Table 2 Assessment of measurement invariance across the group-tested (N = 92, school) and individually tested (N = 535, laboratory) cohorts

From: Collecting big data with small screens: Group tests of children’s cognition with touchscreen tablets are reliable and valid

Models χ2 Comparative Fit Index (CFI) Root mean square error of approximation AIC
χ2 df p CFI Δ LB UB RMSEA Δ LB UB
1. Configural 63.4 28   .971     .0635     11012
2. Loadings 73.8 34 .107 .967 – 0.004 – .024 .001 .0611 – 0.002 – .010 .010 11010
3. Intercepts 98.3 40 <.001 .952 – 0.015 – .038 – .005 .0682 0.007 – .004 .018 11023
4. Residuals 107 47 .273 .951 – 0.001 – .023 .001 .0638 – 0.004 – .008 .005 11017
5. Means 113 48 .015 .947 – 0.004 – .014 .001 .0657 0.002 – .001 .006 11021
  1. Note: The lower (LB) and upper bounds (UB) of the 95% confidence interval for changes (Δ) in CFI & RSMEA are reported, calculated using non-parametric bootstrap resampling (3000 repeats). Random sampling was performed within each group separately for each iteration.