Cookies on Knowhow Nonprofit

We use cookies in order for parts of Knowhow Nonprofit to work properly, and also to collect information about how you use the site. We use this information to improve the site and tailor our services to you. For more, see our page on privacy and data protection.

OK

Skip to content. | Skip to navigation

Community-made content which you can improve Case study from our community

How to present research findings

A network I participate in was recently asked the question:

"if anyone on the list has gone from academia into policy research of any kind and has given presentations based on academic research to think tanks, government departments, NGOs or similar and had any useful insights? "

Now it's some time since I was part of the academic world (I daren't call myself a former academic), but on the basis that I am no expert, the following seem to work for me. What would you add?

Things you'll need

  • Knowledge of your audience
  • Time to practise your presentation
1

Know your audience in advance

Know who is there, their interests/jobs/methodological bugbears, and their names. It's great in a Q&A when a presenter responds to a question and begins with the name of the person who asked.

2

Tailor your presentation to that audience

Surprisingly, many don’t do this and just present 'their research' and expect people to want to think through why it is relevant

3

Highlight the context

If there is a policy or practice context, highlight this - more so than theoretical/methodological context (though you should briefly allude to these to highlight you know your onions).

4

Policy or practice recommendations

If there are policy or practice recommendations, you must draw these out - at the end of the day, people in these organisations are using research-based evidence as a means to making decisions - so help them make those decisions.

5

Include recommendations that are actionable and that help your audience.

Don’t just have recommendations that say 1) we need to do more research and 2) my research has highlighted that this issue needs more funds. Both are acceptable as part of wider recommendations; the latter needs to explain how funds would be spent and on what.

6

Time and practise what you do

If you have a 10m slot, pace out the sections (context/meat/conclusions) so that you don’t spend 8m on context and then overrun.

7

Avoid powerpointlessness

Focus should be you, not your over-detailed slides. When I see someone with a 10m slot and 20 slides I groan (NB: #guilty). Just because a paragraph has a bullet point in front of it, it doesn’t mean it is a bullet point. Try www.prezi.com as a ppt alternative. Its cool. Here's my attempt at a prezi.

8

Visualise your data: try infographics!

Look at manyeyes and informationisbeautiful for inspiration. If you've got data tables, make sure they are ledgible from 10ft! If you put your data tables on google docs it also means your audience can then access them.

9

Keep it simple

This is difficult, but keep it simple, avoid jargon. People will probably only remember 3 points at the most.

10

Don't conclude its all just about the methodology you have used

From my experience, people don’t want to be told that different methods produce different results, that the evidence is inconclusive - ie all the richness that researchers care about. Hopefully you can bring this out in the Q&A.

11

Before and after

The actual 'presentation' is only part of it. Put your slides on www.slideshare.net or your blog so that afterwards people can get them and comment on them. Ask them for questions in advance. Connect with them on www.LinkedIn.com afterwards.

Further information

Finally, this is good: 15 and a half ways to improve your presentation: from Boring to Bravo, blog post.

Contributors

Page last edited Jul 25, 2017 History

Help us to improve this page – give us feedback.

1 star 2 stars 3 stars 4 stars 5 stars 2.9/5 from 1740 ratings