Skip to main content
Search form
  • 00:09

    ADAM D. GALINSKY: Today I'm goingto talk to you about how to give a great research talk,how to present your research to an audience.Now, one of the first questions you have to ask yourselfis, who is my audience?And this is important for a couple reasons.One reason is that different disciplineshave different norms for how to present research.What I'm going to do today is I'm

  • 00:30

    ADAM D. GALINSKY [continued]: going to present to you research from the perspectiveof a social psychologist or organizational behaviorist.The second reason why you want to know who your audience is,is because it's always a good ideato cite the people in your audience.People love seeing their name on a screen.So if you know who is in your audience,you might be able to give them a nice shout outand make them even more likely to be

  • 00:52

    ADAM D. GALINSKY [continued]: enamored with your research.One of the things that you want to think about your researchpresentation is it's not your paper,you're not presenting your paper to an audience.What you want to think about is you're actuallypresenting a trailer of your paper to an audience.Typically, your talks, when you're presenting research,are 10 minutes or 15 minutes, so you don'twant to give all the details.

  • 01:14

    ADAM D. GALINSKY [continued]: You want to give the best details.You want to make it the most enticing in the same waya movie will try to entice the audience to come see itby showing a great trailer.You want entice the audience to then read your paperafter seeing your talk.Again, your research talk is not your paper.So one of the key things that youwant to do when giving a research talk

  • 01:34

    ADAM D. GALINSKY [continued]: is to simplify it as much as possible.You want to make it as easily digestible for an audienceas you can.Now, I, personally, call this the inception rule.If you remember the movie Inception to do inceptionrequires you to come up with the simplest version of your ideaso it can grow naturally in your subjects mind.That's exactly what you want to happenis you want the simplest version of your idea

  • 01:56

    ADAM D. GALINSKY [continued]: to be condensed into your key research takeaway.And then what you want to do is youwant to repeat that key research take away again and againand again so that it actually becomes embeddedin your audience's mind.Now you know your audience.You have your key research takeaway.The next thing you have to ask yourselfis, how do I begin my talk?

  • 02:18

    ADAM D. GALINSKY [continued]: And what I tell people to do is to begin their research talkby focusing on a puzzle.Now, that could be a research puzzle,it could be a real world puzzle, itcould be your own personal puzzle that you have.What I'm going to do today is I'mgoing to start from the beginning of a research talkto the end of a research talk usingmy own research on the psychological effects of having

  • 02:40

    ADAM D. GALINSKY [continued]: versus lacking power.And so I'm going to give you a couple of waysthat you could start a talk with,that I've used in my own research talks,as a way to understand how to present this puzzleto begin a talk.So let me tell you about a little puzzle.In 1960s, poultry farmers would have a lotof different types of chickens.And some chickens were too lazy to produce eggs.

  • 03:02

    ADAM D. GALINSKY [continued]: Some chickens were too anxious to produce eggs.And some chickens were really good at producing eggs.So what these poultry farmers naturally didis they bred their chickens basedon those high egg-producing chickens.But something remarkable happened.Egg production didn't go up.It actually went down.And so what I'm going to do todayis talk to you about what that puzzle was.

  • 03:23

    ADAM D. GALINSKY [continued]: Why would breeding for the high egg-producing chickenslead egg production to go down rather than up?Here's an example of how to start a researchtalk with a personally relevant example.This is actually a story about my collaborator Deb Gruenfeldof Stanford University.And it was really the story that started our entire researchprogram on power.

  • 03:43

    ADAM D. GALINSKY [continued]: One day she got on a plane.She sat down, and a high-powered businessmansat down next to her.And the overhead fan was annoying him.Now, rather than turning it off, he justpushed it out of his way right into Deb's face.And Deb sat there getting colder and colder and colderover the course of the flight.And it wasn't until the end that she realized,I could have actually just turned off the vent myself.

  • 04:05

    ADAM D. GALINSKY [continued]: And so she stewed for a number of days,asking herself two questions.A, why did this man act and she didn't?And B, why did he act in a way thatdidn't seem to take into account anybodyelse on the flight but his own needs?In the last example I just gave you,the business man took action and Deb didn't.And so one of the things that I study

  • 04:25

    ADAM D. GALINSKY [continued]: is when do people take action, and how does power play a role?I want to give you a real-world example that actuallykind of challenges that point.If you remember the famous picture from Tiananmen Square,there's a man standing in front of these long line of tanks.And we know the literature on powersays the powerful take action, yet this man was powerless

  • 04:47

    ADAM D. GALINSKY [continued]: and he was taking action.So we need to solve the puzzle.When do people without power takeaction versus when do people with power take action?Now, once you've given your opening examplethat motivates your research, the next thingyou really need to do is to provide the audiencewith a roadmap.Lots of research shows that roadmapsare critical for helping audiences understand a research

  • 05:11

    ADAM D. GALINSKY [continued]: talk.And so I'm just going to give you one example of a roadmap.I presented three ideas as my roadmap.First, power reduces perspective-taking.Second, the powerful take action.And third, power differences can help teams at various pointsbe more successful.What you can also do is you can use this roadmapto let people know where they are in the talk.

  • 05:32

    ADAM D. GALINSKY [continued]: So in this case, I've grayed out the second two pointsand just focused on the first point.Now, once you've given your opening exampleand provided your roadmap, now youneed to set your theoretical foundation for your talk.Now, sometimes that's pointing to a particular paper.So in this case, my theoretical foundation for a lot of my workis a paper that was written by Dr. Keltner, Deb

  • 05:52

    ADAM D. GALINSKY [continued]: Gruenfeld and Cameron Anderson.And I talk about how that theory has reallymotivated the specifics of my empirical research.Here's another example.I'm going to talk to you today about howpower and perspective-taking might be inversely related.So in one of my research talks I pointed outthat the powerful tend to be really poor social connectors

  • 06:15

    ADAM D. GALINSKY [continued]: and adapters, but perspective-takerstend to be really good at both of these actions.So what you can see is we have this relationshipin literature.The powerful aren't very good at social adaptation.Perspective-takers are very good at it.And that's sort of going to lead into our core hypothesisof whether power reduces perspective-taking.Here's just one more example.

  • 06:36

    ADAM D. GALINSKY [continued]: I'm going to be talking about power differenceshelping groups function more effectively.And so the two theoretical foundations that sort ofrelate to that question, one is, "the hierarchy is functional"theory, and the other as "hierarchy is dysfunctional."So in my research talk I first presentthe evidence and the theory behind why hierarchy

  • 06:57

    ADAM D. GALINSKY [continued]: can be functional for groups, and then I'llpresent the evidence about why hierarchycan be dysfunctional for groups, and thentry to understand when a hierarchy is functionalversus when it's not.So now, you've given your example,you've given your roadmap, you'veset your theoretical foundation.Now comes to a key point that you want to clearly articulateyour hypothesis.

  • 07:18

    ADAM D. GALINSKY [continued]: So in this case, I have a very clear hypothesisfor the first part of my talk, powerreduces perspective-taking.That is my first hypothesis.I often present more than one studywhen I give a research talk.So I'd like to provide a study overview.So and I often like to provide this studyoverview in a pictorial sense.Here you can see that I'm going to presentfour studies in my research talk,

  • 07:39

    ADAM D. GALINSKY [continued]: and I lay out each one of those studies.And then when I'm at each study, Iput a little red box around it.So you can see the first study isgoing to be about power and perspective-taking,and it has a little red box around it.So once you start describing your study,you've already articulated your hypothesis,now you've got to describe your methods.And I always think of describing your methodsas a kind of audience participation task.

  • 08:02

    ADAM D. GALINSKY [continued]: You want your audience to feel like they are actuallypart of your study, they have that same experience.So for example, a lot of times when I'm presenting about powerand perspective-taking, I actuallyhave the audience do a little perspective-taking task.So what I want all of you to do out in the worldis to raise your dominant hand, and Iwant you to draw a capital letter E on your forehead.

  • 08:24

    ADAM D. GALINSKY [continued]: So what I then point out to an audienceis that E that you draw on your foreheadcan be drawn in one of two ways.It can be drawn like this, so it'sa perfect E to an audience.That's the perspective-taking E. Or it can be drawn like this,so it's a perfect E to yourself.That's the egocentric or self-focused E.And so I'm letting the audience understandexactly what my participants went through in this study.

  • 08:47

    ADAM D. GALINSKY [continued]: So I have my hypothesis, I have my sample,I have my independent variable, in this casean experimental manipulation, and thenI have my dependent variable.Then what I often do, especially whenyou're doing an experiment, is I lay outthe experimental procedure.So I say what happens when participantscame into the lab, what did they do first, whatdid they do next, et cetera.

  • 09:09

    ADAM D. GALINSKY [continued]: And what I'd like to do is I liketo show my manipulations to the audience.So you can see here, this is how we manipulated power.We asked people to recall a time in which they had power.And I lay out exactly what the participant saw,or, in the other condition they thought of a timewhen they lacked power.And again, I show the entire manipulationso people can see exactly what our participants saw.So then after I show my independent variable,

  • 09:31

    ADAM D. GALINSKY [continued]: I now show my dependent variable.And in this case, it's the "draw the E on your forehead"task I already presented.Now you've described your methods.It comes to presenting your results.Here's how I like to present my results.I like to show people the figure legendwithout any of the figure bars.And I like to restate my hypothesis.

  • 09:52

    ADAM D. GALINSKY [continued]: I like to say, remember, our hypothesis is that powerwill reduce perspective-taking.So we're going to expect people in the power conditionto draw more self-focused Es than peoplein the low-power condition.And then I say, so let's look at the low-power condition first.You can see here is about 10% of people drew a self-focused Ein the low-power condition.

  • 10:13

    ADAM D. GALINSKY [continued]: But now let's see what happens in the high-power condition.We can see now it actually almost tripledthe number of people in the high-power condition whodrew the self-focused E. Almost a third of the peopledrew a self-focused E.So what you can do is you can present your results slowlyover time so that people can clearly digest and understandexactly what the results were.

  • 10:36

    ADAM D. GALINSKY [continued]: So remember I presented you my study overview.Now we're finished with study 1.Now we're going to go to study 2.I come back that same slide, but now I move the red bar overso now it's over study 2.So people know where they are in the sequence of studies.Now, one of the most important thingswhen you're presenting multiple studiesis you want to make sure there is a consistency between how

  • 10:59

    ADAM D. GALINSKY [continued]: you present each study.I have the same four components again.I have the hypothesis.I have the sample.I have the independent variable, and I have dependent variable.In the exact same format.If you use the same format over and over again,again, it's easier for people to digest it.Here's another example of helpingpeople feel connected to what it was like to be a participant.In this study, we actually, just like Deb Gruenfeld

  • 11:21

    ADAM D. GALINSKY [continued]: in the opening example, we brought people into the lab,we manipulated power, and then we placed them in a roomwhere an annoying fan was blowing in their face.And then I described what our dependent variable was here is,did they act on that fan?Did they move it?Did they turn it off?Did they slow it down?Did they take any action to reducetheir discomfort of having this fan blowing in their face?

  • 11:41

    ADAM D. GALINSKY [continued]: Now, here again you can see consistency in the results.Notice something really important.I have the power conditions in the exact same sequencethat I had in the first study.So low-power comes first, then high-power comes second.And here again I present the figure legend.Then I show the low-power, the numberof people that move the fan, and then I show the high-powerand can point out how almost twice as many people

  • 12:03

    ADAM D. GALINSKY [continued]: in the high-power condition moves the fan.So at this point I'm going to change tracks a little bitin this research talk.So I'm going to say, look, I've shown you so fartwo really important points.A, power reduces perspective-taking.B, the powerful take action.What I'm going to turn to now is whether power differencein a group can actually help that group

  • 12:24

    ADAM D. GALINSKY [continued]: perform more effectively.And I'm actually going to be usingsome real-world archival data from the National BasketballAssociation and Major League Baseball.But because I'm making a transition now,I'm going to step back and tell a little bit more theory again.So in this case, what I'm going to dois I'm going to solve that chicken puzzle that I gave youbefore.So remember I told you that when poultry farmers bred

  • 12:46

    ADAM D. GALINSKY [continued]: for the high egg-producing chickens,egg production didn't go up, it went down.Now, why would that be?Well, it turns out when you've gota lot of high egg-producing chickens in the same coop,they're also the most competitive,status-oriented chickens, and they fightfor territory and for food.And actually, they peck each other to death.

  • 13:07

    ADAM D. GALINSKY [continued]: So breeding for high egg-producing chickens not onlysent egg production down, it sent bird mortality up.And so this really relates to a theory about this ideathat you don't want too much talent comprised in a team,especially when that team needs to work togetherto coordinate their behavior, because you get these statusconflicts exactly as we saw in the chickens.

  • 13:29

    ADAM D. GALINSKY [continued]: Now, this isn't just true for chickens.Lots of companies have figured this out.Google, for example, when they started out,the engineers were like, we hate managers.And they tried to reduce hierarchy,but it was a total disaster because there was chaosand people quit.And so this is something that generallyseems to be true both in the animal worldand in the human world, too.So when we don't have hierarchy, we get competition

  • 13:53

    ADAM D. GALINSKY [continued]: and we get chaos and we get confusion.And so I've coined the phrase called the "too-much-talenteffect."The problem of getting too many superstars on one group or teamtogether, especially when that group needsto coordinate their behavior and work togetherto produce a group outcome.And I want to give you a real-world example of this

  • 14:14

    ADAM D. GALINSKY [continued]: from 2010, when three of the best basketballplayers in the world joined forces on the Miami Heat.LeBron James, Dwyane Wade, and Chris Bosh.Now, they had a big celebration to announce the super team,but you already knew there was a little confusion going onbecause they called it Wade's House, Lebron'sKingdom, and Bosh's Pitt.And as one sportswriter pointed out, it can't be all three.

  • 14:37

    ADAM D. GALINSKY [continued]: And they actually said at that moment,without a defined pecking order, there's notgoing to be any order and they're not going to do well.Now, what's interesting about the Miami Heat,in this first year when they were togetherthey won a lot of games, because they had a lot of talentand they could beat teams.But they were the second-worst teamin the entire NBA at winning close games, when coordination

  • 14:57

    ADAM D. GALINSKY [continued]: really mattered.So building off this example, we actuallydecided to go ahead and test this was some archival data.And what you can see here, again, I'mgoing to be presenting this archival data set.Again, it's got the same structurethat I had in my experimental studies.We have a hypothesis, we have a sample,we have the independent variable,and we have the dependent variable.

  • 15:19

    ADAM D. GALINSKY [continued]: And what I'll describe here is basically a little bit.We took 10 seasons of NBA data.We got a talent measure that reallyfocused on how much talent was on the team.And then we just looked at their performance,what was there win percentage in that season.Now, here we're going to see the data presented a littledifferently.What I'm presenting here is a scatterplot of all the data.

  • 15:40

    ADAM D. GALINSKY [continued]: What you can see is a scatterplot.And what I would do when presenting the scatterplotis, again, to re-articulate the hypothesis, that we werepredicting in the National Basketball Associationthere would be a too-much-talent effect,meaning that talent will help a team but only up to a point,and then it might actually hurt a team.Now what that means from a prediction perspectiveis we expect a quadratic relationship between talent

  • 16:01

    ADAM D. GALINSKY [continued]: and performance.That means that there will be sort of a quadratic curve.And you can see that exactly in our datathat teams that are more talentedare better, but only up to a point,and then their regression slope turned downward.Now I'm going to go into my final study.And the last thing he basically showed a never-enough--talenteffect.And our hypothesis was it was because in basketball, we need

  • 16:23

    ADAM D. GALINSKY [continued]: to coordinate our behavior.So we asked ourself a question.Is there a sport where there's very little coordination?And that's Major League Baseball.In fact, one famous sports writerdescribed baseball as an individual sportmasquerading as a team sport.Now, given this team doesn't-- the players don't really needto coordinate their behavior, now we expect somethingdifferent, a never-enough-talent effect.

  • 16:46

    ADAM D. GALINSKY [continued]: That in baseball, it's always better to have more talent.So now in the final study again, you see the same exact setup-- hypothesis, sample, independent variable, dependentvariable.And now, what I"m going to do is I'm going to show you nowa graph that looks very similar to the last graph.It's a scatterplot.But it's going to have a different relationship.Here again, I would re-articulate the hypothesis.

  • 17:07

    ADAM D. GALINSKY [continued]: Because baseball doesn't require coordination,we predict a linear effect so that more talentwill lead to better performance with no cost over time.And what's really nice in this type of situationis you can actually present both samplesside by side with each other.You can show the basketball side and thenyou can show the baseball side, and you can clearly seethe different pattern of data.

  • 17:28

    ADAM D. GALINSKY [continued]: In basketball, there's a too-much-talent effect.Talent helped performance, but only to a point.In baseball, there's a never-enough-talent effect.So more talent is always better.Now we come to the end of the talk.And you want to usually do four thingsat the end of a research talk.The first thing you want to do isyou want to offer the conclusions.The second thing you want to do is come back

  • 17:49

    ADAM D. GALINSKY [continued]: to your theoretical foundation and mentionsome theoretical implications.Third, you want to consider some of the practical applicationsof your research.And fourth, you want to post some follow-upor next-step questions, some future directionsfor yourself or for other researchers.And then are considered some of the practical implicationsof it.

  • 18:10

    ADAM D. GALINSKY [continued]: So for example, if you want people to take action,you've got to make them feel powerful.But also to give them a little warningthat if you make them feel powerful,they might not take other people's perspectives.And when you want a team to coordinate their behavioreffectively, you don't want too much talent on that team.So these are two practical applicationsof the research I just presented.

  • 18:31

    ADAM D. GALINSKY [continued]: So finally, you'll give some future directions.So for example, you might ask the question of,when does power lead to perspective-taking?Or, what are the benefits for an organization combiningpower and perspective-taking together?Or you might think about the too-much-talent effect and ask,is there a way a leader can act to actually overcome

  • 18:52

    ADAM D. GALINSKY [continued]: this too-much-talent effect?So what I've tried to do today is present to youhow to give a great research talk.Starting from the very beginning to the very end,let me give you some key takeawaysfrom how to present a great research talk.The first thing you have to do is know your audience.So what type of discipline are they focused on.What you want is you want your audience

  • 19:14

    ADAM D. GALINSKY [continued]: to easily digest your talk, so youwant to make it as easy to follow as possible.And so here's some ways to make it easy to follow.You want to give them roadmaps and study overviews, especiallyin the visual form.You want to keep it simple, get to the core key takeawayof your idea, the simplest version of the idea.

  • 19:34

    ADAM D. GALINSKY [continued]: You want to make every time you present somethingas consistent as possible.So you're presenting your methodsin the same consistent way.You're presenting your data in the same consistent way.Now, one of the ways that you can engage your audienceand help them digest it is by helping themfeel like they are a participant in your study,they were part of your study.Now, I mentioned that you want your data presented

  • 19:55

    ADAM D. GALINSKY [continued]: consistently across studies, but you alsowant to present that data visually.Because people have a much easier time processinginformation visually than they do in words.And then the final thing that youwant to do to give a great research talkis remember to highlight your contributions.What did you present today, how does it advance science,how do we understand the world better

  • 20:17

    ADAM D. GALINSKY [continued]: because of your research.Now, if you follow these easy steps from beginning to end,you will get a standing ovation.

Video Info

Publisher: SAGE Publications Ltd

Publication Year: 2018

Video Type:Tutorial

Methods: Professional skills, Professional communication

Keywords: audiences; presentation skills; theoretical approaches; visual aids

Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:



Adam Galinsky gives guidelines on presenting findings from a researcher’s perspective.

Looks like you do not have access to this content.

Best Practices for Presenting your Research

Adam Galinsky gives guidelines on presenting findings from a researcher’s perspective.

Copy and paste the following HTML into your website