The study, published in Frontiers in Psychology , drew on transactional stress-and-coping theory, which treats stress as a dynamic result of how people appraise demands and the resources they believe they have available to manage them. In that framework, coping represents cognitive or behavioural efforts directed at environmental or internal demands, while self-efficacy represents a person’s belief in their capacity to carry out the behaviours required to address those demands. The researchers examined whether efficacy beliefs changed the strength of the association between coping and wellbeing among TVET students, a population that combines competency-based assessment, practical skill development and preparation for workplace requirements.

Participants were recruited through a multistage sampling procedure involving institutions in five Kenyan regions: Nairobi, Mount Kenya, the Coast, Western and Rift Valley. Institutions were sampled within the regions, and trainees were randomly selected within participating institutions. The final analytical sample included 1,118 males and 819 females, with ages ranging from 18 to 37 years and a mean of 22.83 years.

Most respondents were in their second year of study, most were enrolled at national polytechnics, and 73.9% were taking technical courses. A large majority, 79.1%, reported coming from households with a monthly income of no more than 14,000 Kenyan shillings. Coping was measured with the Brief Coping Orientation to Problems Experienced Inventory, general self-efficacy with the 10-item General Self-Efficacy Scale, and mental wellbeing with the 14-item Warwick-Edinburgh Mental Wellbeing Scale.

The instruments had internal consistency coefficients of 0.934 for the retained approach-coping scale, 0.793 for avoidant coping, 0.880 for self-efficacy and 0.91 for wellbeing. Confirmatory factor analysis supported a two-dimensional coping structure distinguishing approach from avoidant coping. The retained model produced acceptable, though not uniformly strong, fit indices: CFI = 0.897, TLI = 0.888, RMSEA = 0.055 and SRMR = 0.083.

The approach dimension included active coping, planning, support-seeking, positive reframing, acceptance, humour, venting, self-blame and self-distraction, while avoidant coping comprised denial, substance use and behavioural disengagement. The two coping dimensions were associated with different patterns in the initial analysis. Approach coping had a large positive correlation with mental wellbeing, r = 0.589, while avoidant coping had a negligible bivariate correlation of r = 0.073.

General self-efficacy had the strongest observed association with wellbeing, r = 0.750, and was also positively associated with approach coping, r = 0.608. Avoidant coping showed no significant correlation with self-efficacy at r = 0.041. When approach coping was included in the model, the unique association between avoidant coping and wellbeing became negative, with an unstandardized coefficient of B = −0.183.

The researchers described this as statistical suppression: the moderate overlap between active and avoidant coping initially obscured the negative association. Moderation analyses used Hayes’s PROCESS Model 1, ordinary least-squares regression with heteroscedasticity-consistent HC3 standard errors, and mean-centred predictors. Each coping dimension was examined in a separate model, with the other dimension entered as a covariate.

For approach coping, the model explained 60.0% of the variance in mental wellbeing. Both approach coping and self-efficacy had significant main effects, but their interaction was negative and statistically significant, B = −0.010, 95% CI [−0.014, −0.006], p < 0.001. The interaction added 0.8% to explained variance, making its magnitude modest despite being detectable in the large sample.

The conditional estimates revealed the pattern at the heart of the findings. At the 16th percentile of self-efficacy, the association between approach coping and wellbeing was B = 0.296, 95% CI [0.235, 0.357]. At the median, it fell to B = 0.225, 95% CI [0.173, 0.276], and at the 84th percentile it was B = 0.133, 95% CI [0.072, 0.195].

Johnson–Neyman analysis indicated that the approach-coping association remained statistically significant across the full observed range of self-efficacy scores. In practical terms, the authors reported that a 10-point difference in approach coping corresponded to approximately 3.0 additional points on the wellbeing scale among trainees at the low self-efficacy percentile, compared with 1.3 points among those at the high percentile. Self-efficacy did not significantly moderate the relationship between avoidant coping and wellbeing, for which the interaction was B = −0.016, p = 0.059 and ΔR² = 0.002.

In that model, avoidant coping remained negatively associated with wellbeing, B = −0.102, while approach coping, included as a covariate, remained positively associated, B = 0.198. Supplementary models adjusted for age, gender, year and level of study, institution type, course classification and family income. The approach-coping interaction was nearly unchanged, with B = −0.010, p < 0.001 and ΔR² = 0.007, indicating that the observed pattern was not accounted for by the measured sociodemographic variables.

Several limitations constrain how the findings should be interpreted. The cross-sectional design cannot establish whether self-efficacy changes the effect of coping, whether effective coping strengthens self-efficacy, or whether other factors influence both. All three central constructs were measured by self-report, creating the possibility of social-desirability and common-method effects.

The retained approach-coping category was also broad, combining strategies with potentially different adaptive value, and the sampling omitted institutional identifiers, preventing the researchers from estimating clustering or applying cluster-robust standard errors. Although the interaction remained after demographic adjustment, unmeasured institutional characteristics and the results’ restriction to the sampled Kenyan TVET setting limit generalisation. The evidence therefore supports a nuanced conclusion: active coping was associated with greater wellbeing at every measured level of self-efficacy, while greater self-efficacy reduced the additional contribution that coping made to the association.

Rather than showing that confidence or coping could replace one another, the study found that both remained linked to wellbeing, with engagement-oriented strategies carrying their strongest measured association among trainees who reported less confidence in their ability to manage academic and occupational demands.

Subject of Research: How self-efficacy moderates associations between coping strategies and mental wellbeing among Kenyan TVET trainees

Article Title: Coping strategies and mental wellbeing among TVET students: the moderating role of self-efficacy

Article References: Owuor, N. O., & Piko, B. F. (2026). Coping strategies and mental wellbeing among TVET students: the moderating role of self-efficacy. Frontiers in Psychology, 17, Article 1951147. https://doi.org/10.3389/fpsyg.2026.1951147

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1951147

Keywords: TVET students, mental wellbeing, self-efficacy, coping strategies, Kenya, cross-sectional survey, student psychology, approach coping, avoidant coping, vocational education, Coping, strategies