Multilevel model

  1. Psychological capital was linked to higher academic achievement
  2. Inflated responsibility linked to postnatal anxiety and breastfeeding
    Cross-sectional investigation of intolerance of uncertainty and responsibility beliefs in relation to postnatal anxiety and infant feeding outcomes in postpartum women.
  3. Resilience was the strongest buffer against stress among academic staff
    Study of 83 faculty members during COVID-19 remote teaching transition reveals resilience as strongest predictor of lower stress, with gender differences and benefits of prior online experience.
  4. Social media addiction moderates happiness and leisure boredom
    Study examines social media addiction as moderator between happiness and leisure boredom perception in youth, finding significant moderation effects.
  5. Family background and region shaped achievement patterns
    Multilevel analysis of TIMSS 2023 data reveals regional differences in how parental education, home resources, gender, and migration status predict mathematics and science achievement.
  6. Youth-oriented voting advice tool improved party position knowledge
    Experimental study shows youth-oriented voting advice application significantly improved party position knowledge among Belgian voters aged 16–30, while generic version had no effect.
  7. Mentalization shows a moderate link to adolescent internalizing outcomes
    Systematic review and meta-analysis of mentalization and mental health outcomes in adolescents with internalizing disorders, revealing a moderate association.
  8. CDS and inattention relate differently to personality and lifestyle factors
    Study examining cognitive disengagement syndrome in Korean children identifies differential relationships with personality traits and lifestyle factors compared to inattentive symptoms.
  9. Formative assessment data predicted standardized assessment performance
    Study examines predictive validity of computer-based formative assessment for standardised test outcomes, finding models explain 30-48% of variance with domain-specific patterns.