Test publishers and users should routinely investigate longitudinal achievement construct invariance over time¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends an ongoing validation practice: "We strongly recommend that achievement test publishers and users continue investigating the longitudinal achievement construct and construct invariance over time in the near future to support valid interpretations of student academic growth." This extends traditional construct validation, which the authors note usually collects evidence at one point in time only, into the longitudinal domain where growth interpretations depend on the construct not shifting across grades or semesters.

Design Implications¶

Context¶

Requirements¶

• Longitudinal repeated-measures data and latent-growth modelling capable of testing factorial invariance across time points

Constraints¶

• The article notes few studies have reported longitudinal achievement constructs, especially for CAT data, so established procedures are limited

Target Learners¶

• K-12 students whose test scores support growth-based interpretations and high-stakes decisions

Target Learning Goals¶

• Valid interpretation of student academic growth from vertically scaled achievement test scores

Related Strategies¶

• Brief researchers and media on how to calculate and interpret months-of-learning figures before using them
• Validate EDM models across demographic subpopulations of students

Examples¶

-

Key Sources¶

• Wang, S., Jiao, H. and Zhang, L. (2013) 'Validation of longitudinal achievement constructs of vertically scaled computerised adaptive tests: a multiple-indicator, latent-growth modelling approach', Int. J. Quantitative Research in Education, Vol. 1, No. 4, pp.383–407. https://www.nwea.org/research/publication/validation-of-longitudinal-achievement-constructs-of-vertically-scaled-computerised-adaptive-tests-a-multiple-indicator-latent-growth-modelling-approach/. doi:10.1504/ijqre.2013.058307