By Daniel Goleman
This is the first of two columns by Daniel Goleman in response to the NurtureShock blog.
Po Bronson and Ashley Merryman are first-rate journalists, and I’m sure NurtureShock is a wonderful book (I haven’t had a chance to see it yet). But in their NEWSWEEK blog the authors have misstated several of my positions. So for the record, let me begin to set the record straight by quoting from my forward to the 10th-anniversary paperback edition of Emotional Intelligence, where I write about one myth “widely repeated: the fallacy that ‘EQ accounts for 80 percent of success.’ This claim is preposterous.”
In the forward I go on to explain that the misinterpretation stems from estimates that IQ accounts for around 20 percent of success in careers. This leaves 80 percent unaccounted for. But this does not mean emotional intelligence explains the rest of career success. As I wrote in Emotional Intelligence, a wide range of elements, from family wealth and education, to simple luck—including emotional intelligence to some degree—and many more factors are at play. Malcolm Gladwell’s recent book Outliers argues for chance opportunity as one such.
“Another common misconception,” I wrote in the forward, takes the form of recklessly applying the importance of emotional intelligence to domains where it matters far less than IQ—academic achievement being the most obvious. When it comes to career success, the picture is more nuanced. IQ scores are the best predictor of what career rung we can manage. That’s what they were designed to do; IQ tests were first widely applied in sorting into the right specialty and rank millions of American soldiers during World War I.
But when it comes to predicting who among a talented pool of candidates within an intellectually demanding profession will emerge as an effective leader, IQ loses is predictive power. This is partly due to the “floor effect,” where in order to enter the top echelons of a given profession or large organization everyone has already been sifted for IQ. At those levels a relatively high IQ becomes a threshold ability—what you need to enter and stay in the game.
In my 1998 book Working With Emotional Intelligence I proposed that EI-based abilities more often than IQ-type abilities or technical skills are the discriminating competencies that predict who among a group of very smart people will lead most ably. This is a key point for anyone running an organization who must decide what abilities to look for in potential leaders. One methodology used in industrial/organizational psychology to make this judgment is called “competence modeling,” which contrasts highly effective leaders with mediocre ones, and determines what specific abilities the stars display that the others lack.
Organizations around the world use the competence-modeling method to make personnel decisions, performing independent analyses of their own employees. As I wrote in the forward, if you scan these competency models, “you discover that IQ and technical skills drop toward the bottom of the list the higher the position” (though they remain stronger predictors of excellence in lower-rung jobs). Competence models for leadership typically consist of anywhere from 80 to 100 percent EI-based abilities.”
I inadvertently may have added to the confusion about EI as a factor in success when I summarized this competence data in ways that were misconstrued as claiming that EI (I generally don’t use the term “EQ”) is more powerful in predicting career success than IQ. Once I realized that people did not understand the limited context and correct basis of this statement, I gave more qualifying information. Still, some press accounts and other secondary sources continue to misrepresent my views, as Bronson and Merryman have done here.
Another point relates to the contrast between executive function and emotional intelligence. Bronson and Merryman seem to say that executive function and emotional intelligence are in some kind of competition as concepts. Actually I would argue they are partly overlapping constructs. My model of emotional intelligence elaborates four domains of ability: self-awareness, self-regulation, empathy, and interpersonal skill. The first two, self-awareness and self-regulation, are themselves elements of executive function, all of which are based in the operations of zones of the prefrontal cortex. Indeed, in writing about self-regulation in my 1995 book Emotional Intelligence, I cite the work of Walter Mischel and his now-famous marshmallow test with 4-year-olds, which assessed their ability to manage impulsivity and delay gratification—two key indicators of executive function. I would expect executive function and emotional intelligence (at least as described in my model—perhaps not with Salovey and Mayer’s) to correlate strongly with EF. That’s a question for further research.
Bronson and Merryman also misrepresent curricula in social/emotional learning as a waste of time. An article by University of Illinois psychologist Roger Weissberg and colleagues at the Collaborative for Social and Emotional Learning, now in press in the journal Child Development, reports on a meta-analysis of more than 200 studies comparing students who had the program with those who do not. The results, as presented in an earlier version of this paper: the programs reduce violence and other antisocial behavior by around 10 percent and enhance positive behaviors like paying attention in class by the same margin—and benefits are greatest in schools that need it the most. Most intriguing, academic achievement test scores go up around 11 percent. That sounds like a program anyone would want their children to benefit from.
Another odd notion put forward by Bronson and Merryman is that Peter Salovey represents the academic side of emotional intelligence and I represent the commercial side. I do not sell any product or service related to EI. The sole exception: like Peter Salovey I have coauthored an assessment tool for EI (this is standard practice among psychologists; the various IQ tests embody differing theories of intelligence and how to measure it and are designed by the theorist). Our assessment tools are available only to professionals. Salovey’s is recognized as the flagship general measure of EI; mine is the ESCI, designed specifically for leadership development. Both Peter (I consider him a friend) and I are members of the Consortium for Research on Emotional Intelligence in Organizations, based at Rutgers. Our hope for the field is that rigorous research will more sharply define the EI construct, its correlates, and its practical applicationsall based on empirical data. That’s the way science grows and evolves. But good science takes time. Give it a decade, Po and Ashley, and let’s revisit these issues.