Thursday, January 26, 2012

So How Do You Really Measure Emotional And Social Competence?

The first, delineated by Reuven Bar-On, was influenced by his interest in the aspects of performance not linked to intelligence; the second, often tied to Daniel Goleman's interpretation, approached EI through competencies; and the third, represented by Mayer and Salovey and colleagues, was influenced by their interest in the relationship between cognition and emotion.

These three approaches have led to diverse and non-overlapping measures of EI characterized as: 1) Personality oriented (e.g, Bar-On Emotion Quotient Inventory); 2) Competency or "Mixed" model oriented (e.g., Emotional Intelligence View 360); and 3) Ability or skill oriented (e.g., Mayer-Salovey-Caruso Emotional Intelligence Test; MSCEIT).

Issues with Ability Based Measures of Emotional Intelligence

  • Independece from personality measures (e.g., five factor models)
  • Weak convergent validity with other cognitive ability measures (i.e., they don't highly correlate with IQ)
  • Scoring issues (i.e., lack of agreement and some controversy on how these assessments are scored)
  • Confounded with a measure of knowledge (i.e., they seem to be measuring what someone "knows" as well as emotional intelligence)

Problems wiht Self-Report (Mixed) Measures of Emotional Intelligence

  • High correlations with five factor personality measures (i.e., the overlap is so high it suggests that some measures of EI are really nothing more than another personality inventory)
  • Limitations of 360-feedback (e.g., inflated self-ratings, moderate correlations between and within rater groups)
  • Limitations of self-report (how do you measure EI in people who lack emotional intelligence?)
  • Tend to ignore context, situation and setting (EI is not a useful predictor of performance in jobs that don't have high emotional labor or are socially demanding)

Our own "mixed measure" of ESC called Emotional Intelligence View 360 based on the Goleman construct has some strengths and limitations as all measures. Our EIV360 appears to be statistically unique from ability based measures (very low correlations with the MSCEIT), correlated with the most popular measures of transformational leadership and predictive of both academic and work performance.

In a review by Joseph and Newman (2010), they found a negative association between measures of EI and work performance when jobs do not require strong social skills. Although the sample sizes for this analysis were rather low (N = 220 and N =223, respectively) it does suggest that EI is important for positions like sales, customer service and leadership and less important in predicting performance and success when high levels of interpersonal interaction are required1.

A newer 2010 meta-analysis by O'Boyle et al. included 65% more studies and twice the sample size to estimate EI and job performance outcomes2.

Their findings extent those of Newman (2010) and suggest that trait, personality and mixed measures demonstrated corrected correlations ranging from 0.24 to 0.30 with job performance. Their research also shows that all measures show incremental validity over cognitive ability and personality measures.

Measurement of emotional intelligence (ability based) is most likely different from other approaches (personality and mixed) but all techniques tend to significantly predict job performance, health and social competence particularly in roles and positions requiring high interpersonal interaction. So, depending on your purpose (e.g., selection versus development of talent) some approaches to measuring EI might be better than others.

The one big lesson from the confusion in the measurement of emotional intelligence is that "it's not HOW smart you are that counts, but how you are smart…Be well….

  1. Joseph, D. & Newman, D. (2010). Emotional intelligence: An integrative meta-analysis and cascading model. Journal of Applied Psychology, 95, 54-78 []
  2. O'Boyle, E., Humphrey, R., Pollack, Hawver, T. & Story, P. (2010). The relationship between emotional intelligence and job performance: A meta-analysis. Journal of Organizational Behavior, 10.1002/job.714 []
Thanks to Ken Nowack / Result Envisia Learning / Envisia Learning
http://results.envisialearning.com/
 
Amazon Magazine Subscriptions Amazon Books Amazon Kindle Store
Amazon Everyday Low Prices, Sales, Deals, Bargains, Discounts, Best-Sellers, Gifts, Household Consumer Products
 
 

No comments: