1. Introduction
The World Café method was first invented by Brown and Isaacs (Reference Brown and Isaacs1995). It is considered an explorative research method (Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022), including elements of participatory (Reference Löhr, Weinhardt and SieberLöhr et al., 2020; Reference Andrew, Warne, Fallon and MoranMcAndrew et al., 2011; Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022) and qualitative research (Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022). The method is beneficial when it comes to developing a topic with the wisdom of the crowd (Reference Brown and IsaacsBrown & Isaacs, 2005). The interactive group dynamic created by the World Café method enables researchers to quickly collect relevant data in large and small groups (Reference Brown and IsaacsBrown & Isaacs, 2005; Reference Gibbs, Press, Wong and CummingGibbs et al., 2020; Reference Löhr, Weinhardt and SieberLöhr et al., 2020; Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022). By bringing together the participants' different perspectives and points of view, new ideas can arise (Reference Brown and IsaacsBrown & Isaacs, 2005). Participants come together in an artificial café (Reference Brown and IsaacsBrown & Isaacs, 2005). After the participants are welcomed by a facilitator (Reference MacFarlane, Galvin, O'Sullivan, Inerney, Meagher, Burke and LeMasterMacFarlane et al., 2017; Reference Ritch and BrennanRitch & Brennan, 2010), they group around different tables to create lively discussions (Reference Brown and IsaacsBrown & Isaacs, 2005). A question will be discussed at each table in several discussion rounds (Reference Brown and IsaacsBrown & Isaacs, 2005). The shared thoughts are written on paper tablecloths (Reference Brown and IsaacsBrown & Isaacs, 2005). After each discussion round, participants change tables randomly so that the wisdom of the crowd can spread (Reference Brown and IsaacsBrown & Isaacs, 2005; Reference Fouché and LightFouché & Light, 2010; Reference Löhr, Weinhardt and SieberLöhr et al., 2020). The discussions are accompanied by a table host (in the following host) per table (Reference Brown and IsaacsBrown & Isaacs, 2005). The behavior of the hosts significantly influences the success of the World Café method and, thus, the quality of the collected data (Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022). The hosts are responsible for creating an atmosphere where participants share their thoughts (Reference Brown and IsaacsBrown & Isaacs, 2005). The overall tasks of the hosts are welcoming the participants, summarizing the previous round, moderating the discussions, and being responsible for note-taking (Reference Brown and IsaacsBrown & Isaacs, 2005).
At first glance, the World Café method looks like a method that is easy to use. However, it is complex and demanding when collecting high-quality data. To guide researchers, Brown and Isaacs (Reference Brown and Isaacs2005) invented seven principles: 1) setting the context, 2) creating hospitable space, 3) exploring questions that matter, 4) encouraging contribution from everyone involved, 5) connecting diverse perspectives, 6) listening together for patterns and insights and 7) sharing discoveries. However, these principles are relatively abstract and lack a structured process-oriented approach. The current literature focuses more on the concept of the World Café method and less on the scientific process. Additionally, research lacks guidance on what to focus on regarding the role of the hosts. This leads to two design questions: 1) How to conduct the World Café generally? and 2) Which activities linked to the table hosts improve the quality of the data collected using the World Café method? For the first question, we designed a structural approach as an artifact of a Three-Phase Blueprint of the World Café method. The artifact shows which phases and sub-steps are needed to collect high-quality data. For the second question, we developed a selection and a three-step instruction plan for hosts. Since especially the instruction plan is a clear intervention in the conduction of the World Café method, this results in the research question of whether or not this intervention is effective. Therefore, we tested the intervention in a practical context. In the next section, we discuss the design and research methods that were applied. In section three, the results of our design exercise can be found: the Three-Phase Blueprint and the selection and instruction plan. In section four, we present the analysis of the test results of the intervention. We finish with a discussion and conclusion. This paper aims to offer researchers a structured approach for conducting a World Café with the help of the Three-Phase Blueprint. At the same time, by testing the role of the hosts, it becomes clear that detailed instructions on their role are necessary to improve the quality of the data conducted. This paper contributes theoretically by supporting the development of the World Café method in different research fields, such as explorative, participatory, and qualitative research, but simultaneously to design science.
2. Method and research design
For designing a structural approach and testing the role of the hosts, this paper encompasses three phases: 1) the design of a process model of a process framework to get clarity on how to conduct the World Café method for scientific purposes, 2) the design and testing of the artifact regarding the role of the hosts to improve the quality of the research output and 3) the evaluation regarding the tested host part in the designed artifact. To design the process model for the World Café in general and the activities linked to the hosts, we carried out a design sprint. As part of the design sprint, we compared the personal experiences of the design team with the World Café method with those of other researchers in the literature. Furthermore, brainstorming and the usage of analogies were applied to shape the design of the framework. The experiences, the collected literature, and the brainstormed approaches of frameworks were finally brought together to create an artifact of the Three-Phase Blueprint. We used action research to test the instructions for the hosts, which are a specific part of the Three-Phase Blueprint. Action research enables one to reflect on the artifact with users while simultaneously creating a direct interaction between academic orientation and practical implementation. (Reference Baum, MacDougall and SmithBaum et al., 2006; Reference Sein, Henfridsson, Purao, Rossi and LindgrenSein et al., 2011). The practical test environment was a German consultancy agency. At this agency, we conducted two World Cafés under comparable conditions in a time spread of 1.5 years. In the first World Café, the instruction was based on the already existing seven core principles according to Brown and Isaacs (Reference Brown and Isaacs2005). In the second World Café, the instruction was based on the designed Three-Phase Blueprint. A general comparison, especially regarding the hosts' preparation, can be seen in Table 1. The research design of testing the artifact was divided into three steps and is visualized in Table 2. We conducted two rounds of interviews with the users of the hosts' instructions and combined that data with field observations of the behavior of the hosts during the World Café method. The first interview round was before the execution of the second World Café to see what the hosts remembered from the first World Café session, where the instructions were based on the seven core principles. The second round of interviews was aimed to test the effectiveness of the newly designed instructions. Content analysis was used to evaluate both data sources, the semi-structured interviews, and the field observations. Brown (Reference Brown2019) three layers of empathy, a cognitive layer, an emotional layer, and a functional layer served as a structural basis for gaining a holistic overview of the analyzed data conducted by the artifact test. The cognitive layer reflects the extent to which the content of the preparation for the hosts' role has been understood, internalized, and implemented. This could develop an understanding of the latent needs of the hosts. The emotional layer reflects how the hosts felt about the instruction. This helps to understand the emotional needs of the hosts. The functional layer relates to the actual behavior and actions of the hosts. The functional layer uncovers the practicability and implementation of the host role, particularly the feasibility of the specified tasks and responsibilities (Reference BrownBrown, 2019).
Table 1. Comparison of the table hosts' role in World Café 1 and 2
Table 2. Research design combining semi-structured interviews and field observations
3. Designing a structural approach to run the World Café method
3.1. Designing an artifact of the Three-Phase Blueprint of the World Café method
To get more clarity on the process and structure of the academic conduct of the World Café method, we designed a Three-Phase Blueprint. The result can be found in this section. Since the role of the hosts is crucial for the quality of the data of the World Café method, we designed a selection and instruction plan for the hosts, which can be found in Section 3.2.
The artifact of the Three-Phase Blueprint includes the existing principles, different roles, and procedures of the World Café method, but it brings them together in a new structure. By comparing our experiences with known principles, roles, and procedures of a World Café, it became clear that the existing understanding of the World Café method is construct-driven. However, a process-related approach is needed to collect high-quality data successfully. It is as if the ingredients for a recipe are there, but there is no step-by-step recipe on how to use them to cook the best possible dish. Using the cooking analogy helped to find a three-phase structure for giving the World Café method a new shape - preparation of the ingredients, the cooking itself, and the cooked dish as a result. The World Café method can also be divided into three phases to gain high-quality data: planning, execution, and evaluation. The foundation of the success of a World Café is laid in the planning phase. All elements and roles, such as questions, the artificial café, and roles, such as table hosts, participants, facilitator, and presenter, get prepared. During the execution phase, the actual World Café will have all its discussion rounds; thus, the data collection will occur. The evaluation phase begins when the data collection is completed. The collected data gets evaluated, analyzed, and prepared for presentation. Figure 1 shows the artifact of the Three-Phase Blueprint, including its main phases, as well as the sub-steps researchers need to know when using the World Café method. The Three-Phase Blueprint is a structural process-driven approach representing the entire research design.
Figure 1. The Three-Phase Blueprint of the World Café method
3.2. Setting selection criteria and instruction procedures for the table hosts
The hosts play a crucial role in the World Café method because they are responsible for creating an atmosphere of collaborative nature and transforming the conversations into high-quality data. Some scholars have discussed the selection of hosts (Reference Broom, Brady, Kecskes and KildeaBroom et al., 2013; Reference Chang and ChenChang & Chen, 2015; Pumpe & Vallée, 2017; Schiele et al.,Reference Schiele, Krummaker, Hoffmann and Kowalski2022) and the preparation and training of hosts (Reference Fallon and BeanFallon & Bean, 2014; Reference Pumpe and ValléePumpe & Vallée, 2017; Reference Ritch and BrennanRitch & Brennan, 2010; Reference Schiele, Krummaker, Hoffmann and KowalskiSchiele et al., 2022). However, there is a lack of structure and clarity in precisely selecting and instructing the table hosts. Thus, concrete approaches were designed to select and instruct the hosts during the planning phase of the Three-Phase Blueprint. Special attention is needed in 'Acquiring & instructing table hosts' because it lays the foundation of the hosts' behavior regarding collecting high-quality data in the execution phase. The selection and instruction plan encompasses three aspects: 1) selecting the hosts based on the required skillset, 2) including the hosts in the research team, and 3) preparing and instructing the hosts. The hosts have to fulfill various tasks. Thus, a comprehensive skillset is needed. We specified and classified the skills needed into three groups. First, hospitality skills. The hosts are responsible for welcoming all participants and making them feel comfortable during the discussion rounds. Second, moderating skills. The hosts have to lead neutral discussion rounds. The host's opinion is not requested, but the participants' opinions are. Third, abstraction skills. The hosts are responsible for getting the participants to talk and keeping the discussion on the topic. After selecting the hosts, they need to be included in the research team. This is significant because, as Schiele et al. (Reference Schiele, Krummaker, Hoffmann and Kowalski2022) already recommended, the hosts have internalized the relevance and main features of the research assignment if the hosts are part of the research team. Once the hosts have been integrated into the research team, they need to get prepared for their role. Detailed instructions are required since the hosts' role directly influences the conducted data. We designed a detailed three-step instruction to prepare hosts for their role. These three steps are: 1) sharing the research aim, 2) communicating the agenda and planning, and 3) instructions about the role and tasks of the hosts. The first step is to communicate the research aim. A written and oral briefing about the research status and the World Café process helps. The second step is about sharing what happens at the World Café. These can be, for example, by handing out a graphic of the process of the World Café method and handing out a guide with helping questions in case the hosts get stuck keeping the discussion rounds alive. The third step is about explaining the role and tasks of the hosts. It is important that the hosts know the expectations they have to fulfill. The hosts are responsible for making the participants feel comfortable and have to create a safe environment. A warm welcome is the first step to go. For the participants to understand the table topic, the hosts explain the question to be discussed and give the participants a summary of the results from round two onwards. The hosts have to keep the discussion alive. This means that the conversation sticks to the actual topic. The hosts must ensure that the discussion is based on cooperation and that individual opinions do not negatively influence or dominate the discussion. Everyone has their say and can express their ideas and opinions. The hosts are independent and exclusively responsible for moderating the discussion. If the discussion stops or progresses slowly, the hosts can stimulate the exchange again by asking specific questions or following up on what has already been said. The hosts can address participants directly or ask questions to the group. It is also possible to use creative discussion techniques such as the 5W (5whys). The host's task is to collect the data entirely. Generally, tablecloths, sticky notes, or moderation cards can be used to write the ideas down. Equipping the hosts with sub-questions to the actual question helps the hosts to discuss further details with the participants and cover the whole range of the topics to be discussed.
4. Results of the testing phase
4.1. The role of the hosts from a cognitive perspective
Regarding the three-step instruction designed to optimize the hosts' role, we tested this artifact's effectiveness. We analyzed the interview and observation data according to the three layers of empathy (Reference BrownBrown, 2019) and show the results subsequently.
The cognitive perspective results in three findings: (1) Clarification of expectation (what), (2) execution of task fulfillment (how), and (3) creation of data awareness (why). At first, the clarification of expectation. This means that hosts know 'what to do'. During the interviews, the hosts were able to explain what was expected of them as hosts. The hosts were aware that they were responsible for creating a hospitable atmosphere in which the participants would feel comfortable and want to contribute to the World Café “(…) first of all, welcoming all warmly, wishing everyone a nice day and expressing that it is nice to have everyone here (…)”. The hosts were able to additionally describe in the interviews that they were expected to play a neutral role “(…) not express our own opinion, that it is about letting the group speak”, to not influence the participants in their discussion and opinion-forming. It is the host's responsibility to moderate the discussion groups. This includes actively listening to the participants, motivating all participants, including those who were quieter, to participate in the discussion, and not evaluating the discussion results or mixing them with their own opinions. In the interviews, it also became clear that the hosts were aware of their execution of task fulfillment to meet the hosts' expectations. Thus, the hosts knew 'how to do' their job. The hosts knew that to moderate the discussion groups and to involve all participants, it is helpful to include the quieter participants by addressing them directly and to slow down the more active participants at times gently “So, I also asked the one or the other more quieter participant directly, but there were not that many (…) I picked one or the other of them up like do you have an opinion about it, or what do you think about it, or how do you see it?” To understand what the participants wanted to say, or sometimes also to better understand what the participants' arguments were, supportive questions should be asked “(…) that we should ask supportive questions (…)”. Giving examples and impulses helps to keep the discussion going if it should falter “(…) if no discussion ensures, that supportive questions will be used to get the discussion started or keep it going.” Being aware of 'what to do' and 'how to do' leads to the result of 'why to do' the host's job. The table hosts could express during the interviews that they had to take a neutral role and give impulses and examples for a lively discussion because the hosts should bring everyone to talk. Only if all participants got involved and were not judged by their thoughts could the data that should be collected. “Otherwise, the whole thing does not make sense in the end if, for example, I end up writing down my own opinion.” This shows that the hosts created the data awareness needed to conduct a World Café successfully.
4.2. The role of the hosts from an emotional perspective
Emotionally, two main results were noted: (1) Feeling that needs have been met and (2) Feeling of behavioral security. The interviews clarified that the hosts' preparation for the World Café met their needs. The table hosts described the preparation as “Good”, “(...) in any case, it helped me.” or “(…) that was perfectly fine with me.” This assessment was justified because the instruction was well prepared from the hosts' point of view “Short, so to say concise, it was well prepared (…)”. Part of the successful preparation was that different methodologies were used to address different types of learning. It was found helpful to present the key facts on a whiteboard, to distribute cheating paper with suggestions for additional questions for the World Café and to discuss the role, tasks, and supporting measures for the implementation of the World Café “Talking about it beforehand was helpful, the most helpful thing, somehow.” The interviews also showed that the introduction to the hosts had created a sense of behavioral security. Due to the well-prepared instructions and the resulting concise introduction of the hosts, all the hosts' fundamental questions were answered directly ”(…) The main points were written down, what the tasks are, what the supporting questions could be, and thus all questions were answered.” All parts of the introduction were equally important and helpful in understanding their role and tasks ”(…) so when I was introduced to it, I thought everything was equally important.” It was beneficial for the hosts to learn that it is not about the hosts' opinion, but about that of the participants. The hosts should consciously take on a neutral role. ”So what has stuck is that it's not about me and my opinion, and I think that helps a lot, so do not sit there and push your opinion through. Precisely because you are a manager, you should encourage your employees to say something, to express their opinions, and to hold back. I believe that this was very helpful and very important for a successful World Café, and to say it so succinctly, well, at least I took it on board succinctly.” For one host, the sense of security could have been increased if the cheating paper had not been handed out on World Café day but during the instruction meeting. This would have allowed the host to add the notes made during the instruction meeting directly to the cheating paper ”Yes, I do not know, maybe just to get a handout right away instead of just telling it. So, I then wrote things down, I then write things down, so if I had, you said, I think, that we would get something. But I always have to see and write things down, then I can remember them better and prepare myself better.” The fulfillment of the hosts' needs in executing the instruction and the feeling of security in the execution lead to the hosts having a sense of self-esteem when conducting the World Café.
4.3. The role of the hosts from a functional perspective
The instruction of the hosts was functional because during conducting the World Café, the hosts were (1) responsible for their role execution, (2) created a lively discussion, and (3) dedicated to the data documentation. While observing the hosts in the second World Café, it became clear that the hosts had fulfilled their role execution responsibly. The hosts focused on their task fulfillment, especially by moderating the rounds and taking on a neutral role. Each round, the hosts welcomed the new participants arriving at their tables. As the hosts were more aware of their tasks, they presented the results of the previous round briefly and concisely to the new participants at the table. In general, the hosts summarized the key facts in one to three minutes and had no trouble summarizing them. The more discussion rounds were held, the longer it took the hosts to summarize the previous rounds. Nevertheless, there was enough time for a lively discussion, even in the last round. The short but concise summary of the previous rounds had the advantage that the participants had more time for the exchange and discussion rounds than in the first World Café. Taking a neutral position was not always easy for every host, especially for the host from table number two. The host was urged to comment on the discussion and give more background information than needed. The other hosts took a neutral role and did not influence the participants' thoughts. It was observed that all hosts managed to create a lively discussion. All hosts motivated the participants to engage and exchange ideas. The hosts asked questions to have the content discussed explained in more detail or to keep the discussions alive. Quieter participants were encouraged to take part. Very talkative participants were sometimes politely interrupted. The hosts reminded the group that discussing off-topic was dangerous and steered the discussion back to the actual topic. It was observed that the hosts knew that the data documentation was important and were dedicated to it. Three of the four hosts took their task of documenting the results of the discussion very seriously. Only the hosts were allowed and were fully responsible for documenting the discussions. Each host had their own documentation style, ranging from dividing the tablecloths into sections, bracketing important topics, organizing notes by keywords, and only making side notes. To create more structure, each host was asked to fill out a summary sheet after the World Café with the following questions: 1) Briefly summarise the key findings of the discussion at your table, 2) How did the participants at your table behave?, 3) What challenges did you face as hosts? and 4) What worked well for you as hosts?. Answering the summary questions helped the hosts to reflect on the World Café and identify the key facts.
5. Discussion and conclusion
The Three-Phase Blueprint's structured approach helps to collect high-quality data while focusing on the data collection and setting up a whole research design. The hosts play a crucial role in collecting high-quality data through their tasks and responsibilities. It is important that the selection and preparation of the hosts be approached in a structured manner. Therefore, we designed the Three-Phase Blueprint, with special attention to the selection and instruction of the hosts. Using Tim Brown (Reference Brown2019) three layers of empathy as a structure for the evaluated data, the following results were obtained: Cognitively, the hosts know what to do because the expectations for them were clarified, and they could explain what was expected of them to fulfill their role. Due to created data awareness, the hosts also knew why their role is important. From an emotional point of view, it was found that the introduction to the role of the hosts met their needs and contributed to their behavioral security while conducting the World Café. Not only did the hosts know what to do, how to do it, and why their role was important, but they were also able to act effectively. Which leads to functional results. The hosts executed their role responsibly, created lively discussions, and documented the data dedicatedly. Thus, introducing the hosts to their role and tasks presented by the Three-Phase Blueprint positively influences the quality of the collected data because the hosts are more self-confident in carrying out their roles and tasks. Table 3 summarizes the key Three-Phase Blueprint artifact elements that positively influenced the quality of the collected data during the World Café if the hosts fulfilled their role correctly. The distributed cheating papers allowed the hosts to focus on their tasks. The hosts know they can look at their cheating paper anytime during a disruption. Imagining and discussing critical situations also enhances the hosts' safety. The fact that the table host welcomes participants with their hospitality skills creates a feel-good atmosphere. As a result, participants feel ready to share their ideas and thoughts. Supportive follow-up questions, motivating questions for quieter participants, and friendly interruptions of very talkative participants lead to a situation in which the participants feel ready to share their ideas and thoughts. The discussion is not colored by a few participants. Steering the conversation to the question that matters ensures that the data collected fits the overall research. The fact that the hosts briefly summarise the key facts of each round and that the World Café consists of several rounds enriches the data quality iteratively from different perspectives and ensures that the data already collected is constantly reviewed.
Even if the Three-Phase Blueprint helped to increase the quality of the collected data, using the artifact test, a number of contraindications were identified. These contraindications show opportunities for further expansion and improvement of the existing artifact. Even though the hosts found it reassuring to have a cheating paper for emergencies, they did not use it. One of the reasons for this was that the cheating papers were not available early enough for the hosts on the day of the World Café. It was more complicated for the hosts to look for the cheating paper during the World Café rounds than to come up with supportive questions themselves. According to one host, it would also have helped if the cheating paper had been made available digitally to all hosts during the instruction meeting. One host had difficulties documenting the results. The conclusion is that the hosts should be more emphatically instructed on the importance of documenting the data. Additional training should be provided. The additional documentation training at the instruction meeting would also have helped the other hosts record the discussed contents more structured and detailedly. The exact documentation of the results is important so that all the topics discussed can be evaluated after the World Café and no data is lost. The main aspect of contraindications is taking a neutral role as table hosts. The hosts mentioned that taking on a neutral role was incredibly challenging. Especially one host struggled with taking a neutral role. If a host is not neutral in the role, the participants are no longer encouraged and supported in expressing their opinions but influenced and directed. This is critical in that the participants may no longer contribute the content they would like to the discussion. Therefore, it would be helpful in the future if hosts already had more moderation experience when hosting or at least completed moderation training before the World Café.
There are also indications that suggest that the artifact of the Three-Phase Blueprint leads to a collection of higher-quality data. The instruction of the hosts meant that they were well prepared when the World Café began. In particular, the balance between theoretical and practical support helped the hosts understand and fulfill their role and tasks. The type of instruction contributed to this. All the important information was presented to the hosts in an instruction meeting. The main points were summarised in written and oral instruction the day before the World Café. During the instruction meeting, there was time and space to clarify questions from the table hosts and to address the hosts' individual needs. In addition, critical situations that regularly arose during a World Café were simulated using imagination. The hosts discussed together why the situation is to be assessed as critical and what possible reactions and recommendations for action could be. Because the instruction meeting took place the day before the World Café, the hosts still had the relevant information present. Being part of the research team increased the understanding of the World Café method. Thus, the examples listed in the Three-Phase Blueprint for preparing the role of the hosts create security for researchers and, at the same time, leave enough room for individual adjustments. The hosts felt better prepared than in comparison to the first World Café. The role of the hosts and the associated tasks and responsibilities were better internalized and understood by the hosts. As a result, executing the role as a host went more smoothly than during the first World Café. It can, therefore, be concluded that the artifact of the Three-Phase Blueprint led to the World Café being conducted more successfully in the second World Café than in the first.
The limitations of this paper also indicate approaches for future research. This research is designed as a design paper that is usually context-specific. This allows researchers to test and apply the Three-Phase Blueprint in industries other than the consulting industry. The data generated during the test phase was analyzed qualitatively. This provides scope for future research to also integrate quantitative elements in the evaluation of the test. For a further perspective on the tested three-step host instruction, the participants could be asked for feedback on the performance of the hosts.
Research has become more of a team game. This applies to the increasing popularity of participatory research and research collaboration. The more important participatory research becomes in research, the more important it is to ensure that the data collected is high-quality data. Empowering the hosts as co-researchers is an efficient way to do research. In particular, participatory research is becoming more important because of the support of co-researchers. Without co-researchers, it will be challenging in the future to collect data from large groups of people in a time-efficient and low-cost way. At the same time, involving co-researchers such as hosts for research also helps to close the scientific-practice gap.
Table 3. The impact of the hosts' activities on data collection and data quality
Acknowledgments
During the preparation of this work, the authors used DeepL and Grammarly to improve the readability and language of the manuscript. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Funding statements
This work was supported by national funds through FCT (Fundação para a Ciência e a Tecnologia), under the project UIDB/04152/2020 (DOI: 10.54499/UIDB/04152/2020) - Centro de Investigação em Gestão de Informação (MagIC)/NOVA IMS.
Open access
