NEPSscaling: Plausible Value Estimation for Competence Tests Administered in the German National Educational Panel Study

Abstract

Educational large-scale assessments (LSAs) often provide plausible values for the administered competence tests to facilitate the estimation of population effects. This requires the specification of a background model that is appropriate for the specific research question. Because the German National Educational Panel Study (NEPS) is an ongoing longitudinal LSA, the range of potential research questions and, thus, the number of potential background variables for the plausible value estimation grow with each new assessment wave. To facilitate the estimation of plausible values for data users of the NEPS, the R package NEPS scaling allows their estimation following the scaling standards in the NEPS without requiring in-depth psychometric expertise in item response theory. The package only requires the user to prepare the data for the background model. Then, the appropriate item response model including the linking approach adopted for the NEPS is automatically selected, while a nested multiple imputation scheme handles missing values in the background data. For novice users, a graphical interface is provided that does not require knowledge of the R language. Thus, NEPS scaling can be used to estimate cross-sectional and longitudinally linked plausible values for all major competence assessments in the NEPS.

Fluid Intelligence and Competence Development in Secondary Schooling: No Evidence for a Moderating Role of Conscientiousness

“…These tests are developed especially for use within the NEPS, based on students’ curriculum, and evaluated in terms of their quality, for instance, regarding test fairness, item difficulties, and reliability. Although tests were reported as fair and reliable, the authors claim that not all competence levels were captured equally well ( Kock et al 2021 ; Scharl et al 2017 , 2021 ). In particular, whereas low performance in reading and mathematical skills was accurately measured, tests were less precise in assessing high-performance students.…”

“…The test for reading skills in German comprised 31 items in grade 4 (SC2), 42 items in grade 7 (SC2 and SC3), and 30 (easy version) or 32 (difficult version) items in grade 9 (SC3) on finding information in the text, drawing text-related conclusions, and reflecting and assessing presented as single multiple-choice, complex multiple-choice, and matching items. The reliabilities of the test were reported as good (WLE reliability = .75–97; Scharl et al 2017 , 2021 ).…”

“…This allows for competence comparisons of adjacent waves and analyses of change across secondary schooling. More information on the tests and scaling procedures is provided by Rohm et al ( 2017 ), Scharl et al ( 2017 ), and Scharl et al ( 2021 ), for reading skills in German, and by Kock et al ( 2021 ), Schnittjer et al ( 2020 ), and van de Ham et al ( 2018 ), for mathematical skills.…”

Causal effect analysis in nonrandomized data with latent variables and categorical indicators: The implementation and benefits of EffectLiteR.

“…Recommended is the inclusion of all (possibly latent) analysis variables as background variables (e.g., Adams et al, 2007;von Davier et al, 2009;Wu, 2005). For customizing PV estimation in relation to the specific research question at hand, the NEPScaling R package (Scharl et al, 2020) was released. The package facilitates the estimation of PV with specific background variables in the NEPS.…”