We use the term “big data” to refer to datasets that are very large and require special software tools to manage them. The SGP database is much bigger than a typical academic research database, but not nearly as big as a Facebook dataset. We are working on ways to move the data sgp into permanent, archived repositories.
SGPs are a popular way to assess student growth in the United States, but they have several limitations. First, they depend on standardized test scores that are error-prone measures of latent achievement traits. These errors lead to noisy, aggregated estimates of a student’s true SGP. We can reduce the noise by using a more accurate measurement of a student’s current latent achievement trait (e.g., the sum of a student’s test score on each of the individual tests used in his or her evaluation).
In addition to estimating SGPs, the sgpData database contains latent traits for each of the individual tests that students take in MCAS, as well as the prior performance of each of those students. This allows us to compare the performance of a student across different grades and subjects, as well as to identify trends in student growth.
We also collect latent traits for the teachers who have taught each student, as well as teacher characteristics and classroom characteristics. This allows us to analyze how student performance is related to each of these factors, as well as to determine whether teachers are more or less effective in a particular subject or grade level.
A common question is whether or not SGPs correlate with a student’s background characteristics, such as race or socioeconomic status. In general, the correlations between true SGPs and these variables are very small. However, it is possible that there are underlying relationships between background variables and SGPs that are not captured by our existing covariate analyses. These relationships could be exploited in future research to improve the accuracy of SGPs estimated from standardized tests.
Finally, it is important to note that SGPs are calculated using mean instead of median calculations, which may influence how students and schools are ranked. While this does not change the underlying calculations, it is an important consideration for those who are using SGPs to evaluate schools and teachers.
The sgpData website is a powerful tool for anyone interested in using the SGP data. Its functionality and design make it easy to navigate for even the most novice user. In addition to this, the site has been optimized for mobile devices, making it even more convenient to access the SGP data. In addition, the site is secure and features a virtual private network for users to protect their identity.