagement in public policy consultation: The crowd or the few? Aust J Public Adm. 2017; 76(1): 33-47.

25.

Adamczyk

Bullinger

Moeslein

. Commenting for new ideas: Insights from an open innovation platform. Int J Technol Intell Plan. 2011; 7(3): 232-49.

26.

Lührs

Albrecht

Lübcke

Hohberg

. How to grow? Online consultation about growth in the city of Hamburg: Methods, techniques, success factors. Lecture notes in computer science. Prague, Czech Republic: Springer. 2003; 79-84.

27.

Peixoto

. E-participatory budgeting: E-democracy from theory to success? 2008.

28.

Irvin

Stansbury

. Citizen participation in decision making: Is it worth the effort? Public Adm Rev [Internet]. 2004; 64(1): 55-65.

29.

Allegretti

Herzberg

. Participatory budgets in Europe: Between efficiency and growing local democracy. Amsterdam: Transnational Institute and the Centre for Democratic Policy-Making. 2004.

30.

Mook

Schugurensky

. Concurrent or integrated hybridity? Exploring offline and online citizen participation in invited spaces. Int J Organ Theory Behav. 2016; 19(4): 514-34.

31.

Graham

Bachman

. Ideation: The birth and death of ideas. Hoboken, NJ: Wiley. 2004; 240.

32.

Franke

Shah

. How communities support innovative activities: An exploration of assistance and sharing among end-users. Res Policy. 2003 Jan; 32(1): 157-78.

33.

West

Lakhani

. Getting clear about communities in open innovation. Ind Innov. 2008 Apr; 15(2): 223-31.

34.

Kuratko

Goldsby

Hornsby

. Innovation acceleration: Transforming organizational thinking. Pearson. 2011.

35.

Aitamurto

Landemore

. Democratic participation and deliberation in crowdsourced legislative processes: The case of the law on off-road traffic in finland. Proceedings of the 6th Conference on Communities and Technologies (C&T) Workshop: Large-Scale Idea Management and Deliberation Systems. Munich, Germany. 2013.

36.

Frey

Lüthje

. Antecedents and consequences of interaction quality in virtual end-user communities. Creat Innov Manag [Internet]. 2011; 20(1): 22-35.

37.

Barber

. Which technology and which democracy? Democracy and new media. Jenkins

Thorburn

, editors. Cambridge, Massachusetts: MIT Press. 2003; 33-47.

38.

Murray

Wing

Woolf

Wise

Clark

Osterweil

. A prototype facilitators dashboard: Assessing and visualizing dialogue quality in online deliberations for education and work. Proc 2013 Int Conf e-Learning, e-Business, Enterp Inf Syst e-Government. 2013.

39.

Hayes

Krippendorff

. Answering the call for a standard reliability measure for coding data. Commun Methods Meas. 2007; 1(1): 77-89.

40.

Garcia

ACB

Vivacqua

Tavares

. Enabling crowd participation in governmental decision-making. J Univers Comput Sci. 2011; 17(14): 1931-50.