Don’t Think Too Highly of Yourself

A few years ago, in the 2006 Brown Center Report on American Education: How Well Are American Students Learning? researchers found a correlation that went against 40 years of prevailing wisdom in education circles.  When it comes to students and their performance, higher confidence does not go with better math scores.  The report made international comparisons using TIMSS data and concluded:

“countries with more confident students who enjoy the subject matter–and with teachers who strive to make mathematics relevant to students’ daily lives–do not do as well as countries that rank lower on indices of conficence, enjoyment, and relevance.”

It turned out that some of the highest-confidence eighth-graders were some of the worst performers, and the contrary case held as well.  U.S. students rated themselves much more highly than did students in Korea, Japan, Hong Kong, Singapore, the Netherlands, and Chinese Taipei, but they scored well behind that insecure group.  While 93 percent of U.S. eighth-graders failed to achieve an advanced score on the test, only 5 percent of them “Disagreed a lot” with the statement that they “do well in math.”

Another report that just came out suggests that the same discorrelation between confidence and performance may hold in reading.  It appears as “Calibration of reading self-concept and reading achievement among 15-year-olds: Cultural differences in 34 countries” in the September issue of Learning and Individual Differences (volume 19, issue 3, pages 372-386).  Authors Ming Ming Chiu and Robert M. Klassen took data from many countries, examined “self-concept” and “students achievement” in reading, and found that students with under-confidence scored higher than the average in their country, while students with over-confidence scored lower.  This leads them to a firm conclusion.

“Higher reading achievement is linked to not only calibration accuracy but also calibration modesty (i.e., underestimations of performance levels). Students that were modest in their self-appraisals relative to their reading achievement – who were, in fact, underconfident – usually had reading scores that exceeded their country mean while overconfident students often had reading scores below the mean. In this regard, optimistic self-beliefs may signal a developmental lag in cognitive self-awareness.”

While the authors agree that self-confidence may be important for long-term endeavors in which motivation has to carry through from beginning to end, for “specific academic tasks . . . an optimistic or overconfident approach might be dysfunctional.”  While deep-seated doubts can, indeed, be disabling, deep-seated confidence can sometimes “mask real academic difficulties.”  In other words, over-confidence can be a sign not of prior superior achievement, but of inferior achievement, a defense mechanism against poor performance and skill level.

How are teachers in English Language Arts supposed to tell the difference?

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428

For subscription service to the printed journal
Phone (617) 496-5488

Copyright © 2024 President & Fellows of Harvard College