Table 8.
Review of selected studies on psychometric properties of the Emotional and Social competence Inventory (ESCI).
| Citation | Participants (N, age, occupation, gender ratio etc.) | Study design | Reliability and validity evidence | Comment |
|---|---|---|---|---|
| Technical Manual for Emotional and Social Competence Inventory (ESCI) (Boyatzis and Goleman, 2007—manual updated 2011) US and UK |
Non clinical The manual reported the development of the ECI and ESCI utilizing 3 primary samples: N > 4,000 self-assessments from managers and professionals, and more than 10,000 other assessments (1998). N = 116 self-assessments, 1,022 other assessments (2007). N = 5,700 self-assessments and 62,000 other assessments (2,010). |
The studies reported in the technical manual relate primarily to the factor structure, reliability and validity of the ESCI. The factor structure and reliability studies utilize a cross sectional design. Validity studies comprised a combination of self-report and other-reports, including other reports of performance (e.g., supervisor ratings). |
Internal consistency: The internal consistency of the 12 scales of the ESCI range from 0.79 to 0.91 (n = 52,363; published in technical manual as well as Boyatzis and Gaskin, 2010). This is for the “other” assessments. Test-retest reliability data was only given from the ECI, this ranged from 0.41 to 0.92. Construct validity: This was assessed by examining correlations with similar constructs from the MBT (N = 18 paramedics). MBTI intuiting types scored highly on several EI competencies including emotional self-awareness and self-control. MBTI feeling types scored highly on self-awareness, empathy and others. Evidence for discriminant validity was based on low correlations with subtests of the Watson-Glaser Critical Thinking Appraisal (N = 90 executives). Evidence for predictive validity was found for self-reports of job success, life success and salary across a range of sectors N < 300. |
The technical manual is available online1. It reports a combination of data from published industry reports as well as published peer-reviewed academic articles. |
| Morrison (2008) USA |
Non clinical N = 92 Sample: registered nurses from 3 healthcare facilities in South Mississippi. Gender: of the 92 participants, 71 were female and 21 were male. Age: age ranged from 20 to over 60 years, with 47.8% of participants between 20 and 30 years of age. Ethnicity: the majority of the nurses were Caucasian (85.9%). Education: 72.8% had a Bachelor degree in nursing. Experience: over half of the nurses had four or less years of work experience. |
Cross-sectional correlational design completed by both the participant and peer reviewer. The purpose of this study was to determine if a relationship exists between emotional intelligence (EI) and preferred conflict-handling styles of registered nurses. Each participant completed the ECI 2.0 (later renamed to ESCI) as well as an instrument to measure conflict handling. The participants handed a second ECI 2.0 instrument to a known manager, peer or subordinate they had worked with in the past year. The peer, subordinate or supervisor was asked to evaluate the participant using the ECI 2.0 instrument. |
Internal Consistency: The authors reported existing Cronbach's alpha scores. Cronbach's alpha for others-rating ranged from 0.73 (Trustworthiness) to 0.92 (Empathy), with an overall (other-rating) average internal consistency coefficient of 0.85. The internal consistency coefficients for self-rating ranged from 0.61 (accurate self-assessment) to 0.85 (service orientation), with an overall average internal consistency coefficient of 0.75. Construct Validity: The study indicated a positive and statistically significant relationship between collaboration and all four of the EI clusters: self-awareness (r = 0.25), self-management (r = 0.32), social awareness (r = 0.31), and relationship management (r = 0.37). |
Small sample size which could limit the generalizability. The authors also noted that an organizational climate survey could be administered to assess whether the organizational climate affects how a registered nurse responds when faced with conflict. |
| Reed et al. (2015) USA |
Non Clinical N = 40 Sample: First year pediatric and medicine/pediatrics interns from a Children's Hospital. Gender: 32 females and 8 males. |
Cross-sectional, self-report and peer-assessment design. The aim of the study was to determine: (1) performance of first-year pediatric residents in the delivery of bad news in a standardized patient (SP) setting; and (2) the role of EI in these assessments. Skill in bad news delivery was assessed via SP encounters using a previously published assessment tool and being exposure to a scenario. Residents completed the ESCI as a measure of EI. The ESCI was administered via a self-assessment to residents online, with eight relevant peers/supervisors invited to complete a confidential assessment of the resident. For each resident, a minimum of five other assessments were obtained and averaged to create the Other assessment. The Other assessment score was used for analysis. |
Internal Consistency: The authors noted that the internal consistency of each ESCI subscale for Other assessments (i.e., peer ratings) was consistently high (Cronbach's alpha was not reported). Construct Validity: No ESCI subscales were significantly associated with residents' death notification skills, demonstrating no construct validity. |
Limitations of this study include a small sample size drawn from a single institution, the use of a single SP encounter, and a reliance on only one type of bad news scenario (i.e., death notification) which is arguably among the most difficult. The study did not account for differences in trainees' previous experiences and/or training in breaking bad news and death notification. Additionally, limited reliability and validity data were obtained. |
| Boyatzis et al. (2017) USA and Northern Europe |
Non clinical N = 40 Sample: Engineers in a multi-national manufacturing company. Gender: 37 males, 3 females. Age: Age ranged from 25 to 64 years with the modal age range of 35 to 44. Employment: The average tenure in the organization was 13 years. Country: 33 were from USA and 7 from Europe. |
Cross-sectional, self-report and peer-assessment design. The survey was administered online. The total number of peers completing the surveys for the 40 engineers was 168 (average of 4.2 per person). Peers reported on the perceived emotional intelligence of the participant using the ESCI as well as perceived effectiveness measured with the Reputational Effectiveness Scale (RES). Self-report measures included job engagement, cognitive intelligence, personality and quality of relationships. |
Internal Consistency: Coefficient of 0.95. This was based on the overall ESCI score since subscale scores were not used in this study. Construct Validity: ECI correlated with engineer reputational effectiveness (r = 0.70, p < 0.01) but not with general mental ability or personality. |
This study presents a number of limitations. There was a small sample size which may limit the generalizability of the findings. The low response rate (5% valid responses) may have resulted in more of a volunteer bias than is often encountered in survey research in organizations. Further, due to the limited sample, it may be possible that the findings may be a function of organizational culture. Statistically speaking, the ESCI was completed by subordinates, so there could be an inflated effect due to common source. |