Reference
Lenhard, W., & Lenhard, A. (2021). Improvement of Norm Score Quality via Regression-Based Continuous Norming. Educational and Psychological Measurement, 81(2), 229-261. https://doi.org/10.1177/0013164420928457
Review
In their article, Lenhard and Lenhard (2021) compared semiparametric continuous norming (SPCN) with conventional norming methods to improve the quality of norm scores in psychometric tests. The authors simulated test scales with varying item numbers and difficulties and used random samples to model norm scores through either conventional ranking procedures or SPCN. Cross-validation and error measures were computed using a representative sample of 840,000 individuals.
The results indicated that both approaches benefitted from an increase in sample size, but SPCN achieved optimal results with smaller samples compared to conventional norming. Furthermore, conventional norming performed worse in terms of data fit, age-related errors, and missing values in norm tables. The authors questioned the general recommendations for sample sizes in test norming due to the varying data fit in conventional norming with fixed subsample sizes.
Lenhard and Lenhard (2021) concluded that test norms should be based on statistical models of raw score distributions rather than simply compiling norm tables via conventional ranking procedures. This study highlights the potential for SPCN to improve norm score quality in psychometric testing, which may lead to better interpretation and utility of test results in educational and psychological settings.
