Introduction

A major part of being a scientist is communicating your work to others. There is a lot that has been written about both verbal and written communication, thus here I’m just providing some key high-level advice and encourage you to dig through the extensive literature, websites, videos, etc. on the subject.

Key prerequisites for developing communication skills

There are two key points I want to start with:

1. Good communication requires deliberate and sustained PRACTICE.
2. Good communication requires careful preparation that takes TIME.

It is critical that you take every opportunity you have to give presentations, whether it is poster presentations at conferences (large or small), or scientific talks at meetings, or informal presentations to friends, at clubs, etc. Also practice describing your work to people who are not in your field, whether it is colleagues who are not familiar with computational biology, or family and friends who may not even have a deep understanding of biology or computation. Even that 3 minute description of your work to an uncle as you wait to grab food at a family event counts!

The practice does have to be deliberate: Pay attention to how your message is received. What would you change the next time you present your ideas? How would you change your message depending on your audience (e.g., your uncle versus participants at a conference in your field).

The next point is that preparing your presentation/paper takes time. At the bottom of this page you’ll find a rough timeline for a typical paper or 1-hour presentation.

Importantly, part of the time you must account for is time you ask from others: co-authors or other people whose feedback you seek. Every time you ask someone to look at your slides or text, you must allow at least a few days if not a week to allow them to review your materials and provide thoughtful feedback. Just put yourself in their shoes – how would you feel if someone gave you a 10 page paper one morning and asked you for feedback on it before a deadline that same day? Afterwards, you must also account for your time to digest the feedback received and incorporate it in your presentation.

My general advice – you should have a nearly-complete paper/slide deck 1 month prior to the actual journal/conference deadline.

Key principles about communication

It’s not about you

To be very blunt – what you say doesn’t matter. Ok, hopefully it matters to you. What matters more, however, is what people hear/understand from you.

How people perceive what you say depends on a lot of factors. Do they understand the words you say/write? Are they distracted by things you say or the way in which you say them? Are they open to hearing your words in the first place?

Communication is only effective if you and your audience share trust and use a common language. Whether you write or speak you need to get your audience to trust you, and you need to communicate to them in a way that they can understand. To do both things you need to understand your audience: know what engenders trust, and know what language they understand (and I’m not meaning here whether the audience can speak English, but whether they have the necessary background to understand the scientific terms used in your presentation). To be a good communicator you must, therefore, develop empathy (that’s a useful skill in life more generally).

Developing trust

When giving an oral presentation, trust relates to the ways in which you interact with your audience: the volume and tone of your voice, the structure/style of your slides, and even the way you dress or wear your hair. If you look and/or talk like most of your audience, they are more likely to be willing to hear from you.

Importantly – I’m not arguing that you must change who you are to adapt to each separate audience, but it’s critical that you create opportunities to highlight commonalities between you and your audience.

When writing, building trust seems harder to do – you ultimately cannot control who reads your work. In this setting, you build trust by choosing the appropriate publication venue (e.g., don’t submit an algorithmic paper to a journal that primarily publishes results of clinical trials) and by making sure that your presentation matches the presentation style of other papers in that journal. To get a sense for differences in style, pick papers on a very similar subject that are published in Nature, Science, Genome Research, Nucleic Acids Research, or other seemingly similar journals. Do you notice any difference between the way the papers are structured, presented, etc.?

As a final point about trust, for both oral presentations and papers, it is critical that your presentation is ethical – don’t make claims that are not supported by the data you present, don’t overstate the abilities of algorithms you describe, don’t hide inconsistencies between data and your “story”, and make sure to fairly acknowledge prior work in the field. Also, freely share data/code/analysis details in a way that allows others to repeat and validate your work.

It’s all about storytelling

Think back on the last movie or TV show that you watched. Think back on the last literary book or short story/essay you read. Did any of these start with a table of contents explaining to the viewer/reader what is coming next? It’s highly likely they did not. You kept watching or reading through without knowing what was coming next (or because you didn’t know what was coming next), driven by the “narrative arc” set out by the author(s). Your papers and presentations should do the same – guide the reader effortlessly through a story about the work you are presenting. If you feel the need to have “table of contents” in the introduction of your paper or in your presentation (as all too many scientific papers/talks do), then you have not structured your arguments well.

Thus, before you write any part of your paper or put together any slide in a talk, write down on sticky notes all the points you feel you want to make about the work you are about to present. Now move the notes around, grouping them together according to shared themes, and arrange them into a linear thread that will be the backbone of your presentation. Experiment with different arrangements and see which one you like more. Think about your audience. Would certain information flows work better for one audience or another? As you do this, you may find out that there are ideas you forgot to write down – add them into your story. Also, you will find that certain ideas don’t fit well in the narrative arc – discard them (or set them aside in case you change your mind). Of course, you don’t necessarily have to use physical notes – they could be bullet points that you shuffle around in an electronic document. You’ll find that once you’ve come up with a compelling high level story, the difficult work of writing out the text of your paper or designing the slides for your presentation will be a lot easier.

As you convert your bullet points/sticky notes into sentences and paragraphs, it’s useful to also pay attention to the internal structure of your text, making sure that the exposition follows effortlessly from sentence to sentence, from paragraph to paragraph, and from section to section of the paper. This article by George Gopen and Judith Swan has great examples for how the flow of information can be organized in a paper.

Rough timeline for preparing a paper/presentation

his represents a very rough approximation of the time needed to prepare a typical journal paper or 1 hour presentation. The actual time varies significantly depending on your writing speed, graphical design expertise, and familiarity with the subject. Despite the difference in the amount of text you may write in a paper or place on slides, the overall “time cost” can be roughly similar – it’s frequently easier faster to write text than to summarize that text in a few informative bullet points. There’s a famous writing adage “I’m sorry for the length of this text, but I didn’t have time to write anything shorter.”

This timeline is adapted from the book Slide:ology, where it referred to a 1-hour, 30-slide presentation.

Prep time (6-20 hours)

• 6-20 hours (~1-3 days full-time) – Do background research – this is in addition to the research you’ve done to get the results you are presenting in a paper or slide deck.

Laying out the story (8+ hours)

• 1 hour – Figure out your audience needs. For papers – What do papers in the venue(s) you are targeting look like?; For presentations – Who do you expect to be in the audience? What do they care about? What is their level of expertise in the topic you are presenting?
• 2 hours – Generate ideas/points you want to present and write them on sticky notes
• 1+ hours – Organize the ideas into a “story”. You may drop points that don’t fit in the story, and create different “flows”.
• 1+ hours – Get feedback on the ideas and the tentative story
• 2 hours – Refine the structure of the presentation into a sketch/storyboard/outline

Actual writing (20-60 hours)

• 20-60 hours (up to ~2 week of full-time work) – Write slides/text

Refine the text/presentation (8-16 hours)

• 4-8 hours – get feedback on text/presentation
• 4-8 hours – refine text/presentation based on feedback

Note: when you get feedback – you need to account for the time needed by others to read and comment.

Further reading

Mihai’s paper writing advice

Slide:ology by Nancy Duarte (O’Reilly media, ISBN 9780596522346, available at McKeldin) – Mostly written for business, but very applicable to academic talks. I like that it shows different versions of slides indicating how design/presentation choices impact what is being presented.

Even a geek can speak by Joey Asher (Persuasive Speaker Press, ISBN 0-9785776-0-4) – Largely targeted at a business audience but many principles apply to academic presentations.

On speaking well by Peggy Noonan (Harper Collins, ISBN 9780060987404) – Written by Ronald Reagan’s speech writer. Focuses nicely on the mechanics of giving great presentations.