The same bifactor DL structure holds for primary and secondary students under partial measurement invariance, permitting comparison of performance levels across cohorts¶

Claim · All claims
Evidence · 1 study · associational r3 · q2
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Full measurement invariance across P5 and S2 students was not quite supported, but a partially constrained model with equal parameters and unconstrained means showed much better fit and was adopted for both cohorts. → Frank Reichert 2020

Evidence¶

Frank Reichert 2020¶

Frank Reichert, Nancy W. Y. Law, Gary K. W. Wong, Jimmy de la Torre. (2020). Exploring the structure of digital literacy competence assessed with authentic software applications. Education Tech Research Dev, 68, 2991–3013. https://doi.org/10.1007/s11423-020-09825-x

q2 · i? · associational · r3

Multiple-group analysis in Results comparing P5 and S2 cohorts: partial measurement invariance with unconstrained means fit better (RMSEA = .048, CFI = .931) than full invariance, so "the same dimensional structure of DL was adopted for both age cohorts".

"The assumption of full measurement invariance was not quite supported (RMSEA = .056, CFI = .914, SABIC = 15,963), but a model where all parameters except the mean levels of student performance were held equal showed a much better fit to the data (RMSEA = .048, CFI = .931, SABIC = 15,710)."

Discussion¶

Related Claims¶

• Technical digital literacy measured with authentic software applications is not a unidimensional construct; a four-dimensional exploratory MIRT model fits best — related
• DL performance comprises a general dimension plus four tool-specific residual dimensions, with the specific dimensions defined by the software applications used — related
• Secondary students outperformed primary students on general DL and online information seeking, but primary students performed slightly better on presentation-related DL — a narrower finding that bears on this claim
• The five-dimensional student model achieved metric but not scalar measurement invariance across the seven countries, permitting association comparisons but not latent mean comparisons — related