Development and validation of an ideological-political literacy scale for university students in the New Era using multidimensional item response theory

Introduction Existing measures of ideological-political literacy among Chinese university students have limited dimensional clarity and cross-group comparability. This study developed and evaluated a five-dimensional scale covering political identification, patriotic sentiment, moral cultivation, rule-of-law awareness, and cultural confidence using multidimensional item response theory. Methods Framework-based qualitative coding, two-round Delphi consultation, content-validity screening, and pilot testing yielded 47 candidate items. A stratified cluster sample of 3,716 undergraduates from 28 universities across 12 provinces provided calibration data. A compensatory multidimensional graded response model (C-MGRM) was estimated by marginal maximum likelihood, with Metropolis-Hastings Robbins-Monro estimation as a same-sample robustness check. Competing dimensional specifications, reliability, measurement invariance, and differential item functioning (DIF) were examined. Results The 47-item oblique model yielded RMSEA = 0.043, CFI = 0.957, and TLI = 0.951. After four items were removed, the final 43-item model yielded RMSEA = 0.039, CFI = 0.964, and TLI = 0.958; marginal reliabilities ranged from 0.82 to 0.91. Although bifactor models showed slightly better approximate fit, their indices did not support a substantive general-factor interpretation. Scalar invariance was reported across gender, grade, and disciplinary clusters. Eight items showed DIF across urban-rural or party-membership comparisons, leading to proposed wording revisions or interpretive caveats. Discussion The scale provides dimension-specific scores and a basis for further measurement development. Same-sample model selection, incomplete external-validity evidence, and unadjusted clustering in the MIRT and DIF analyses limit interpretation. Independent validation is needed, and the findings do not establish readiness for an operational computerized adaptive test.