Wednesday, 27 February 2008

How do psychologists study what we know about ourselves?

Dr. Virginia Kwan of Princeton University, with the latest in our ongoing series of guest features for students.

One of the most direct ways in which psychologists learn about how people think about themselves is by simply asking people about themselves (e.g. “How smart do you think you are?”). There are many advantages to self-reports in studying self-perception. They are simple, inexpensive – there are no fancy machines or complicated experimental setups – and revealing; after all, who knows you better than you?

Nevertheless, self-reports have their flaws. One problem is that self-reports are subject to social desirability concerns, making them vulnerable to misreporting. When people know that someone else is going to hear their response to a question, they may change their answer, even unknowingly. Another issue concerning self-reports is whether people are consciously aware of their self-perception and whether they are able to report it accurately.

No single measure is perfect, which is why psychologists often use both self-reports and implicit measures. Implicit measures are designed to be free of social desirability concern by tapping into unconscious aspects of self-perception. For example, during the Rorschach inkblot test, people are presented with ambiguous inkblot images and are asked to interpret them. The interpretation of ambiguous stimuli is thought to reflect personality characteristics and emotional states (see also prior Digest item on implicit test of attitudes).

Why should we care about self-perception? Psychologists have studied self-perception extensively because many believe it is essential for human functioning. One question that has endured, however, is whether we are better off seeing ourselves accurately or through a rose-coloured glass (see Block & Colvin, 1994; Sedikides et al pdf; Taylor & Brown, 1988 pdf). Recent research suggests that overly positive self-perception, known as self-enhancement, may be a mixed blessing for mental health (Bonanno et al; Kwan et al pdf; Paulhus, 1998 pdf).

But overly positive compared to what? Self-perception is an inherently social phenomenon. The way we see ourselves and the ways we are seen by others are closely intertwined. To examine self-enhancement, my colleagues and I (pdf) asked study participants to rate themselves and each other on personality attributes following a group interaction. Comparing all of these ratings allowed us to study the effects of self-enhancement by taking into account the ways people perceive others as well as how they are perceived by them. Our findings suggest that seeing oneself in an overly positive light compared to the social reality leads to maladjustment, but on the other hand, seeing oneself more positively than we see others leads to higher self-esteem and other intrapsychic benefits.


  1. Regarding self-enhancement...there is a story about Al Maguire the late, former basketball coach of Marquette University and color commentator for NCAA basketball. The story is as follows:
    Al was speaking to reporters about his basketball team and had his son playing for him. His son was one of the starting 5. One of the reporters asked him why his son was starting in front of another player because most of the writers felt the other player was "a little bit better" than his son.

    Al replied with a sly smile:
    "If he is going to start ahead of my son, he has to be ALOT better!" So it is with self enhancement. If it is only a little bit of distortion in your favor, that is a good thing; a lot would be a problem.

  2. Anonymous2:47 pm

    I can't believe you actually mentioned the Rorschach test as a research instrument. That test has never been scientifically demonstrated valid nor reliable and has no business being used in scientific research. Also, not all self-reports are prone to social-desirability. For the most part, when people are asked about very personal and socially undesirable habits (for example, drug use, sexual behavior), they are surprisingly honest. Sure there's some under-reporting, but not as much as one would think. Lastly, in any kind of global self-report measure, smart researchers would include a lie scale to see if each respondent is reliably reporting on him or herself.

  3. Yes, Anonymous, on Rorschach (or other storytelling with little prompt) you´re right.
    But WHICH lie scale?
    Have those tests which ask "Do you believe in god" and rate answeing "No" a liar being changed recently?
    And those about "sinful thoughts"?
    When I do not believe in the religious concept of sin I am lying?????