Facilitation at Astro Hack Week

Astro Hack Week intentionally spreads itself very wide in terms of the participants we admit, and the projects we encourage participants to work on. For example, we have both early undergraduate students all the way to faculty members and senior industry experts, and the differential in computational and data analysis skills can be quite wide (though they don't have to be: Astro Hack Week has seen undergraduates teach senior faculty about things like databases!). In this context, Astro Hack Week operates on the principle that every participant has valuable skills to contribute, and that their unique life experience make them a valued member of our community.

It's also worth nothing that participants come to Astro Hack Week with very different goals: some come primarily to learn, some come to make progress on a specific (PhD) research project, some come to collaborate with specific people, and others want to learn more about how to run a hackathon. The breadth of indv

Perhaps the key challenge for us as organizers and facilitators is helping each participant find and contribute to a project that they are both interested in and that matches up with their goals for learning and participation. In addition, not all participants feel equally comfortable participating, and impostor phenomenon is a common occurrence at Astro Hack Week. The goal of facilitation at Astro Hack Week is to provide an environment where all participants feel welcome, accepted and their contributions are valued. With this in mind, each year there is at least one person whose primary task is facilitation. This includes setting the tone of the meeting and framing the community building in the larger context of the goals of the workshop, leading the group through activities such as ice breakers and hack pitches, and going around the room during project time to make sure the teams are functioning well and all participants are able to participate to the level that they wish to. Usually, the person facilitating does not work on a hack themselves (though in 2019, the facilitation was shared between three organizers to allow hack time for each person).

The first day is particularly important in the context of facilitating the event. It sets the tone for the rest of the week, and many participants will meet each other for the first time ever. This can be particularly intimidating to junior participants, and so making sure that everyone has the chance to meet others is a core goal. In addition, the design and pitching of projects, as well as the team formation happens on the first afternoon. This, too, requires careful facilitation to ensure the teams are off to a welcoming, healthy and productive start.

As organizers, it is very difficult to know what's in the participants' minds, and the participants' lived experience might lead them to react to certain activities very differently than we as organizers expect. We have a number of feedback mechanisms that we have implemented to receive feedback both during the week (which may allow us to make changes immediately) and at the end of the workshop (for implementation during a future event). In addition, it is worth noting that all group activities are framed as invitations: we do not insist that participants take part in events that require social interaction, which would be actively hostile to our participants and also counter-productive. We rather invite participants to take part based on how well they feel able to do so. One notable exception to the rule is the Code of Conduct: we do require all participants to adhere to the Astro Hack Week Code of Conduct, and ask participants to confirm their willingness to do so upon applying to the workshop.

With the above considerations in mind, we have implemented some activities to guide participants through Astro Hack Week. These are all based on the following ground rules that we as organizers and facilitators strive to follow and implement during the week:

• The code of conduct sets the ground rules and minimum standard for all interactions
• Remind participants repeatedly to think about
  • their interactions,
  • the dynamics within their interactions and groups
  • the space they occupy (relative to others)
  • the distribution of talking time/space to participate
• Also remind participants to pay particular attention to known dynamics (e.g. with respect to academic seniority, gender, race, …)
• Do not interrupt speakers (unless they won’t shut up)
• Encourage questions and give others the benefit of the doubt (don’t tell people “I can’t believe you don’t know this!” or something of the sort)
• Encourage participants to engage critically with and think about how they interact with others

Story Board

Add story board here!

Detailed Descriptions of Activities

Community Building: Impromptu Networking

• Description: Impromptu Networking
• When: Very early on, before the first coffee break (as needed potentially every morning)
• Duration: ~15-20 minutes
• Why: Provide a non-threatening environment for first interactions between participants, reduce barriers to interactions, gives them a chance to introduce themselves to others and get to know them in a short while. The short amount of time for each interaction reduces the chance for participants to “panic” about making a good first impression. Also aims to set the tone for the week and get participants to think about their own goals and obstacles for the week. The responses have to be fast, so they’re designed to be stream-of-consciousness and somewhat unpolished.

• What: Invite everyone to stand up and spread through the room. Ask them to find someone they haven’t met before. They have 120 seconds to introduce themselves to each other and complete the sentence displayed on the slide (see below for examples). After 120 seconds, ring bell and get participants to switch to a new partner, with a new sentence. Repeat ~7-8 times. Remind participants that 120 seconds is short, and that they’re responsible for giving the other person enough time for their answer.

• Materials: Bell, slides with sentences to be completed.

• Sample Statements:

  • What first inspired me in this work is …
  • One thing I’d like to learn this week is …
  • A challenge I hope we make progress on is …
  • I belong at Astro Hack Week because …
  • Something I am proud of is …
  • Something I plan to give to this community is ...
  • My hope is that people will experience me as …
  • Astro Hack Week will be a success for me if …
  • For fun, I …
  • In my free time, I like to ...
  • When it’s all said and done, all I want is …

• Additional statements, if the exercise is repeated in the mornings throughout the week:

  • Yesterday, I learned that …
  • In my interactions with others, I particularly enjoyed ...
  • Today, I’d like to try to ...

• Observations and Feedback: Participants took part in this at Astro Hack Week fairly enthusiastically. Not all used the prompts, some just introduced themselves and chatted about what they planned to do this week, and that's perfectly fine, too. There were two important observations and feedback that are relevant for future uses of this activity:

  • Some of the statements might be triggering for participants with social anxiety and mental health issues (e.g. in the first version of this activity one of the prompts read "A voice I need to stop listening to is ..."). We've since tried to eliminate those statements, and remain attentive to feedback regarding prompts that might make our participants uncomfortable.
  • One issue with this activity is that 50 people all talking to one another at the same time can get very loud! Ideally, do this activity outside if weather permits, or in a space that allows participants to spread out and that doesn't echo too much.
  • Something that we haven't solved very well so far is how this exercise might work for participants with mobility issues. Because it requires standing and moving about, this might unintentionally exclude some participants. There could be a variant where this exercise involves paired chairs, and only some participants move about. The invitation would be for roughly half of the participants to stand up and find a partner, while the remainder stays seated. In either case, however, it's useful to get participants to move away from the tables they might normally be sitting at and away from their computers, if possible.

Introducing and Framing Astro Hack Week

• When: At the beginning of the week, right after the Impromptu Networking
• Duration: About 30 minutes
• Why: Many participants come in with just a vague idea of what Astro Hack Week is and what its goals are. This introduction aims to give participants an overview of the workshop, the practical logistics (schedule, bathrooms, lunches, social events etc.), and especially the rules and expectations for working together and for building an inclusive learning environment.
• What: About a half-hour presentation by the organizers, covering the elements and structure of the workshop, logistics, the code of conduct and other aspects of working together for the week. At Astro Hack Week, it has been important to us not only to reinforce that the code of conduct is to be followed at all times (both during the workshop itself and social times outside of it), but actively goes into detail how we want participants to behave and interact with one another. This includes welcoming questions and providing useful answers, experimenting with Astro Hack Week itself, trying out something new, listening to each other, respecting each other's pronouns, being kind to one another, and being open to learn about one another, both in terms of technical knowledge as well as from other people's lived experiences. This discussion also includes a note on impostor phenomenon, which we've found to be prevalent at Astro Hack Week.
• Material: Slides (an example of our slides from Astro Hack Week 2018 is available here)

Building Positive Learning Environments Through a Data Visualization Tutorial

During Astro Hack Week 2018 and 2019, the first tutorial of the week was a data visualization tutorial. Because data visualization is one of the less technically complicated topics, and because it allows for many opportunities to practice group interactions, we chose it as a vehicle to also model and practice inclusive group work with our participants. We guided our participants through a series of group activities that allowed them to self-discover important data visualization principles and put them into practice, while also providing them with structures to help them with the project work later.

Community Building: 1-2-4 All for Data Visualization

• Description: 1-2-4-all
• When: at the start of the data visualization tutorial
• Duration: 20 minutes (silent reflection: 2 minutes, pairs: 4 minutes, groups of 4: 6 minutes, whole group: 8 minutes; see also feedback + observations below)
• Why: Best practices and rules are most easily accepted when they are grounded in the students’ own experiences. Especially with data visualization, more than other fields, there is likely a lot of expertise scattered through the room already, and most participants, even students, are likely to have an intuitive sense of whether a particular visualization is effective at communicating information. This exercise is aimed at co-creating best practices for effective data visualization. The secondary goal is to give participants a structure and a framework for group interactions and working in teams, which they may adopt for the free-form hacking later during the day/week, and reinforce good practices for communicating and cooperating.
• What: Invite participants to sit down next to someone they don’t know. Each participant gets one data visualization to look at: some sample visualizations are fairly competent at communicating information, some are not. Each participant is then asked to answer three questions (see below) in silent reflection (2 minutes). In the next step, participants are grouped in pairs, and discuss their answers, co-creating a common set of best practices (4 minutes). In the third step, groups of four discuss the common set of best practices and (6 minutes) rank them by importance. In groups of four, a pen or crayon will serve as a talking object. Only the person with the object speaks, and each person mentions one rule, then passes on the object. One person has to be a note-taker (for the shared online document). Participants should invite the most junior person to speak first, but they’re free to say “pass” and let someone else begin. Finally, each team shares a practice with the entire group, which a note-taker records in the shared online document. Once each team has had a chance to speak, if there are further best practices, we go through the groups again and allow each group to mention another best practice.
• Material: pens/crayons, print-outs of data visualizations, set-up of a shared online document

• Questions:

  • What information does this data visualization convey
  • What does this data visualization do well in communicating information?
  • Can you think of ways in which you would improve this data visualization?
  • What general rules can you derive for communicating information visually?

• Observations and Feedback: 20 minutes is way too short for this exercise, partly because participants require more time to take in the visualization and answer the questions, but also because they are excited to share their conclusions both in pairs and groups of four. For the second iteration of this exercise, we restricted ourselves to just two data visualizations, and gave roughly half the participants each. I think 45 minutes is a better time estimate for this exercise in total (5 minutes silent contemplation, 10 minutes in pairs, 15 minutes in groups of four, 15 minute for summary with the group). One suggestion has been to give this exercise more practical structure by providing participants a page with 4 quadrants, each signifying one of the questions. The participants are then asked to note down their observations in each quadrant, and circle one observation in each quadrant that stood out to them.

Community Building: Improve a Data Visualization

• When: In general, the 1-2-4-all exercise above is followed by a bit of a lecture-style introduction into some concepts that give the participants a bit of practical background on some of the visualization principles they co-created (e.g. pre-attentive processing, Gestalt psychology). Following this, the participants get the opportunity to take an existing data visualization and improve upon it.
• Duration: 60 minutes
• Why: Learning new principles and ideas functions best through practice and trying it out. In this activity, participants will work in teams of 3-4 to take an existing data visualization and improve upon it visually. Again, this structure aims to give participants a framework for how to interact in teams and learn some of these principles of constructive interaction in a non-threatening setting.
• What: Participants will be asked to team up in groups of 3-4 and pick a data visualization. It can be a plot they have made (voluntary!), or they can pick one of the ones that the instructors bring along. During the first ten minutes, participants identify improvements to be made in the existing visualization, using the same rules as during the group-phase of the 1-2-4-all: each person gets to say one thing, then passes the talking pen off to the next person. A person who would not like to add anything can say “pass” and pass on the pen. Participants should invite the most junior person to speak first, but they’re free to say “pass” and let someone else begin. One person takes notes. Once all the possible improvements have been put onto paper, participants vote on their top 5 improvements (each gets 3 votes), and start collecting ideas for how to improve in a similar manner as before. They take a piece of paper and pin it to the wall, then use the crayons to draw out a sketch of the new visualization. If there is time, they can attempt to create it on a computer, but emphasis here is on creating a prototype together on paper. During the last 15 minutes, a member of each team will share the results with the rest of the groups.
• Notes: Some of the design elements of this exercise were chosen particularly to enable positive group collaboration and facilitate the best learning outcome. Do not pick figures from your own field as bad examples for visualizations, unless you can be absolutely sure that the author is not in the room (for example, by choosing your own). Having a figure you might have spent weeks on publicly dissected without being warned that this might be happening is absolutely terrifying and has a chilling effect. You do not want this to happen. Thankfully, the internet is full of terrible, professionally produced visualizations for you to pick from. The exercise explicitly uses crayons on paper, because with computer-based exercises, the group can be too easily side-tracked into specific questions of how to plot one particular thing with that specific plotting library. In addition, team members might have different skill levels in computer-based visualization, which may lead to some participants programming, while others sit at the sidelines being excluded from participating. Secondly, we chose to pin paper to the wall specifically because it means that participants are all facing the same way, toward a common goal. This is markedly different from paper that's spread on a table, with people standing around. In the latter case, participants disagreeing with someone else's suggestion do so while facing them, which can feel more confrontative than while everyone is facing the same way towards the paper on the wall.
• Material: pens/crayons, print-outs of data visualizations, large sheets of paper, tape, empty walls
• Observations and Feedback: Expect ~10 minutes to pick a figure, ~30 minutes to prototype a figure, and another ~15 minutes to collect the results (depending on group size, of course; this was for ~50 participants). Participants really enjoyed both the figures and the crayons. It made this much more playful and much less intimidating than it would otherwise have been (and some participants noted that specifically in the feedback). One out of ten groups chose to use their own figure instead of one of those I provided. Putting the (empty) posters on the wall was really helpful, I think, because people started talking about the poster, rather than toward each other (which suggested might be less confrontational). We also observed that many participants who didn’t want to share their group’s conclusions during the first exercise participated enthusiastically in the drawing, so it seemed to have succeeded in getting people to interact.

Facilitating the Hacking

The project work component of Astro Hack Week requires careful facilitation in particular, because it is otherwise very free-form and can be very intimidating to new participants and those who are introverted and/or shy. In particular pitching potential projects to the entire group requires a lot of self-confidence in one's ideas and ability to convey them in less than a minute.

We start this part of the workshop with a short introduction into what a hack is, and the characteristics of a good hack (see also intro slides linked above). We alert participants before the workshop as well, and encourage them to start submitting hack ideas to a shared document (though not many do). I think in the future, it would be worthwhile to have a hack development workshop before actually going into the activities described below, but we haven't tried that out yet. The goal here would be to get those with hack ideas to fill out a hack template that asks relevant questions about the project, to be answered before the project can commence.

Hack Pitch Preparation: Impromptu Networking

• When: Directly after the intro about hacking or a hack development workshop
• Duration: 6 minutes
• Why: Standing up in front of the entire group and sharing an idea is a daunting task. Some participants might not be sure whether their idea is something they really want to work on, or whether it’s good enough to share. The round of impromptu networking is an opportunity for participants to test out their idea on a small number of individuals first, refine how they present the core objective of their idea, and decide whether they would like to share it with the group.
• What: Impromptu networking in this iteration only involves only a single question: “What is it you’d like to pursue this week?” Each participant is invited to find another person, ideally one they haven’t interacted with yet. Within two minutes, they answer this question to each other, either in the form of a hack idea, or something else they’d like to achieve. This is more free-form in the sense that the other person can ask questions and it can be more of a conversation. We need to remind participants to be mindful of their time and share it with their partner. After two minutes, participants find a new partner and do the same thing again, 3 times in total.
• Material: Slides

Hack Pitches

• Duration: as needed, about 15-20 minutes
• Why: At some point, the group needs to know about the different project ideas, so they can pick which they'd like to work on. We haven't found a good way other than asks people to present their ideas to the group, ideally within about 45 seconds each.
• What: Each participant gets the opportunity to stand up in front and share their hack idea with the group. Ideally, one would start with 3-4 experienced participants to model how this is best done, then asks some of the newer participants to share their ideas, then go back to some experienced participants. During the hack development phase, participants could also write down short descriptions on a set of slides, or fill out paper versions of the hack template. Those that don't want to stand up in front of the group could ask the organizers to pitch either from the slides or from the shared notes.
• Material: Potentially slides with hack pitches, a microphone
• Observations and Feedback: Pitching hacks is scary for some participants, and we haven't yet found a way to lower the barrier for those that are uncertain about stepping in front of the group to pitch their idea. This is definitely an area of future development.

Hack Sorting

• Duration: about 30 minutes
• Why: Once all hacks are pitched, participants need to sort themselves into groups to actually start working on these hacks. Many participants might be interested in more than one hack, and many might not be sure where they want to contribute. This phase of the process allows participants to go around and chat with those that proposed hacks and decide whether they want to work on that particular project for the week.
• What: Participants who proposed hacks station themselves at different points throughout the room. In four 7-minute intervals, participants can go through the room and check out a hack. We suggest they aim to answer three questions to themselves: What do I want to work on this week? Can I contribute to this project? Do I want to work with this group? After 7 minutes, participants are invited to switch and explore a different hacks. Those that have already decided on a hack may remain with the group, but are asked not to monopolize the conversation with details of the project, but rather allow newcomers to ask questions as well. It is important to remind participants and project leaders to give each other time to ask questions, and not dominate the discussion.
• Observations and Feedback: The time people needed was very uneven. It is difficult to keep some projects from just going ahead and starting to work. We don't have a good sense at the moment for how to prevent that from happening.

Throughout the Hacking

One key component of facilitating the hacking is for a designated facilitator to spend the project time looking out for needs that the groups might have. This involves checking in with the different hack teams regularly and ask questions. How are things going? Is the group missing any expertise? Observe the group interactions and find out whether there are participants who are being (unintentionally or not) excluded from group discussions and work. If so, gently steer the group leaderships towards including these participants, or, if necessary, have a discussion with them in private. Also watch out for participants who might be on their own. Ask them how their week is going, whether they need help connecting to a project. Back off if they indicate that they have some work they wish to do on their own.

We generally repeat hack pitches every day after lunch. Some hacks finish early, others fail early (we encourage participants to try out new things, and fail quickly), and so new hacks might emerge during the week that are being pitched. These pitch sessions are also an opportunity for teams to ask for help, if they are missing specific expertise to make progress on their project.

Wrap-Up

• When: short wrap-up sessions during the week as needed (usually not on Mondays, but perhaps Tuesday, Wednesday, Thursday), and then a big final wrap-up session on Friday afternoon.
• Duration: At least 60 minutes, but longer as needed
• Why: Sharing outcomes of projects is a big part of the workshop. The wrap-ups provide participants to share outcomes: a scientific result, a tutorial, a software tool, a product, a figure, or simply something they've learned. It is particularly important to encourage participants to share projects that failed. Too often, hackathons showcase the most polished and most shiny result, which can reinforce impostor syndrome. It is important to acknowledge that most hacks fail, and that failed hacks are valuable learning experiences that we encourage to be shared.
• What: Participants are encouraged to submit a wrap-up slide to a shared presentation that summarizes their project. These wrap-up slides are often written during the last lunch break on Friday, or shortly before the wrap-up begins. Each slide includes a title, the participants that contributed, a short summary (1-2 sentences) and what the outcome was. Often, these slides also include an image or a figure that signifies the outcome. For online projects and software, short (1 minute) demos are also permissible. Participants are then encourage to come to the front and summarize their project to the group (in about 1 minute), while the group cheers.
• Material: shared presentation with a template slide

Feedback

Feedback is an important component of the hack week. We don't know if something is going wrong if we don't have a mechanism for people to tell us so! It is helpful to stress at the beginning of the week that feedback is always welcome, and that the organizers are always open to receiving feedback.

In addition, we have two mechanisms to allow participants to provide feedback:

Feedback Box

• When: every evening after the hack show-and-tell
• Duration: <5 minutes
• Why: getting feedback during the week is important for us to react and adjust quickly. Feedback should be anonymous in order for people to be able to give it more openly. Emphasise that this feedback is important to us, because we’re experimenting with different forms of interactions here and are trying to learn what works/doesn’t work.
• What: Ask participants to take a post-it, and think about these two questions: “What did you observe that worked well today? What would you suggest to improve any part of the day?” Participants can add their sticky notes to an anonymous box that the organizers empty and read each evening.
• Observations and Feedback: This has worked really well for things that needed adjusting on-the-fly during the workshop. The box needs to be somewhere that is easily accessible to participants, and where they won't be easily exposed as writing a feedback note in order to allow them to feel comfortabel submitting especially critical feedback. Participants have also used this to give positive feedback, which is great for us organizers, both in terms of encouraging us and in terms of helping us know what worked well.

Feedback Survey

• When: Ideally directly after the last lunch break (note: you can run this after the workshop ended, but response rates go down significantly)
• Duration: about 20-30 minutes
• Why: The survey enables us to get feedback on a broader range of topics related to Astro Hack Week, and enables us to learn about how participants interact at and with the workshop. We have also been tracking different metrics related to learning, open science and reproducibility across the different years and across different types of hack weeks. In addition, the surveys have been invaluable in improving hack weeks and helping us secure funding in the future.
• What: a survey on Qualtrics (questions to be finished during the week); participants will be asked to fill out this survey (anonymously) after lunch on the last day, in order to get as large a sample as possible.
• Material: survey on qualtrics (see the GitHub repository associated with this documentation for a template) Observations and Feedback: the survey has been the single most invaluable resource in helping us adjust and improve the workshop every year. We have received an enormous amount of candid and constructive feedback that has helped us design our facilitation strategies better. In addition, Astro Hack Week runs activities, especially with respect to facilitation, as experiments. We re-design parts of the survey every year in a way that allows us to test whether these experiments have positive or negative outcomes.