Episode 34: Opinions of Ourselves with Ken DeMarree


Ken DeMarree studies how opinion science applies how we see ourselves. He’s an associate professor of psychology at the University at Buffalo. In our conversation, we talk about how opinion science can be used to understand things like self-esteem, how people sometimes desire opinions they currently disagree with, and how some people just tend to be pretty confident in their views.

Things we mention in this episode:

  • California’s Self-Esteem Task Force (GuardianNYTThe Cut)
  • The psychology of strong opinions can help us understand how people see themselves (DeMarree et al., 2007)
  • More “accessible” self-esteem is more durable and impactful (DeMarree et al., 2010)
  • Seeing yourself in both positive and negative ways makes your self-esteem more susceptible to influence (DeMarree et al., 2011)
  • When we want an opinion we don’t already have, it makes us conflicted (DeMarree et al., 20142017)
  • Some people just tend to be more confident in their views than others (DeMarree et al., 2020)

Transcript

Download a PDF version of this episodes transcript.

Andy Luttrell:

In 1986, California signed into law some legislation that created the “State Task Force to Promote Self-Esteem and Personal and Social Responsibility.” As a political science professor put it at the time, it’s just so California. I can’t imagine Idaho having a task force on self-esteem. But the intentions seemed pure. Looking to the research in psychology, California state legislator John Vasconcellos discovered that self-esteem seemed to protect against a host of negative outcomes. So, the task force was created to understand self-esteem and how we can nurture it to solve a bunch of social problems. 

It was a sign of a cultural shift to come. Self-esteem started creeping into the curriculum as schools developed programs to boost kids’ self-esteem. For example, playing games where everybody complimented each other to make themselves feel good. Now, the research on self-esteem makes it clear that the outcomes of it are not always sunshine and happiness. Sometimes, the pressure to have high self-esteem can be problematic on its own. But nevertheless, the notion of self-esteem and the prize we put on it has spread far and wide. 

You’re listening to Opinion Science, the show about our opinions, where they come from, and how they change. I’m Andy Luttrell, and you might be thinking, “What was all that self-esteem stuff at the top?” Well, what else is self-esteem but an opinion we have of ourselves? I’m excited to talk to my friend, Dr. Ken DeMarree. He’s an Associate Professor of Psychology at the University at Buffalo, and in his early days of studying the psychology of attitudes and persuasion, he started to notice some parallels to the psychology of the self. This insight has inspired a bunch of interesting studies, so in our chat today we’ll talk about how opinion science can be used to understand things like self-esteem, how people sometimes desire opinions that they currently disagree with, and how some people just tend to be pretty confident in their views. 

So, we had talked before about what to cover in your expansive line of work. 

Ken DeMarree: 

Disjointed, disorganized. 

Andy Luttrell: 

But I think the self idea is a through line that cuts through a lot of it, right? And that, as far as I understand, emerged pretty early on. So, could you talk a little bit about… I guess maybe first, for people who are outside of social psych or aren’t super familiar, when we say, “the self,” that social psychologists study, “the self,” what do we mean? And I know that’s a loaded, giant question, but I’m gonna force you to answer it anyway. And what are the questions that fall under that? 

Ken DeMarree: 

I mean, the self is really challenging to define, because I can’t even think of an easy way to define it without somehow using the word self in the definition, but you know, in terms of like our self-conceptions, what we’re talking about is kind of our mental representation of ourselves. And so, just like any mental representation, whether it is your conception of what is a chair, to your conception of different groups, like a stereotype, your mental representation of yourself is going to have all memories that are relevant to yourself, the inferences, and beliefs that you have about yourself, the goals that you have about yourself, so a variety of different sets of relevant constructs, I guess, so whether they’re… Again, like these memories, these inferences, these goals, and so on, important relationships, and so our mental representation of ourself is probably one of our most elaborate, largest mental representations. 

And because of this, and obviously it’s centrally important to most of our decisions, and so because of these combinations of these things, it’s really important to study and understand. In terms of what we’re interested in, oftentimes as attitudes researchers is self-esteem, and so self-esteem, now we’re getting at that evaluation of the self, the attitude of the self. If you want to go with the title of your podcast, the opinion that we have of ourselves. And I think some of my initial observations were just the idea that we could take these initial concepts that we’ve studied in the attitudes literature to understand this really important attitude object, the self, because this seems to be a critically important… Self-esteem is a critically important evaluation. 

You know, predicts a lot of mental health outcomes, well being outcomes, and so on, so I thought it was interesting to understand how I specifically got started in applying attitudes to the self. Research was actually started in a self seminar I started, or I took in graduate school. And I remember reading… I think it was Michael Kernis, 2003, dating myself, it was fresh off the press papers at the time on optimal self-esteem. And in that paper, he was talking about how essentially self-esteem with particular qualities was particularly adaptive to have. And I don’t know if I could remember the full list of all of those qualities, but it included self-esteem being relatively stable. Self-esteem that was evaluatively congruent. He talked about it in terms of implicit and explicit congruence. I think there was authenticity and was probably one or two other features that he talked about in that paper on what optimal self-esteem was. 

And I just saw this as linking nicely to a lot of the research on attitude strength, but also failing to appreciate some of the lessons that we had learned in the research on attitude strength, in particular lessons about how different features of the attitude, although they may predict the strength of the attitude, they can be relatively independent and in the case of some of your own work, they can even interact with each other. And so, this initial conception of optimal self-esteem, there seemed to be kind of lumping together things that potentially were different. I kind of saw that as at least raising opportunities for a little bit more nuance in terms of understanding different variables like ambivalence, and certainty, and accessibility, and so on. So, that was really how that work got started. 

Andy Luttrell: 

So, was Kernis saying that optima self-esteem is defined by those properties of being internally consistent and stable? Is that what you’re saying? 

Ken DeMarree: 

That’s my memory. It’s been a long time since I read that paper, honestly. 

Andy Luttrell: 

Which is all to say that like I think what you’re getting at is if we define let’s say optimal self-esteem as being internally consistent, you could say, “Oh, well, internal consistency might be an informative quality of your self-esteem, but it is not itself necessarily the determinant of what’s optimal.” Is that kind of what you mean?

Ken DeMarree: 

Yeah. Exactly. Exactly. And you know, I mean, Kernis had a different interest in focusing on this optimal self-esteem. He was looking at things like defensive responding, for example. And I kind of… Where my initial thoughts were that, well, this really sounds a lot like attitude strength applied to the self, which means that there’s a lot of different outcomes we potentially could look at, such as whether it’s stability over time, resistance to change, the ability of my self conceptions to predict relevant outcomes. And you know, arguably some of the things that Kernis talked about might have fit under those umbrellas. I’m sure if you look back at my term paper from 2003, I talked a little bit about that. 

But so that kind of motivated me to start including measures of self strength. In other words, these measures related to the strength of our self conceptions, just exporting those attitude strength variables to the self. 

Andy Luttrell: 

It might help to give an example as well, so I guess there are… The one that always comes to mind for me is the ambivalence one, the self-ambivalence one. I don’t know why that one in particular, but maybe you could walk through like-

Ken DeMarree: 

As someone who has several papers on ambivalence-

Andy Luttrell:

Yeah. Maybe that’s the one. But so, I guess it’s just a good example to be able to show like, okay, what does attitude strength mean? What does it mean for things to predict a strong opinion? And how could we just take that, lift it, and place it on the self and make new predictions that aren’t that new, if you know about attitude strength? 

Ken DeMarree: 

Yeah. Yeah. I mean, one of the things we know, so ambivalence is ambi valence. So, you’ve got both valences, positive and negative within an evaluation. So, if it’s an attitude towards my cell phone, I’ve got some positive aspects of it that I like, some negative aspects of it that I do not like, so whether it’s the positive things like the camera quality, the negative aspects like the battery life, and so I may have an ambivalent attitude towards my cell phone if I have a lot of both positives and negatives at my head at the same time. 

And one of the things that this might predict is how malleable my attitude is, how susceptible my attitude is to persuasion attempts, for example. So, if I see an advertisement for a competing cell phone model that says mine is crappy for some reason, that I should buy theirs, I might be very  susceptible to that, because I’m already ambivalent. I’ve already got some positives and some negatives. And maybe this might help me resolve that ambivalence by latching onto these new arguments, especially if they provide new information. 

And so, that’s one way that ambivalence in an evaluation generally might predict instability, for example. And so, we’ve looked at this in one paper where we measured people’s ambivalence towards the self, so I don’t know if you want to get into the details of like objective ambivalence versus subjective ambivalence, like the objective ambivalence, that’s the kind of that presence of both positives and negatives, where the subject of ambivalence, that’s more the experience of feeling conflicted, feeling torn, feeling confused, and so on. And so, we had tried to look at more objective ambivalence in some work that was using relatively subtle change manipulation, so some of those we used conditioning, so kind of pairing positive and negative words with the self. Me, myself, I forget what the exact words were, but it’s probably very close to my, myself, and I. 

So, pairing positive words in one condition, neutral words in another condition, and the more people felt ambivalent about themselves, the more likely they were to change. I guess the more they had the objective ambivalence, and not feeling ambivalent, but rather the more they had both positive and negative features, the more susceptible to change they were in response to that manipulation. So, yeah, we did that in a couple of different ways. It’s an older paper, so I don’t know, if I was to do it again I would certainly do it with better sample sizes and better, tighter methods than I would do it today. 

Andy Luttrell:

But the idea is pretty straightforward, so if I already see myself as having good qualities and bad qualities, I’m more susceptible to being nudged either to see myself as a little more positive or as a little more negative, right? Because I’m already kind of… I don’t have a clear, perfect, fully constructed sense that I’m either good or bad, and so I can be nudged to see myself in a slightly different way. 

Ken DeMarree: 

Exactly. And I don’t know, honestly, if we were… We didn’t include the negative condition, at least in that case. That was, you know, there’s some ethical concerns, or at least with a paradigm that we’d adapted, like they were also unwilling to try to make people dislike themselves. And you know, I would guess that these effects are not particularly long lasting, but it’s better to be safe anyway and not do that. 

Andy Luttrell:

So, are there other qualities of the self that someone could look at in order to know, for example, how changeable someone’s self-esteem would be? 

Ken DeMarree: 

Yeah. Certainly. So, we talked about ambivalence. Yeah, we talked about ambivalence, but you can look at accessibility, or certainty, or importance, and a lot of these have actually been looked at in the self literature. Some by myself, some by other people at different times, so like with the example of self-esteem accessibility, so we’re talking about how easily one’s attitude towards the self comes to mind, is kind of based off of earlier work by Russ Fazio, who looked at attitude accessibility in general and what his work has found is that the more accessible a person’s attitudes are, the more those attitudes guide behavior, so if your attitude towards a political candidate is very accessible in your mind, you’re more likely to actually go out and vote for that candidate. More likely to predict information processing, so if you saw that candidate giving a debate performance, you’re more likely to interpret that performance to be in line with your attitude to the extent that the attitude is active in your mind. And also, more stable over time, more resistant to change, and so on. 

So, when we applied that work to the self, we had people complete self-esteem measure, or self attitude measure, sometimes we’d just use like the Rosenberg Self-Esteem Scale, sometimes we’d adapt standard attitude semantic differential scales, so positive, negative, favorable, unfavorable, in people’s attitudes towards the self. And when we do that approach, we can kind of compare how quickly they respond to those specific attitude items that refer to the self, to how they respond to very similar questions on other topics to get an idea… You know, I think we put in things as mundane as paper plates, Mexican foods, college football, though with college students that might be a highly important attitude that they might respond to very quickly on. 

But we try to see, compared to their baseline level of accessibility of attitudes, how accessible is the self, or the self-esteem specifically. 

Andy Luttrell:

So, just to pause, when you say compared to the baseline, is it actually like a difference? Like if my self positivity comes to mind quicker than how much I love tacos or whatever you said, that’s what I’m talking about? Or is it just sort of like you just control for how fast do I respond to stuff? 

Ken DeMarree:

We actually… This was something that… I forget which paper it was, but Fazio and his collaborators had used this in one of their papers, simply computed a within person Z score, so like across the 10, 12 issues that we had people respond on, where for their mean and their standard deviation response time on this same nine-point attitude scale, where in that distribution does their self evaluation response time fall? 

Andy Luttrell: 

So, like do I respond to myself much more quickly than I respond to other stuff? Much less quickly? Relative to just how I tend to evaluate stuff. 

Ken DeMarree: 

Exactly. Exactly. So, yeah, for each person, on average people respond if I remember correctly reasonably quickly on the self compared to some of those other topics, but you know, if my self-esteem is especially accessible, if I’m especially responding quickly compared to you, even if our levels of self-esteem is different, if I’m responding more quickly I may be more likely to act in line with my self-esteem, for example, than you are. And so, one of the ways that we looked at acting in line with my self-esteem is we gave people ambiguous personality feedback. So, kind of these Barnum statements like the sort of statements you’d see in a horoscope, like, “There are times when you’re this, but at other times you’re this other thing.” 

And people read these and generally find them to be quite accurate, but we also found that people tend to interpret them in line with their preexisting views of the self. So, if people are higher in self-esteem, they saw that information as portraying them in a more positive light compared to if they have lower self-esteem. And so, that’s kind of the general attitude predicting and information processing bias. An attitude, in this case self-esteem, predicting how they interpret this attitude-relevant information. This feedback that’s specifically about them, supposedly. 

But what’s critical is that it’s the people whose self-esteem is more accessible, more active in their mind, they were the ones who were particularly showing that bias, and the people whose self-esteem was less accessible, who were responding less quickly, they didn’t show that bias or didn’t show it as much. 

Andy Luttrell:

So, being able to go from the ideas of accessibility and ambivalence from attitudes to the self, kind of requires you to say that self-esteem is an opinion like any other, which I know is the part of this that can ruffle feathers, especially for people who are self-esteem researchers. And so, I guess the question is to what extent do you think we should say self-esteem, this thing we’re constantly talking about, is an opinion of yourself just like I have opinions of restaurants in my neighborhood, of political candidates, of anything else. 

Ken DeMarree: 

I’m certainly willing to make that statement. Having done so quite a bit already. Yeah, so I think… I don’t know that it’s necessarily very controversial, because if you look at a lot of definitions of self-esteem, they talk about it as an evaluation of the self. Some people do specifically define it as people’s feelings towards the self, so they kind of get at more affective in nature, where a general evaluation we would think is broader than just affect. We’d have cognition, maybe behavior as well as part of the attitude. 

But I think for a lot of definitions of self-esteem, attitude is very clearly related, very clearly relevant. Yeah. So, I don’t think that it’s necessarily controversial, but just because definitionally it’s not controversial, people may still have a hard time believing that this literature is relevant. And I think some of those concerns are valid. You know, if you look at some of our persuasion studies where you’re using completely novel issues that you’re presenting to undergraduates to read about, well, okay, part of a set of factors can lead someone to change their attitude. I’m putting change in scare quotes here because it’s an attitude they may not have had before, because it’s a novel issue. Where this self-esteem, people have an immense amount of relevant information, experience, that feeds into that attitude. 

So, I think in a lot of ways there are additional properties that probably exist on a continuum, but that the self-esteem is at the far end of that continuum in terms of how important it is, how much experience it is, how embedded it might be in a person’s social networks, and social roles, and so on. Because of that, it may not be an easy direct translation of ideas from attitudes to the self always, and I think the same could apply for example to prejudice, where our attitudes towards a social group, like that’s essentially what prejudice is, but they’re often imbued with so much additional stuff beyond what we would often look at in more mundane persuasion studies, for example, in the lab. 

Andy Luttrell: 

To that point, it’s making me think that it’s surprising that anyone would have a weak sense of self at all, right? If there was any issue that followed me around every minute of my life, that I have that immense backlog of information about… You’d say, “Well, I mean, you’ve got the makings of having a strong take on this issue.” Do you have any sense of how common it is that people would have a relatively weak sense of self and why? How do you get there? I guess the ambivalence thing I could see being the road. But the accessibility thing, how do I just like, “Oh yeah, that’s right. I’m here.” 

Ken DeMarree: 

Well, I mean, accessibility if you’re ambivalent, it’s very hard to have an accessible summary evaluation. You know, if the things that come to mind are a mixture of positive and negative, then there’s going to have to be some process that resolves that ambiguity before you can say this is where I stand. And I don’t know in terms of work on accessibility, there’s very little that’s looked at component accessibility, so Newby-Clark has some work suggesting that yeah, when both positives and negatives are accessible, people feel conflicted, but it doesn’t necessarily go the additional step of what are the consequences of having accessible positive and accessible negative? Does both the positive and negative bias ambiguous information? Does both the positive and negative impact your behavior? 

And so, those are things that are less clear, especially with respect to self-esteem. I think we understand a little bit more with attitudes, but probably still not enough. 

Andy Luttrell:

So, it sounds like you’re saying ambivalence is like the road to having a weak sense of self. 

Ken DeMarree: 

Yeah. I think ambivalence is a big part of it, but probably not the only part. So, our self-esteem is also embedded in a social and cultural context that communicates certain messages to us, and especially in the West, those messages are that you’re supposed to like yourself. You’re supposed to love yourself. You know, this is why we get participation trophies, so that we continue to feel good about ourselves. And so, if I don’t feel good about myself, I’m still getting these messages that I’m supposed to feel good about myself. I might adopt them as evaluative goals, like I want to like myself. I want to have a positive opinion towards myself. But I’m faced with a reality that I don’t particularly have a positive towards myself. 

And certainly, in some of my work, we’ve shown that this does predict the experience of ambivalence over and above the presence of positive and negative evaluation. It’s just kind of this desire to be more positive also adds another layer to me feeling conflicted or feeling like I don’t have a clear sense of who I am. So, I think that’s another way that it can be problematic, but ultimately I think ambivalence is a big one, an important one, especially just because it gets at what is inherently the problem, like that there’s so much knowledge and information that we have about the self that it’s sometimes hard to have a clear, coherent, succinct evaluation, and yeah, I think that ends up being a big part of it. 

Now, looking at more specific self views, so we’re talking about self-esteem here, which is very global, but we can think about self evaluations in specific domains, and those domains can differ in a variety of different ways. So, imagine like the musical domain. So, being musical or having… I evaluate my musical abilities as relatively low, but I also view them as unimportant to who I am, like I’m not invested in my low musical abilities. So, I think when you get at more specific evaluations beyond the global self-esteem, then things like importance or knowledge and so on can start to become important, as well, because in more specific domains, I may not have a lot of knowledge about who I am or where I stand in a particular domain, and a particular domain could vary in how important it is to me and so on. 

Andy Luttrell: 

I wanted to go back to the idea of this prescriptive part of self-esteem, that it’s this thing we’re supposed to have, and you go, “You’d better want to feel good about yourself.” Which tees us up to talk about the other work that we talked about talking about, which is-

Ken DeMarree: 

It sounds like cognition, how we talked about talking about it. 

Andy Luttrell:

Oh my God. Yeah, so there’s another direction. But this idea that we can want opinions that we don’t already have, and so the example I have of this is from you had given a talk on this, and I think I reached out to you about this, so you had given a talk. Later that day, I went home, I turned on a TV show that was super popular, and I was just so underwhelmed by just not really liking it, and I caught myself in that moment saying, “I wish I liked this show.” And I went, “Wait a minute. I know someone who knows something about that.” So, what does it mean sort of in general, this idea of desiring attitudes, and where did the idea that that is important come from? 

Ken DeMarree:

I’ll start with the where the idea originally came from was I was doing all of this work that was applying concepts from the attitudes literature to understand the self. And I thought it really should be a two-way street, and so I sat down with Christian Wheeler when he was visiting Columbus, this was towards the end of grad school, and we just kind of had a brainstorming session talking about what concepts from the self literature aren’t really being studied and aren’t really being understood in the attitudes literature. And one of them was just the general idea of self regulation, where there hadn’t been a lot done on attitude regulation beyond kind of like motivated reasoning sorts of things. And so, we didn’t even really have this idea that like do people have evaluative  goals? And so, that was where the original idea came from and so I think we started off to some extent using Higgins’ conceptualization that you have who you actually are, and then you have different standards that you might have, like who you feel you ideally would be, or who you feel you ought to be. 

So, in our very first studies, we just applied those kind of Higgins Selves Questionnaires, I think it is what it’s called, to attitudes. So, whether it was… I just included these as kind of filler measures in other studies, so okay, I need a filler, so we’ll ask people a bunch of questions about abortion and included in those are questions of sometimes the attitudes we actually have might be different from the attitudes that we want to have, and sometimes these might be the same. So, using the scales below, please indicate the actual attitude you have towards abortion. Please indicate the attitude you feel like you ideally want to have towards abortion. And please indicate the attitude you feel you ought to have towards abortion. 

And so, that was kind of the initial starting point, was just looking at first, are people gonna give us different answers to begin with? And you know, we certainly consistently find that they do give us different answers, so obviously we’re trying to pick issues that people may have discrepancies on, but typically the neighborhood of 50% is not at all uncommon. The self is actually one of the more common ones, where sometimes two thirds of participants will report wanting a different attitude. Almost universally they wish they were more positive towards themselves than they actually are, but for other issues you see the opposite pattern, or a mixture of people wanting to be more positive or more negative. 

So, some of the initial work looked at… This is essentially an evaluative conflict. It’s a different one than we normally talk about. It’s not just the presence of positives and negatives in my actual opinion. It’s that I want to be somewhere else. And so, if I’m acting on who I want to be, then that’s gonna pull me or push me in a different direction from how I actually view myself. And so, that kind of creates some evaluative tension, so we’ve found that these discrepancies pretty consistently predict the experience of ambivalence, the subjective ambivalence. Feeling conflicted across a variety of different issues, including the self and self-esteem. 

But you know, then kind of tried to see, well, do people do things that might help them to get where they want to be? So, you know, if I want to like broccoli more, will I engage in things that might help me to actually like broccoli more? So, we’ve never done this with broccoli. We did do it with coffee, at least. Brought people into the lab, told them it’s a taste testing study. The very beginning of the study, we had them fill out a little form, like, “Okay, we’re gonna do a taste test study, but we want to make sure that we prepare the coffee in a way that suits your preferences. Please fill out this little form to say how you want your coffee prepared.” And so, we showed them the 12 ounce cup that we were going to fill up with coffee, and so how many creams and sugars do you want in your coffee, essentially. 

And so, what we were thinking is that if people want to like coffee more, that they might try to change coffee in a way that might make it more enjoyable. So, it’s the same idea like if you want to like exercising more, you might choose to exercise with a friend, because that’s gonna make the activity of exercising more enjoyable. So, if I want to like coffee more, putting sugar and dairy into the coffee might cause me to like it more. I think we had both dairy and non-dairy creamer, so vegans were allowed to make their coffee more palatable, as well. And so, then while the researchers were making the cup of coffee for them, and you know, we had a Keurig machine, so it was like a standardized cup of coffee, that’s when people reported their actual and their desired attitudes, and so, even though we hadn’t made their actual desired attitude salient ahead of time, people who wanted to be… The more people wanted to be positive towards coffee, the more they had somebody doctored up their coffee, the more they had added to the coffee. Especially if they were committed to being more positive towards coffee. 

So, in some of the later work on that topic, we’ve kind of looked at that commitment variable to see is it just a different way… When we’re asking, we’re measuring people’s actual and desired attitudes, are we really just measuring their attitude in different ways? Or is their desired attitude really something that’s kind of goal like? So, the measure of commitment was a way to try to get at that idea, and so the commitment measure predicted stronger influence of people’s desired attitudes, but it didn’t have the same moderating role for people’s actual attitudes, so kind of getting at the idea that this is a goal that people have, or at least the people who are committed to it have.

Andy Luttrell:

The thing about it that seems a little different from the other kind of evaluative conflict, where it’s just there are pros, and cons, and that makes me conflicted, is that I don’t… I still don’t have an end state that I’m looking for other than to understand and to pick a side, right? Whereas with the desired attitude, I go, “I want to go there.” So, I’m looking to move in this direction, whereas regular old ambivalence, I go, “I just… Push me in one direction. I don’t really care. I just want to resolve this idea.” So, one is directional, and one is directionless. 

Ken DeMarree: 

Yeah. I mean, the ambivalence stuff is not necessarily completely directionless. If I’m leaning towards one way, I know it’s easier to go that way than to go the other way. But yeah, these desired attitudes definitely have… Directionality is a much more clear aspect of this. And you know, one of the things that we want to study, we’ve got some ongoing work trying to understand where do these desired attitudes come from, either in terms of purely come from… Sometimes with these things that are embedded in our ideological networks or other things, it may be hard to get at the true cause and effect relationships. But we have some preliminary data at least that my desired attitudes tend to be more consistent with my ideology than my actual attitudes are, for example. 

So, like our ideologies, especially these days, are very much like identities. So, our social identities, our other goals, these are things that we think may be very relevant drivers of our desired attitudes. In the case of coffee, for those undergraduate students wanting to like coffee more could have been for social reasons, like, “Oh, my friends hang out at all the cool coffee shops and I want to enjoy my time with them, so if I like coffee more it would be better.” Or it could be just purely utilitarian, like, “Man, if I drank coffee, I could stay up later at night and cram for my final a little bit better.” But either way, it would be in the service of some goal. 

Andy Luttrell:

I never put this together until now, but there’s research showing, as you know, that if my opinion is different from a loved one’s opinion, that creates this feeling of tension. But I wonder if you’d say that all of what that is is a desired attitude, right? Really, the reason I feel conflicted is because I wish we agreed. Maybe I wish you would come to my side, right? I guess that could be part of it. But it could also be like, “Oh, wouldn’t it be nice if I saw things your way?” That would just make everything go more easily. And that’s what the tension is about. 

Ken DeMarree: 

I mean, that’s a good question. We certainly, in our original paper looking at ambivalence, we did measure this interpersonal ambivalence in at least one study. I think it was the topic was practicing safe sex, so we asked people for the attitudes of their current or most recent romantic partner, because obviously not just leading to ambivalence, like it could lead to actual interpersonal conflict that you and your partner disagree in your attitudes towards practicing safe sex. In that work, we found that objective ambivalence, interpersonal ambivalence, and these actual desired attitude discrepancies independently predicted feelings of conflict. But that’s not to say that interpersonal ambivalence couldn’t be reduced to desired attitudes, because desired attitudes could also be coming from people’s health and safety goals, the pressures that they feel to be an upright citizen, so those desired attitudes may be coming from a variety of sources, one of which could certainly be their romantic partner’s preferences. 

Andy Luttrell: 

Yeah. It strikes me just that these social goals would be a pretty potent predictor of desiring an attitude you don’t already have, right? In a sort of yearning for belonging, right? Like when the world loves a TV show that I think is stupid, you go, “It would be way easier if I liked it because I could talk to…” Or I feel this way about like football, right? I have no real affinity for it, but I know how much small talk I would be able to make if I really did like it, right? Yeah, so these social goals would seem like they would create desired opinions that are distinct from the ones that I have. 

Ken DeMarree:

Yeah. I absolutely agree with that, so yeah, whether it’s individual relationship goals, or broader social goals related to our identities, for example, I think that can be a very powerful driver, and that’s in part why we’re starting with ideology to look at this, also because ideology has more clear set of attitudes that might be related to it, where other social goals you might have to look at in a more idiographic way, rather than rely on these kind of broader normative… Conservatives tend to be more in favor of the death penalty and liberals tend to be more opposed. You know, that way of thinking is a little bit easier for us to look at methodologically. 

Andy Luttrell:

I want to, just by way of maybe wrapping up, I also wanted to ask you about the certainty, your very recent certainty paper, in part because I know it’s… There’s a story to it in that it’s been a sort of a… Maybe not, story is the wrong way to put it, but just that it’s been an evolving project over a long time, and so I’m curious to kind of get your take on where that came from. But also, because as I was thinking about it in the context of asking you about opinion and the self, it sort of strikes me that like… Oh, it on the surface isn’t about the self and isn’t exactly about these other things, but it’s this idea that’s still like me, the person, I’m either a generally confident person, like that’s part of who I am, or I’m the type of person who tends not to see, tends not to feel super confident in the things that I’m thinking. So, could you just sort of give a little bit of a look into why go down the road of looking at certainty as something about a person? And what new insight that brings to our understanding of opinion. 

Ken DeMarree: 

Yeah, so I’ll start off by just kind of describing what the project is, and the project essentially found that there is a disposition, or at least a tendency for people to be certain across a wide range of their opinions. So, you know, if I am certain of my attitude towards canoeing, I’m also likely to be certain of my attitude towards Japan, which presumably are relatively unrelated issues, but if I’m the type of person who tends to be certain in general, I’m likely to be certain of both of those attitudes. 

And as you mentioned, this is a bit of… It was an evolving project. Where it originally started was actually with that self strength work, where we found that people who were reporting that they were certain of their level of self-esteem seemed to be acting certain in a variety of other ways. And I think it took us a while to kind of… The reason it took so long is because we initially found it with self-esteem certainty, I think we initially anchored on that more than we should have and didn’t kind of realize that this seems to be a general evaluative certainty. And so, the version of the paper that is published kind of neglects the earlier kind of sidetrack in our thinking, but from before we were willing to just take a step back and see, think about what we really had. 

But ultimately what we had is that yeah, people who are certain of one attitude tend to be certain of other attitudes, and if you know how certain a person is in their attitudes in general, you can predict how certain they’re going to be in an attitude they haven’t yet formed yet. So, like in that study, we would measure the disposition to be certain just by essentially asking people 16 different issues, their attitude, and their associated certainty towards things like kayaks, and Japan, and cold showers, and a variety of other reasonably unrelated topics. And you know, the certainty in these different attitudes, they hang together, but they also, when participants later read some Amazon.com reviews for a made up microwave oven, their resulting attitude and critically their certainty in that attitude, we can predict from those earlier measures. 

So, if you were certain in your attitude towards Japan, and kayaks, and cold showers, you’re likely to be certain of your attitude towards this microwave oven. Not only that, but you… It seems that people who are certain of their attitudes in general seem to be more likely to act on their attitudes in general. I will admit that the findings were mixed. We found those pretty consistently with undergraduate samples, found them less consistently with a more general range of the adult population. 

And I think part of it is the issues that we were asking about, things like eating meat, so how many servings of meat did you have in the past week, how many cups of coffee did you drink in the past week, these… A lot of these are habitual behaviors, so this is our thinking. We’ll fully admit that it’s post hoc conjecture, at least for now, is that these are things that might be driven by attitudes for someone who is 18, 19, 20 years old, but for your typical 40-year-old like myself, or 40-something like myself, it’s just a habit, like every morning I have coffee. I don’t have to consult my attitudes. So, that’s our thinking, but it’s fascinating that at least for college students, knowing how certain they are in general allows us to predict whether other attitudes will predict corresponding behaviors. 

So, there’s a general tendency for people who are more certain to have their attitudes guiding their behaviors to a greater extent than those people who report less certainty. So, yeah, we found it kind of interesting, fascinating, so found that both with behavioral intentions, so people’s plans to behave in the coming week, as well as not real behavior, but the reports of behavior within the past week. 

Andy Luttrell:

Does that cut across domains? I’m trying to think like your… If I know how much certainty you have about your food preferences, does that tell me how much certainty you have about your political preferences? Right? It sounds like that’s where we’re heading, but I’m just curious, how far do you think that that extends?

Ken DeMarree: 

That’s a good question. I would imagine that there’s going to be some specificity, but the specificity is going to be driven less by your general tendency and more by kind of idiosyncratic factors. And so, the reason that we’re looking at things like kayaks, and Japan, and cold showers, is those are not issues that people have thought a lot about necessarily. Or at least on average, American participants have not thought a lot about those specific issues. Those specific issues may not be tied to their identities and so on. 

So, looking at these dispositions to be certain, we’re not saying that these dispositional factors are all that matter. And so, when you get into specific domains, other things start to matter, so politics and food I think were the two examples you just gave. Well, for someone who’s a foodie, because they’re really interested in food, they may think a lot about topics related to foods. But for someone who’s an NPR junkie, maybe it’s news that fascinates them, politics fascinate them, so they may have thought a lot about politics, and those are not reflective necessarily of their general tendencies. Those are reflective of more specific interests or how much knowledge and expertise they have on a specific issue, so looking at this dispositional work, we’re not saying that the disposition is all that predicts how certain a person will be. That it’s at a default level for things that people don’t have those more idiosyncratic experiences with, or unique ties to identity, or importance, or whatever it might be. That’s where the dispositions are gonna matter most. 

And where those other factors become increasingly important, then issue specific kind of interactions with the topic are gonna matter. 

Andy Luttrell:

Great. I want to be mindful of your time. I feel like I’ve taken more of it than I promised. But I just wanted to say thanks for stopping by to talk about all this stuff and I’ll be curious to see what the next steps are as always. 

Ken DeMarree: 

Thank you very much. I appreciate it. I enjoyed talking with you. Yeah, easily could have talked much longer, but I ran out of water. 

Andy Luttrell: 

All right, that’ll do it for another episode of Opinion Science. Thank you so much to Ken DeMarree for coming on. As always, check out the show notes for a link to his lab’s website and links to the research we talked about. You can support the show by doing all the podcast things. Subscribe to the show, learn more at OpinionSciencePodcast.com, and follow on social media @OpinionSciPod. Ooh, this outro was short and sweet today. I like that. Okay, see you in a couple weeks for more Opinion Science. Bye-bye! 

alluttrell

I'm a social psychologist.

Get in touch