With the current Scientific Use File of Starting Cohort 5 (First-Year Students, version 14.0.0), so-called Plausible Values (PV) were published for the first time for the main domains of competence data. For Starting Cohort 6 (Adults, version 11.0.0), a similar data offering will follow very soon. For further NEPS starting cohorts, the addition of Plausible Values is already in preparation. Besides the provided PV variables, the freely available R-package „NEPSscaling“ provides the possibility to generate independently suitable Plausible Values for own analysis purposes.
Further informationen under „Overview and Assistance“
Plausible Values can be used to describe the competencies of people at a group level. On the basis of the individual responses to the respective competence tests as well as additional background characteristics, the probability distribution of each person's competence is first determined and several values are randomly drawn from this. The analysis is then calculated individually for each Plausible Value and summarized into an overall result. This allows estimates of effects at the population level to be adjusted for measurement errors and is therefore particularly suitable for inferential statistical tests in correlation and mean value analyses. For further information on the estimation and handling of Plausible Values in the context of the National Educational Panel Study (NEPS) as well as instructions and an example of the use of the R-package „NEPSscaling“ for generating your own plausible values, please refer to the NEPS Survey Paper by Anna Scharl, Claus H. Carstensen and Timo Gnambs.
NEPS Survey Paper (preprint) regarding „Estimating Plausible Values with NEPS Data“