Humblebragging: A Distinct – and Ineffective – Self-Presentation Strategy

Humblebragging – bragging masked by a complaint – is a distinct and, given the rise of social media, increasingly ubiquitous form of self-promotion. We show that although people often choose to humblebrag when motivated to make a good impression, it is an ineffective self-promotional strategy. Five studies offer both correlational and causal evidence that humblebragging has both global costs – reducing liking and perceived sincerity – and specific costs: it is even ineffective in signaling the specific trait that that a person wants to promote. Moreover, humblebragging is less effective than simply complaining, because complainers are at least seen as sincere. Despite people’s belief that combining bragging and complaining confers the benefits of both self-promotion strategies, humblebragging fails to pay off.


Humblebragging: A Distinct -and Ineffective -Self-Presentation Strategy
Thanks to social media, we are frequently confronted with a particular type of selfpromotion: "Hair's not done, just rolled out of bed from a nap and still get hit on, so confusing!"and "Graduating from 2 universities means you get double the calls asking for money/donations.So pushy and annoying!"Such instances of "humblebragging" allow actors to highlight positive qualities while attempting to appear humble by masking it in a complaint; in the second example above, note how the brag (I have two university degrees) is couched in a complaint (getting calls is annoying).While people humblebrag to make a good impression on others without appearing vain, we suggest that humblebragging frequently fails.In fact, because observers find the strategy insincere, humblebraggers are less likeable than those who straightforwardly brag -or even those who simply complain.

Humblebragging to Manage Others' Impressions
An extensive body of research demonstrates people's motivation to manage the impressions they make (Jones & Pittman, 1982;Leary & Kowalski, 1990).People wish to be viewed positively (Baumeister, 1982;Schlenker, 1980) and attend closely to how they present themselves in social interactions (Goffman, 1959).A commonly used impression-management strategy is self-promotion, which allows individuals to bring their good qualities to others' attention (Jones & Pittman, 1982;Leary et al., 1994).
At the same time, successful impression management can be a balancing act.Because modesty is a highly valued quality (Ben-Ze'ew, 1993;Schneider, 1969;Wosinka, Dabul, Whetstone-Dion, & Cialdini, 1996), self-promotion efforts such as bragging can backfire: people who brag may be perceived as conceited (Powers & Zuroff, 1988;Tice, Butler, Muraven, & Stillwell, 1995).Therefore, people often seek to present their qualities and accomplishments Humblebragging 4 indirectly (Schlenker & Weigold, 1992); for instance, people may glorify the accomplishments of and give credit to others (Cialdini, Finch, & DeNicholas, 1990;Tetlock, 1980).We suggest that humblebragging is an understudied yet ubiquitous indirect strategy that attempts to mask a brag in the guise of a complaint.Used correctly, complaining can be an effective means of eliciting sympathy and attention from others (Alberts, 1988;Alicke et al., 1992), such that combining bragging and complaining may offer a "sweet spot" for self-promotion.
We suggest, however, that humblebragging is ineffective.Prior research has shown that the success of an impression-management strategy depends on whether actors are able to mask their ulterior motive of being viewed positively (Eastman, 1994;Giacalone & Rosenfeld, 1986;Jones & Pittman, 1982) -and complaining, used incorrectly, has just such potential to backfire (Kowalski, 1996).When observers believe that actors' primary goal is making a favorable impression, such actors are considered insincere; perceived sincerity is a critical factor in determining the success of self-promotion (Crant, 1996;Nguyen, Seers, & Hartman, 2008;Turnley & Bolino;2001).As a result, we hypothesized that humblebragging -attempting to mask a brag with a complaint -generates negative impressions because the strategy seems insincere.

Overview of the Research
We test our predictions in five studies.We first asked raters to evaluate tweets from a dataset of humblebrags from Twitter, assessing whether the more raters view a tweet as humblebragging, the less they liked the humblebragger (Study 1a).In Study 1b, we analyzed answers to the question "What is your biggest weakness?" in job interview contexts to document the commonality of humblebragging, and its effectiveness in the eyes of observers.In Study 2, we explore the mechanisms underlying humblebragging, testing whether humblebraggers are liked less than complainers and braggers because they are seen as less sincere.We then examine whether humblebragging may even give a weaker impression of a desired trait than straightforward bragging (Study 3), and whether people's dislike of humblebraggers extends to less generous behavior toward them (Study 4).

Study 1a: Humblebragging on Social Media
We first examined humblebragging in the communication channel where it seems most ubiquitous: online (Alford, 2012;Buffardi & Campbell, 2008).People employ a wide array of strategies to construct a positive image online (Lampel & Bhalla, 2007;Schau & Gilly, 2003).
We analyzed a dataset of statements categorized as "humblebrags" on Twitter, predicting that the more a tweet is perceived to be a humblebrag, the less liked and the less sincere the humblebragger would be.

Method
Procedure.We constructed our dataset of humblebrags using a web page (http://twitter.com/Humblebrag)that lists tweets categorized as humblebrags between June 2011 and September 2012 for the book Humblebrag: The Art of False Modesty (Wittels, 2012).This resulted in a dataset of 740 tweets; 68.4% were made by males (seven tweets lacked gender information).Examples include "It's been 10 years but I still feel uncomfortable with being recognized.Just a bit shy still I suppose," and "Graduating from 2 universities means you get double the calls asking for money/donations.So pushy and annoying!"Two independent raters evaluated each statement based on its text alone, without receiving any additional information about the tweeter.Raters answered the following questions for each statement: (1) "How much do you like the person who shared this tweet?"(2) "How competent do you find the person to be?" and (3) "How sincere do you find the person to be?" all on 7-point scales (1 = not at all, 7 = very much).The two raters also assessed the statements on the following dimensions using the same 7-point scale: (1) "To what extent do you think this person is complaining?"(2) "To what extent do you think the person is bragging?"and (3) "To what extent do you think the person is humblebragging?"For the last question, we provided raters with our definition of humblebragging: showing off in the guise of complaint.Because the inter-rater reliabilities for each question were high (ranging from .85 to .94),we averaged the two ratings for each item.

Discussion
The negative correlations between liking, perceived sincerity, and perceived competence with ratings of humblebragging suggest that despite their efforts, these Twitter users were unsuccessful in using humblebragging to self-promote.These findings offer initial evidence that the more individuals are perceived to use humblebragging the more they come across as less likeable, sincere, and competent.In Study 1b, we explore the ubiquity and effectiveness of humblebragging in another paradigmatic impression-management situation: job interviews (Stevens & Kristof, 1995).

Mean
Interviewees faced with the common question, "What is your biggest weakness?"are torn between admitting a real weakness or humblebragging (e.g., "It's hard for me to work on teams because I'm such a perfectionist.")Study 1b had three primary aims.First, we wanted to document how common humblebragging is in these situations, assessing how many people chose to respond with a humblebrag rather than an actual weakness.Second, we hoped to show that people who engage in humblebragging do so because they believe it is an effective strategy.
Finally, we tested whether this strategy would pay off, predicting that observers would prefer to hire people who confessed to a real weakness than those who attempted to strategically humblebrag.

Method
Participants.One hundred and twenty-two students from a university in the Northeastern United States (67% female, M age = 21.34,SD = 1.80) enrolled in an online study in exchange for a $5 Amazon Gift Card.To ensure that participants paid attention during the study, we included comprehension filter questions, all of which participants passed.Prior to beginning data collection, we targeted a recruitment of approximately 100 individuals based on previous pilot surveys.
Design and procedure.First, participants were instructed to write in detail how they would answer the question "What is your weakness?" in a job interview context.Second, we asked them to explain the reason for their response: "Why would you answer the question 'What is your weakness?' in this manner?"Both questions required open-ended responses.Finally, participants answered demographic questions.
We recruited two independent coders -blind to our hypotheses -to analyze the content of participants' open-ended responses.We asked the coders to categorize participants' responses as either humblebragging, in which they talk about a positive quality but frame it as a weakness; or as a real weakness, in which they talk about an objectively negative characteristic.Coders agreed 97% of the time and resolved disagreements through discussion.
Next, coders independently rated responses to the following questions on 7-point scales: "To what extent do you think this person is humblebragging?"(1 = Not humblebragging at all, 7 = Definitely humblebragging) and "To what extent do you think this person is bragging?"(1 = Not bragging at all, 7 = Definitely bragging).Interrater reliability was high both for humblebragging and bragging ratings (αs = .86and .85).We averaged ratings to create humblebragging and bragging scores.
Coders also analyzed participants' answers to the question "Why would you answer the question 'What is your weakness?' in this manner?"In particular, the coders coded for whether participants indicated that they were being honest (e.g., "This is really my weakness") or strategic (e.g., "I want to get hired").The coders agreed 98% of the time and solved disagreements through discussion.Coders also categorized participants' responses in terms of the main characteristics they mentioned, such as being a perfectionist, working hard, being nice and helpful, being honest, other positive characteristic, or a real weakness.
To analyze the extent to which these answers affect the likelihood of being hired, we recruited two additional independent research assistants, also blind to the hypotheses, and had them rate each response on: "To what extent would you want to hire this person?"on a 7-point scale (1 = not at all, 7 = very much).Reliability was sufficiently high (α = .71);we averaged ratings to create a composite measure of likelihood of being hired.

Results
Decisions to humblebrag.As we expected, the majority of participants chose to humblebrag rather than disclose an actual weakness: 77% of participants' responses to the question "What is your weakness?"were coded as a humblebrag and just 23% as a real weakness, χ 2 (1, N = 122) = 35.71,p < .001.The most common humblebrags centered on being a perfectionist (32.8%), working too hard (24.6%), being too nice and helpful (14.8%) and being too fair and honest (4.9%).The 23% of participants who reported real weaknesses named a range of actual negative attributes (see Table 2 for examples.)

Humblebrags
Real Weaknesses I am a perfectionist at times, it is so hard to deal with.
I'm bad at saying no.I always get all of my work done, but I find myself doing a lot of favors for others.
My inability to be not nice to coworkers.
I'm too demanding when it comes to fairness.
Sometimes I am so confident that I'm right that I don't allow others to work through their opinions.
I'm not always the best at staying organized.
Sometimes I overreact to situations.
My biggest weakness is that I sometimes tend to procrastinate, which leads to work piling up towards the end of deadlines.

Effectiveness of humblebragging.
Humblebragging is thus both ubiquitous and believed -by participants -to be a strategic decision.But did this strategy pay off?Coders rated humblebrags as higher in both bragging and humblebragging (Ms = 5.94 and 4.86, SDs = 1.37 and 1.59) than real weaknesses (Ms = 3.50 and 2.86, SDs = 2.24 and 2.01), ts(120) = 7.05 and 5.49, ps < .001.Perhaps most importantly, the second set of coders -who simply read the answers and indicated their likelihood of hiring the candidate -were less interested in hiring humblebraggers than candidates who had reported an actual weakness (Ms = 3.03 and 3.93, SDs = 1.34 and 1.49), t(120) = 3.03, p < .01.

Discussion
Study 1b shows that humblebragging is ubiquitous not only on social media but also in job interviews.More than 70% of participants humblebragged in response to a question about their greatest weakness, and more than 60% reported doing so because they believed the strategy would boost their chances of getting hired.Independent raters, however, preferred honest candidates who reported a real weakness.Study 1b offers a strong test of the negative effects of humblebragging: raters preferred people who had confessed a true workplace liability to humblebraggers.
Our definition of humblebragging -bragging in the guise of complaint -suggests that people believe that by combining bragging with complaining, humblebragging dominates both as a self-presentation strategy.In Study 2 we tested the relative efficacy of these two strategies against humblebragging.We predicted that humblebrags would be less effective at inducing liking than both complaints and brags because these are at least perceived as sincere.We Humblebragging 12 assessed perceived sincerity as a meditator of the relationship between humblebragging and liking.

Method
Participants.Three hundred and two participants (M age = 36.97,SD = 11.84;41.5% female) recruited on Amazon's Mechanical Turk participated in this study in exchange for $.50.
We included several comprehension checks to ensure that participants paid attention and eliminated five participants who failed these checks.We recruited our sample size based on an estimate effect size of f = .1 and 80% power.
Design and procedure.Participants were told that they would be evaluating another person.All participants were randomly assigned to one of three conditions -humblebrag, brag, or complain -in a between-subjects design.Participants in the humblebrag condition viewed the following statement from the target: "I am so bored of people mistaking me for a model." Participants in the brag condition viewed the brag portion of the humblebrag, "People mistake me for a model", while participants in the complain condition viewed the complaint portion, "I am so bored."The humblebrag was in the Twitter dataset from Study 1a; we created the brag and complain versions.
After viewing one of these statements, participants rated how much they liked the target on a 7-point scale (1 = not at all, 7 = very much).Then, participants answered a two-item measure of perceived sincerity on a 7-point scale (1 = not at all, 7 = very much): "How sincere do you think this person is?" and "How credible do you think this person is?" (α = .92;Chan & Sengupta, 2010).The order of liking and sincerity questions was counterbalanced; order did not impact our results.
Next, as manipulation checks, participants rated the extent to which they thought the person was complaining, bragging, and humblebragging on 7-point scales (1 = not at all, 7 = very much).Finally, participants answered demographic questions.

Results
Table 3 provides means for all dependent measures by condition.
Confirming our definition of humblebrags, ratings of complaining were higher in the humblebrag condition than in the brag condition (p < .001).
Liking.As predicted, an ANOVA revealed a significant effect on liking, F (2, 298) =15.02, p < .001,η 2 = .09.As depicted in Figure 1, participants in the humblebrag condition liked the target less (M = 2.62; SD = 1.36) than did participants in the brag condition (M = 3.11, Humblebragging 14 SD = 1.24; p < .02)and in the complain condition (M = 3.57, SD = 1.06; p < .001).Participants in the complain condition liked targets significantly more than participants in the brag condition (p = .02).Participants in the brag condition rated targets as less sincere than participants in the complain condition, p < .001.Mediation.To examine whether sincerity mediated the effect of humblebragging on liking, we followed the steps recommended by Baron and Kenny (1986).The first and second criteria specify that the independent variable should significantly affect the dependent variable and the mediators.The prior analyses showed that these two criteria were met, as humblebragging had a significant effect on liking and sincerity.To assess the third and fourth criteria, we conducted a hierarchical ordinary least-squares (OLS) regression analysis (including a dummy variable for the bragging condition) predicting liking from the independent variable of the mediator should significantly predict the dependent variable while controlling for the independent variable.The results met this criterion: controlling for the humblebragging condition and the bragging condition, we found that sincerity significantly predicted greater liking (β = .67,t = 12.995, p < .001).
To complete the test of mediation for sincerity, the fourth criterion holds that the effect of the independent variable on the dependent variable should decrease after controlling for the mediator.After controlling for sincerity, the effect of humble bragging on liking decreased significantly (from β = -.35,p < .001 to β = .05,p = .44).To test whether the size of the indirect effect of humblebragging on liking through sincerity differed significantly from zero, we used a bootstrap procedure to construct bias-corrected confidence intervals based on 10,000 random samples with replacement from the full sample (Preacher & Hayes, 2004).The 95% biascorrected confidence interval excluded zero (-1.41, -.79), indicating a significant indirect effect.

Discussion
Individuals who humblebrag -couching a brag in a complaint -are viewed more negatively than those who straightforwardly brag or complain.Moreover, as in Study 1a, insincerity plays a mediating role: while people do not love braggers or complainers, they at least see them as more sincere than humblebraggers, such that perceptions of insincerity drive lower ratings of humblebraggers.Including sincerity in the model significantly reduced the effect of humblebragging (from β = -.20,p = .005,to β = .01,p = .76),and perceived sincerity was a significant predictor of liking (β = .80,p < .001).A 5,000-sample bootstrap analysis revealed that the 95% biascorrected confidence interval for the size of the indirect effect excluded zero [-.97, -.31], suggesting a significant indirect effect (Baron & Kenny, 1986;Preacher & Kelly, 2011).

Discussion
These results suggest that humblebragging is not only costly in terms of general liking and perceived sincerity, but also is less effective than bragging at signaling the single trait one wishes to convey.Thus humblebragging is ineffective as an impression-management not only due to the lower levels of liking and perceived sincerity it induces, but also its failure to communicate specific qualities.

Study 4: The Behavioral Costs of Humblebragging
In Study 4, we aimed to understand whether the costs of humblebragging extend beyond interpersonal evaluations and influence behavior, causing individuals not only to dislike humblebraggers but also to treat them less positively.We predicted that people would allocate less money to humblebraggers than to braggers when given $5 to split in a dictator game.
Replicating the previous studies, we again predicted that perceived sincerity would drive lower levels of liking -which in turn would lead to less money allocated.

Method
Participants.The study employed two phases.One hundred and fifty-four individuals (M age = 33.26,SD = 9.36; 35.1 % female) recruited on Amazon's Mechanical Turk participated in the first part of the study in exchange for $.50.We included several comprehension checks to  "I am the know-how person at work.People come to me first." "I am really qualified for the job market."  "I got accepted to all the top schools."  "I get elected to leadership positions all the time."  "People frequently tell me that I look like a celebrity." "People keep telling me how cute I am." Participants in the second part of the study were all assigned to the role of Player B and were informed that they would allocate $5 between them and Player A from another session.
They were randomly assigned to one of the two conditions in a between-subjects design, such that they played the dictator game with an individual who either sent humblebragging messages or bragging messages.After reading the messages, participants allocated $5, then rated how much they liked Player A and Player A's sincerity (α = .70),using the same measures from Studies 2 and 3. Finally, participants answered demographic questions.

Humblebragging 22
Mediation.A path analysis revealed that perceived sincerity and liking mediate the relationship between condition and allocation (Figure 3).Higher perceived sincerity led participants to like their partner more, which led to higher allocation amounts in the dictator game.When we included perceived sincerity in the model, predicting liking, the effect of condition was reduced (from β = -.20,p = .013,to β = -.07,p = .33),and perceived sincerity was a significant predictor of liking (β = .49,p < .001).The 95% bias-corrected confidence interval for the size of the indirect effect excluded zero [-.56, -.14], suggesting a significant indirect effect.When we included perceived sincerity and liking in the model, predicting allocation, the effect of condition was reduced (from β = -.17,p = .034,to β = -.07,p = .35),and both perceived sincerity (β = .20,p = .029)and liking (β = .22,p = .014)were significant predictors of allocation.The 95% bias-corrected confidence interval for the size of the indirect effect excluded zero [-.14, -.01], suggesting a significant indirect effect (Baron & Kenny, 1986;Preacher & Kelly, 2011).

Discussion
These results demonstrate that the costs of humblebragging extend beyond interpersonal evaluations, impacting behavior.Humblebraggers are seen as insincere, leading them to be less liked, which results in stingier treatment.As in the previous studies, straightforwardly bragging produces better outcomes than humblebragging.

General Discussion
Both correlational and causal evidence demonstrates that humblebragging fails to leave a favorable impression on others.Although people frequently resort to humblebragging when attempting to impress others (Studies 1a and 1b), the strategy is less effective than simply bragging or complaining in producing either global or specific desired impressions (Studies 2 and 3), while also garnering more negative behavioral responses (Study 4).Across the studies, perceived sincerity is a key predictor of liking: the disingenuousness signaled by humblebragging manifests in dislike.
Taken together, our results shed light on the common phenomenon of humblebragging, an increasingly ubiquitous strategy utilized by people attempting to self-promote.Although a large body of prior research has documented different impression-management strategies, humblebragging is a previously unexplored -and, as we show, uniquely ineffective -form of self-praise.While previous research has identified other indirect means of self-promotion, such as praising close associates (Cialdini et al., 1990;see Schlenker & Weigold, 1992), we document a new type of indirect speech that does not divert attention to other people bur rather attempts to divert attention from the bragging nature of the claim -via a complaint.
This research makes several contributions to the literature.First, we introduce and examine the psychology underlying a unique impression management tactic, adding to the Humblebragging 24 literature on impression management (e.g., Arkin, 1981).Second, we shed light on the pivotal role that perceived sincerity plays in impression management.Specifically, we show that sincerity plays a critical role in determining the success of three seemingly different selfpromotion strategies: humblebragging, compared to bragging or complaining, is perceived as insincere -the critical factor that causes humblebragging to fail.Third, our research contributes to research on indirect speech; prior research has shown that indirect speech and indirect selfpromotion attempts are more likely to lead to favorable judgments when they occur naturally and spontaneously (Holtgraves, 1986;Tal-Or, 2010).Our research shows that in contexts where this naturalness and spontaneity are absent -such as in job interviews (Study 1b) -indirect attempts can backfire.
While our research focused on the reactions of recipients of humblebragging, future research should examine the emotional experiences of humblebraggers.For example, previous research has shown that self-promoters experience positive emotions while bragging (Scopelliti, Loewenstein, & Vosgerau, in press).Humblebragging may constitute a particularly miscalibrated case in which humblebraggers experience positive affect from bragging and the positive feeling that they are not actually bragging, while recipients react negatively to both the self-promotion and the attempt to mask it.

Conclusion
The proliferation of humblebragging in social media and everyday life suggests that people believe it an effective self-promotional strategy.Yet, our results show, people readily denigrate humblebraggers.Faced with the choice to (honestly) brag or (deceptively) humblebrag, would-be self-promoters should choose the former -and at least reap the rewards of seeming sincere.

Figure 1 .
Figure 1.Liking and perceived sincerity by condition in Study 2.

Figure 2 .
Figure 2. Dependent measures by condition in Study 3.