Table 1.
Main characteristics of the studies included in the systematic review.
| References | Sample size | Mean age (years) | Percentage of men | Study country | Risk behavior instrument | Risk domain | EI instrument | EI model | Primary outcomes | Effect size [Pearson's correlation coefficient (r)] |
|---|---|---|---|---|---|---|---|---|---|---|
| Alipour and Mijani (2013) | 285 | No reported | No reported | Iran | Researcher-built questionnaire | Finances | Shrink questionnaire | Mixed | Significant positive relationship between EI and risk behavior. | 0.15 |
| Anwar et al. (2016) | 225 | 17.41 | 48.00% | Pakistan | HRBQ | Health | SEI | Mixed | Significant negative relationship between EI and risk behavior. | −0.48 |
| Dinç Aydemir and Aren (2017) | 496 | No reported (20 years or above) | 55.40% | Turkey | Researcher-built questionnaire | Finances | SSRI | Self-report ability | No relationship between EI and risk behavior. Significant positive relationship between EI and risk behavior when this relationship was included in a structural equation model along with the variables of locus of control, risk aversion, and financial literacy. |
0.00 |
| Fernández-Abascal and Martín-Díaz (2015) | 855 | 34.27 | 21.98% | Spain | HBC | Health | TMMS and TEIQue | Self-report ability and Mixed | Significant negative relationship between the TMMS dimension of clarity and risk behavior. Significant positive relationship between the TMMS dimension of attention and risk behavior. Significant negative relationship between the TEIQue dimensions of emotionality and self-control and risk behavior. The remaining dimensions of the TMMS and TEIQue were no related risk behavior. |
For TMMS EI instrument: −0.02 For TEIQue EI instrument: −0.09 |
| Hayley et al. (2017) | 179 | 29.85 | 55.00% | Australia | BDDS and DDDI | Health | SUEIT | Self-report ability | No relationship between EI and risk behavior. | 0.01 |
| Lana et al. (2015) | 275 | 22.40 | 11.60% | Spain | Researcher-built questionnaire | Health | SSRI | Self-report ability | The group of participants scoring higher in the risk behaviors of excessive alcohol consumption and unsafe sex showed lower EI. The risk behavior of illicit drug use was not associated with EI. |
Not applicable. |
| Lando-King et al. (2015) | 253 | 15.60 | 0.00% | USA | Researcher-built questionnaire | Health | BarOn EQ-i: YV | Mixed | Significant negative relationship between the BarOn EQ-i: YV dimensions of intrapersonal and interpersonal skills and number of sex partners. Significant negative relationship between the BarOn EQ-i: YV dimension of stress management and inconsistent condom use. The remaining dimensions of the BarOn EQ-i: YV were no related to number of sex partners or inconsistent condom use. |
For the risk behaviors of “number of sex partners”: 0.16 For the risk behaviors of “inconsistent condom use”: −0.05 |
| Malinauskas et al. (2018) | 1,214 | 22.36 | 49.17% | Lithuania | HBC | Health | SSRI | Self-report ability | Significant positive relationship between all the dimensions of the SSRI and the risky driving behavior. Significant negative relationship between the SSRI dimensions of optimism, appraisal, and utilization and the risk behavior of substance abuse. No relationship between the SSRI dimension of social skills and substance abuse. |
For the risk behaviors of “traffic risk taking”: 0.16 For the risk behaviors of “substance risk taking”: −0.05 |
| Micklewright et al. (2015) | 34 | 39.9 | 94.12% | UK | DOSPERT | General risk perception | SSRI | Self-report ability | The group of higher risk-perceivers showed higher EI compared to the group of lower risk-perceivers. | Not applicable. |
| Panno (2016) | 94 | 17.23 | 79.00% | Italy | Cold CCT | Gambling | TEIQue-ASF | Mixed | Significant positive relationship between EI and risk behavior. | 0.25 |
| Panno et al. (2015) | 158 | 21.64 | 24.00% | Italy | Hot CCT | Gambling | TEIQue-SF | Mixed | No significant direct relationship between EI and risk behavior. Significant positive indirect relationship between EI and risk behavior via negative mood and anticipated fear. |
0.09 |
| Rivers et al. (2013) | 243 | No reported (between 18 and 19) | 25.10% | USA | CSLSS | Health | MSCEIT | Performance-based ability | Significant negative relationship between EI and the risk behaviors of substance abuse, adjustment problems, and aggressive behavior. | For the risk behaviors of “substance abuse”: −0.18 For the risk behaviors of “adjustment problems”: −0.16 For the risk behaviors of “aggressive behavior”: −0.25 |
| Vaughan et al. (2019) | 269 | 21.80 | 57.62% | Ireland | CGT | Gambling | SSRI | Self-report ability | Significant negative relationship between the four dimensions of the SSRI and risk behavior. | −0.2 |
| Yip and Côté (2013) | 52 | 24.00 | 37.00% | USA | IGT | Gambling | MSCEIT | Performance-based ability | No relationship between EI and risk behavior. | Not applicable. |
| Zavala and López (2012) | 829 | 13.60 | 47.50% | Mexico | MACI | Health | BarOn EQ-i: YV | Mixed | Significant negative relationship between EI and risk behaviors associated with eating disorder and substance abuse. | For the risk behaviors of “eating disorders”: −0.20 For the risk behaviors of “substance abuse”: −0.29 |
HRBQ, Health Risk Behavior Questionnaire; HBC, Health Behavior Checklist; BDDS, Distracted Driving Scale; DDDI, Dangerous Driving Index; DOSPERT, Domain Specific Risk Taking; CCT, Columbia Card Task; CSLSS, College Student Life Space Scale; CGT, Cambridge Gambling Task; IGT, Iowa Gambling Task; MACI, Millon Adolescent Clinical Inventory; SEI, Scale of Emotional Intelligence; SSRI, Schutte Self-Report Inventory; TMMS, Trait Meta Mood Scale; TEIQue/-SF/-ASF, Trait Emotional Intelligence Questionnaire/-Short Form/ -Adolescent Short Form; SUEIT, Swinburne University Emotional Intelligence Test; BarOn EQ-i: YV, Bar-On Emotional Quotient Inventory: Youth Version; MSCEIT, Mayer-Salovey-Caruso Emotional Intelligence Test.
For those studies that did not show a global score of EI, primary outcomes were reported separately for each EI dimension and effect sizes were averaged across EI dimensions within each study in order to provide an approximate effect size for the global EI (see Results of the meta-analysis section).