The current study aimed to investigate complex links among a large set of anxiety-related variables and identify targets for well-being interventions in a large sample of male and female vocational education training students. In total, 28 psychological constructs, such as self-esteem, parental pressure and dissatisfaction and motivation, were assessed in four groups of VET students (mode age: 16). The sample included 3069 females in ASSP schools (nursing and caring); 2108 females and 1772 males in Commerce schools (sales and management); and 2262 males in MELEC schools (electricity and maintenance). We used Gaussian Graphical models (GGMs) that allow for building sparse models of links among multiple variables and detecting targets for interventions via the identification of the most central nodes. We showed gender differences in absolute means for some variables (higher self-esteem and math grades in males; higher anxiety and error sensitivity, but stronger endorsement of mastery approach achievement goals in females), as well as in network structure. GGMs suggested that the key nodes were self-reported math competence for females in the ASSP group, self-regulation for females in Commerce, and mastery approach goals for males in both MELEC and Commerce groups, and that these should be differentially targeted by educational interventions in these populations.
Likhanov et al. (2026) studied this question.