Remidius Ruhinduka, a lecturer and research fellow at the University of Dar Es Salaam, participated in a short-term visiting fellowship with CEGA-EASST at UC Berkeley from April 18-22, 2016. Over the course of the week, Dr. Ruhinduka was able to meet with CEGA faculty across UC Berkeley and Stanford University to discuss research and potential avenues for collaboration.

Written by Fitsum Mulugeta, Junior Research fellow at the Ethiopian Economics Association and EASST Visiting Fellow in Fall, 2012.

There were a number of reasons that led me to choose research as a career path.  Importantly, it is filled with new set of challenges every day and that there is nothing routine about it. In addition to living each day differently, research in development gives me a satisfaction of being able to contribute towards changing the lives of those who are less fortunate. Yet whether we admit it or not, deep down we also want to achieve immortality by contributing to the body of knowledge; contributing something that would be cited even long after we pass away.

Even if it has been a long standing tradition in science and research, it is only recently that I have discovered the ways in which researchers want to share in a different light. This has reignited my reasoning in being and staying a researcher. In my opinion, humanity is greedy in general, as we seek to gather all we can and keep it for ourselves. However, there remains at least one exception: knowledge. Generally speaking, people are happy to share their knowledge, be it in the form of oral literature in traditional communities or in the form of writing books, publishing papers, teaching in universities or conference presentations.

In the realm of research, people seem to be very happy to talk about the projects that they are working on, the papers they have published, the results they have found, and others’ works in similar domains. I’m continually amazed at how willing to share and helpful academics are when it comes to their work. Thus my renewed sense of pride in being a researcher – it is indeed a blessing to be in the group of people who sees sharing as a way of life in the world of greed for finding and keeping things for one’s self.

These were all thoughts that came to mind during the BITSS – IPA workshop on research transparency, held in Kenya this past March. The new wave of sharing knowledge is to share not only what we know, but how we come to know what we know. The workshop covered the following:  how to ensure learning as much as we can from research that doesn’t necessarily  have ‘significant’ results, how to make research reliable (by sharing the entire research process and reproducing results), what tools do we have to make and share pre-analysis plans, codes, data and reports (which include pre-analysis plan registries, best practices for keeping and sharing codes, tools for version control and collaboration, and tools for making dynamic documents).
​
There is more to be said about the workshop but I would like to limit myself to the idea of sharing, which is what we love to do in research. The movement towards transparency in the way we do research is the present and the future of research. While it may mean more upfront work, it is an opportunity to say more about our research than what a limited number of pages of published article or few minutes of presentation would reveal. Eventually, the world of social science research will join this path and I think it is better to start the practice soon and learn it as we go rather than trying to catch the train after it has left the station.

For more information on the workshop, you can read the blog posted by IPA here.
​

Markus Goldstein and David Evans via the World Bank's Development Impact blog recently posted useful tips on how researchers can summarize their work into a 15-20 minute presentation:
​​

1. You have 15 minutes. That’s not enough time to use the slides you used for that recent 90-minute academic seminar. One recent presentation one of us saw had 52 slides for 15 minutes.    No amount of speed talking will get you through this in anything resembling coherence. (And quit speed talking, anyway. This isn’t a FedEx commercial!) There is no magic number of slides since the content you’ll have and how you talk will vary. But if you have more than 15 slides, then #2 is doubly important.
2. Practice. This is the great thing about a 15-minute talk: You can actually afford to run through it, out loud. Running through it once in advance can reveal to you – wow! – that it’s actually a 25-minute talk and you need to cut a bunch. Of course, the first time through the presentation it may take a bit longer than you will when you present, but if you have any doubts, practice again (bringing your prep time to a whopping 30 minutes plus a little bit).
3. You need a (short) narrative. What is the main story you are trying to tell with this paper? Fifteen minutes works better for communicating a narrative then for taking an audience through every twist and turn of your econometric grandeur. Deciding on your narrative will help with the discipline in the points that follow.
4. A model or results? Even if your audience is all academics, you don’t have academic seminar time. So the first thing to do is to figure out which is more important to get across – your model or your empirical results. Then trim the other one down to one slide, max. If the results are your focus (usually the case for us), give the audience a sense of how the model is set up, and what the main implications are as they pertain to the results you will show. Conversely, if it’s the model that’s more important, the empirical results will come later and you can just give the very brief highlights that bolster the key points.
5. The literature. Really, really minimal. If you do it at all, choose only the papers that you are either going to build on in a major way or contradict. For some types of discussants, it may help to include them, even if they don’t meet the other criteria. Marc Bellemaretakes an even stronger stance: “Never, ever have a literature review in your slides. If literature reviews are boring to read in papers, they are insanely boring to listen to during presentations.”
6. Program details. Here it’s a bit of a balance. The audience needs a flavor for the program, they need to understand what it did and how it’s different from other things (particularly other things with some kinds of evidence). But only in exceptional cases (as in, it’s a really different program for theoretical reasons, or you don’t have more than process results yet) do you want this to eat up a lot of your time.
7. You don’t have time to go through the nitty gritty of the data.   We get that every detail about the survey was fascinating (we spend a lot of our lives thinking about this).   But if it’s not key to the story, save it for a longer presentation (or another paper). And if you’re doing a primarily theoretical paper, this is a bullet on one slide.
8. Balance and summary stats. Key summary stats that tell the audience who the people are might make the cut, but 3 slides of every variable that you’ll use are going to be slides you either rip through (telling the audience nothing) or waste most of your time on. Summarize the summary stats. On balance tests: you are either balanced or not.  If you are, this gets a bullet at most (you can also just say that). If you’re not, tell us what’s up and why we should or should not worry.  
9. Pre-analysis plan. If you had it, mention it (quickly). If not, don’t. It’s not critical here.
10. A picture may be worth 1,000 numbers. Sometimes, taking that really packed table which is currently in 12 point font and turning it into a graph is going to help you with self-control and help your audience with comprehension. Put the significant results in a bar chart, and use asterisks to tell folks which are significant.  
11. A special warning about presenting your job market paper. When I (Markus) submitted my job market paper to a journal, the referee report came back noting that this was surely a job market paper since it had 40(!) tables. Key example of how everything matters when you just spent four years of your life collecting each observation. Discipline. You have (or will have) an elevator pitch from the job market – use this to trim your presentation. 

A few bonus resources from others

• Marc Bellemare has a great series of “22 tips for conference and seminar presentations,” many of which apply to short presentations: “Always provide a preview of your results. This isn’t a murder mystery: it’s only when people know where you’re taking them that they can enjoy the scenery along the way.”
• Jeff Leek has a great guide to giving presentations of different lengths, and what your goal should be: “As a scientist, it is hard to accept that the primary purpose of a talk is advertising, not science.” This is doubly true for a 15-minute talk.
• The AEA Committee on the Status of Women in the Economics Profession has a top 10 list. “Never cut and paste a table from your paper onto a slide. These tables are never easy to read and only irritate your audience. Instead, choose a few results that you want to highlight and present them on a slide in no smaller than 28 font.” We’ve pretty much all done this. It’s bad practice. (“I’m sorry you can’t read this table.” “Oh really, then why did you cut and paste that giant table from your paper into the presentation?!”)
• I (Dave) go back and re-read Jesse Shapiro’s guide on “How to Give an Applied Micro Talk” from time to time. It’s more geared toward a full-length seminar, but the advice is so good I can’t resist plugging it here.