Monday, 25 July 2011

On when it feels good to be under-estimated

Imagine that you're a media mogul and you over-hear people estimating how much of the newspaper market you control. Would it be preferable, do you think, to hear an under-estimate or an over-estimate? Xianchi Dai at the Chinese University of Hong Kong and his colleagues have addressed this very question, producing some findings that they say have practical social lessons for us all.

Based on classic findings in psychology showing that most of us like to see ourselves in a positive light, the researchers said you'd think that on a valued measure, people would always like to hear over-estimates about themselves. Yet the researchers' new studies have demonstrated that the opposite can also be true. In the first, recent graduates in a large Chinese city reported feeling happier after a former class-mate under-estimated their salary. In a second study, business students in the USA said they'd feel happier in a hypothetical scenario in which they heard class-mates under-estimate their GMAT score (a standard admissions test to management school), as oppose to over-estimate it, or guess spot on.

Why should people sometimes like to hear others under-estimate their success? Dai's team propose an explanation based on the precise circumstances. If you don't actually know the answer to whatever valued measure others are estimating about you, then the researchers believe it is always preferable to hear an over-estimate because of the ego-boosting effects this will have. On the other hand, if you do know the precise amount or score about yourself that other people are estimating, then the researchers believe your preference for an over- or under-estimate will depend on whether your priority is (a) creating a good impression or (b) the true answer. For (a) they think an over-estimate will again be preferred because of the advantage for your image, but for (b) they think an under-estimate will be preferred because it has the effect of making the true answer, which is higher, seem more appealing to you.

Dai's team tested these predictions. Over two hundred business students in the USA were asked to imagine that they over-heard colleagues estimating the size of their (i.e. the participant's) annual bonus. As you'd expect, those participants told they didn't yet know the size of their bonus reported feeling happier after hearing colleagues estimate that they were to be awarded a larger amount. For participants told to imagine that they already knew their bonus size ($15,000), their preference for hearing an under-estimate ($3000) or over-estimate ($30,000) depended on whether they'd additionally been told their priority was that they needed extra cash for a property purchase, or that they wanted their colleagues to think they were valued by the boss.

For the participants seeking to impress their colleagues, hearing an over-estimate was preferred. By contrast, for participants saving for a property purchase, hearing an under-estimate was preferred, presumably because it made the true, larger amount seem all the more gratifying.

These findings suggest some rules for the etiquette of guessing, as the researchers explained:
"Imagine that, at a party, you have learned that your friend, Linda, is selling her house, which you believe is worth approximately $500,000. Her friends, including you, are guessing how much she can sell it for. Because Linda does not yet know the truth (the actual sale proceeds), we suggest that you should guess high, if you intend to make her happy. You might say, 'it's such a nice house, I guess you can sell it for $600k.' Now, imagine an alternative scenario, in which Linda has just sold her house, is desperately in need of money, and cares more about the actual proceeeds than others' impressions. In this case you should guess low. Rather than saying, 'It's such a nice house. You must have sold it for $600k,' you should say, 'I'm not sure. Would $400k sound reasonable?' Now Linda can say to herself, 'Wow, I am glad I sold it for $500k ...' and savour the pleasure."

ResearchBlogging.orgShen, L., Hsee, C., Zhang, J., and Dai, X. (2011). The art and science of guessing. Emotion DOI: 10.1037/a0022899

This post was written by Christian Jarrett for the BPS Research Digest.


Rachel said...

This is interesting and the theory is plausible, but I'm not sure that experiment really tests the theory. The task demands are a little too strong for my liking. The researchers explicitly asked participants to imagine specific values. This pretty much tells the participants that they should use these imagined values when answering the question that followed.

I would be more convinced if a follow-up study measured participants' own values (preferably after the crucial question, not before) rather than highlighting 'values' that researchers suspect will be important.

Maybe the study was more subtle; I confess I haven't read it and I'm basing my comments only on the description here, which will necessarily pick out key features.

Anonymous said...

Yah.. quite logical it is! ;)

JerQS said...

This idea is evident when guessing someones age: always underestimate.

Post a Comment

Note: only a member of this blog may post a comment.