Contrary to popular belief in psychology, there is no relationship between self-insight — how accurately you can judge your own abilities — and certain areas of adjustment, indicated by measures such as life satisfaction. These potentially groundbreaking findings are from a recent U of T study published in Nature Human Behaviour.

Various competing perspectives exist in psychology about the relationship between self-insight and adjustment. Self-insight refers to how well people’s self-view, or beliefs about their levels of ability, match with their actual levels, while adjustment is essentially how well people function in life.

Self-insight is valued in institutions such as schools and workplaces, where individuals may be given feedback on their work, and are often encouraged to reflect on their strengths and weaknesses to improve performance.

But is it really best for your life satisfaction to accurately know your levels of ability? Could overestimating yourself and ‘self-enhancing’ boost confidence and be more beneficial? Or would another combination of high abilities and accurate self-views be optimal?

These are the types of questions that inspired coauthors Joyce He, a PhD candidate at U of T, and her advisor, Professor Stéphane Côté, to begin this study nearly two years ago.

Results do not support existing theories

He and Côté tested five main competing perspectives. The first, the self-insight perspective, proposed that adjustment is highest when self-views and abilities match. This enables individuals to perform confidently in their strengths while being aware of their weaknesses.

The second, the optimal margin of illusion perspective, suggests that regardless of abilities, adjustment is highest when positive self-views exceed ability by a certain amount. This provides enough confidence to motivate individuals while remaining realistic to their actual abilities.

The third and fourth perspectives posit that only positive self-view or only high abilities are related to high adjustment, while the other variable is irrelevant. These are named the positive self-views-only, and high abilities-only perspectives, respectively.

The fifth perspective proposed that the relationship between abilities and self-views is non-existent — rather, both variables are independently related to higher adjustment. Named the positive self-views and high abilities perspective, the hypothesis suggests that one’s abilities and accurate self-views of one’s abilities are both positively related with adjustment, after controlling for one another.

But all five perspectives were unsupported by the co-authors’ high-powered study.

An implication of the lack of support for all five perspectives, according to the co-authors, would mean that providing feedback to students and workers about their cognitive and emotional abilities — or enhancing their self-views — may not enhance their adjustment.

The many competing perspectives on self-insight and adjustment exist mainly due to two limitations of past research.

First, some studies determined self-insight by assessing how well people’s self-views matched with peers’ perceptions of them — which can be biased or inaccurate — or with their views of other people. This is also problematic, because discrepancies between one’s ratings of one’s own abilities and others’ abilities might be due to actual differences between people’s abilities, and not a lack of self-insight.

A second limitation stemmed from how past researchers used the difference between self-view and ability, the square of their difference, or other measures, in statistical analyses. These “difference scores” can conceal information by merging variables, and the correlations found can interpreted in various ways.

How the researchers overcame these two limitations

Using an online recruitment source, He and Côté surveyed 1,044 participants from the United States. This large sample size was calculated to provide high statistical power, at 95 per cent.

“Statistical power is essentially how much power you have to detect an effect, if it is there,” He explained to The Varsity. The sample size and size of the effect are two important factors affecting power — the number of participants should be large enough to detect the effect being studied.

To address the first limitation, the researchers measured the abilities objectively through a timed test of emotional ability. The test required participants to identify the emotions expressed by 72 photos of actors with different facial expressions, as well as a 20-minute cognitive ability test with 15 perceptual problems.

Self-views were measured by asking participants to rate how they think they scored on these tests. The researchers measured levels of psychological, interpersonal, and institutional adjustment by requiring participants to rank aspects of their life satisfaction, quality of relationships, and career satisfaction, respectively, in a daily diary format.

To reduce error from different types of biases, these measurements were collected over the span of a week. If a participant was in a bad mood, for example, it might have affected their responses. But by taking multiple measurements at different times, the data would be more representative of the participant’s general situation.

To overcome the second limitation, He and Côté analyzed their data in a new way, using polynomial regression to model the data and response surface analysis to generate a three-dimensional plot, showing every possible combination of abilities and self-views, and their relation to adjustment.

Analyzing the response surface graphs revealed that they did not meet the conditions required to support any of the five hypotheses. They did notice some patterns that may support self-enhancement, a relatively new perspective not included in the hypothesis, which posits that individuals whose self-views exceed their abilities will be better adjusted.

This perspective is somewhat counterintuitive, as it predicts that individuals with low abilities and high self-views would be the most adjusted, while individuals with high abilities would be less adjusted because their self-views cannot exceed their abilities by as much.

“One possibility here is that these self-enhancers are rating everything on a higher level,” said He. “So they’re rating their abilities higher, their self-views about their abilities are higher, but they’re also rating their life-satisfaction, career satisfaction, and relationship satisfaction higher.”

Study was published as a Registered Report aimed for transparency

He and Côté’s study is one of the first two Registered Reports published in Nature Human Behaviour. Traditional papers are submitted to journals after the study has been completed, whereas Registered Reports have researchers submitting their introduction, proposed methods, and plans for analysis before conducting the study.

Following the submission, reviewers and editors then consider the proposals and can make suggestions. If a proposal is approved, researchers conduct the study, with guaranteed publication of their results — significant or not.

Registered Reports are part of a push for greater transparency in psychological research, according to He. This design can help studies that may face difficulty with publication if they produce non-significant results.

“With the Registered Report, the editors are really trying to put more emphasis on the research questions that you have,” He said. “A lot of authors, they might have this really important question, [and] they [design a study to] test it.”

“But [if] they find null results, then they [may not] actually end up publishing [them],” she continued. “Because in our field, at least, you’re kind of incentivized to publish interesting results.”

Future steps following the study

Next steps could include studies designed to confirm evidence supporting the self-enhancers perspective. These might measure adjustment with variables that are not self-reported, such as peer opinions and objective performance at work.

In terms of potential applications of these findings to policy, education, or management, He believes that more research needs to be done.

“Once we see from a few studies, or a bunch of studies, that we see the same patterns over and over again, then I think that that’s when we can actually draw the conclusion.”

Editor’s note (October 18, 2019, 6:15 pm): The article has been updated to reflect that the fifth hypothesis, which contends that the relationship between abilities and self-views is non-existent, is based on existing research literature.