In order to gain insight into Earth’s environmental history, researchers need a large and well-vetted dataset that is tractable to multivariate statistical analyses. Data sgp has taken a unique approach to achieving these goals by assembling a consortium of geoscientists with the knowledge and skill necessary to construct such a dataset.
Achieving this goal requires a new generation of geologists with the necessary skills and tools to build such a database. It also requires a large investment of time and effort to train students in how to use this type of dataset. While the initial phases of the project have focused on establishing the consortium and increasing the data density to the point where it can be used, future work will expand the scope of geologic time, data types, and geography that is available for analysis.
Data sgp is an online platform that provides teachers and students with access to a wealth of educational resources. It allows them to track student growth and achievement in both academics and social-emotional learning (SEL). It also provides students with personalized recommendations for learning activities that are tailored to their individual needs.
The sgpData_LONG data set contains student assessment records in LONG format for 8 windows (3 windows annually). These are anonymized data sets, and include the variables VALID_CASE, CONTENT_AREA, YEAR, ID and a SCALE_SCORE, GRADE and ACHIEVEMENT_LEVEL (required if running SGP analyses). There is also a DATE variable that indicates the date associated with each record.
The SGP model estimates a latent achievement trait for each student by comparing their current test score with the scores of students with the same prior latent achievement. Because of the limitations of standardized tests, these estimates are error-prone and noisy measures of the true latent achievement trait. The figure below shows the RMSE (relative standard error) of these estimators as a function of the amount of prior and current test scores conditioned on. This plot is adapted from a publication by Akram, Erickson, and Meyer (2013). It is important to note that the reliability of the SGP model deteriorates rapidly as the amount of prior and current test scores is reduced. Despite these limitations, the availability of SGP makes it possible to conduct research on student performance that would otherwise be impossible. This is a valuable contribution to the scientific community. The authors are grateful to all who have contributed to this project and to the students and educators who have utilized it. The authors are also grateful to the National Science Foundation for support of this work. This publication is based on the work funded under Grant No. 1142217. Any opinions expressed herein are those of the authors and do not necessarily reflect those of the National Science Foundation. The author acknowledges assistance from the University of California, Merced, in constructing this article. –Sally A. Schreiber, Ph.D., is a professor of statistics at the University of California, Merced. She is the co-director of the UC Merced SGP Consortium and is a principal investigator on a NSF funded SGP project.