Measuring Digital Stress in Children: Construct Validity, Model Comparisons, and Measurement Invariance of a Multidimensional Scale (DSS-CH)

<b>Background:</b> The use of digital media in childhood offers both opportunities and risks. Digital stressors-such as excessive screen time, constant availability, information overload, and social media pressures-affect primary school children but have been rarely studied systematically. Despite growing research, no validated instruments adequately capture how younger children perceive and express digital stress. This study presents the development and validation of a three-dimensional instrument for children under 14: the "Digital Stress in Children" scale (DSS-CH). The DSS-CH is theory-driven and child-appropriate, with three interrelated but distinct dimensions: (1) excessive screen time, (2) compulsive media behavior, and (3) approval anxiety. <b>Methods:</b> In a cross-sectional survey of <i>n</i> = 907 Swiss primary school children (grades 5-6; ages 10-14), participants completed an online questionnaire in class. Latent variable modeling with cluster-robust inference accounted for classroom nesting. Competing models (1-, 2-, 3-factor CFA; ESEM; bifactor-ESEM) were evaluated. <b>Results:</b> The 1-factor CFA fit poorly (CFI ≈ 0.81; RMSEA ≈ 0.15), while the 3-factor CFA showed acceptable fit (CFI ≈ 0.96; RMSEA ≈ 0.07). Allowing cross-loadings improved fit substantially in the 3-factor ESEM and bifactor-ESEM (CFI ≈ 0.999; RMSEA ≈ 0.01), supporting a general digital stress factor alongside facet-specific variance. Subscales showed good reliability (ordinal α ≈ 0.81 - 0.89) and moderate intercorrelations (r ≈ 0.28 - 0.47). Scalar invariance across gender and age was supported (ΔCFI ≤ 0.003; ΔRMSEA ≤ 0.012). <b>Conclusions:</b> The DSS-CH demonstrates good reliability, model fit, and measurement invariance. It provides valid evidence for interpreting children's digital stress as three related facets and can help identify elevated stress profiles to inform preventive efforts.