Data sgp is a large database that contains information about student growth in academic achievement. This information is useful for educators, school leaders, and parents because it helps them identify students who need additional support. It also helps them determine which strategies will be most effective for each student. Typically, these strategies include tutoring and mentoring.
The SGP package offers an efficient means of organizing longitudinal (time dependent) student assessment data into statistical growth plots. It supports two common formats for this data: WIDE and LONG. WIDE format is used by the lower level SGP functions studentGrowthPercentiles and studentGrowthProjections, while higher level wrapper functions like sgpData utilize the LONG data format. In general, it is more practical to use the LONG data format for operational analyses if you plan on running them year after year as the data requires more extensive preparation and storage.
In this article, we explore the properties of a model for latent achievement attributes and their true SGPs, as well as describe some of the ways in which these relationships can be assessed from the data. We focus on the case of estimating math SGPs from grade-level assessment scores, but we also discuss relationships between latent traits and covariates for ELA SGPs.
One of the key properties that we establish is that the mean SGPs for a given teacher tend to be closer to 50 than their median SGPs, reflecting the fact that schools with high averages have more teachers who are above or below the average. In addition, we find that relationships between latent trait and covariate characteristics are stronger for ELA than for math.
This is likely due to the fact that there are more ELA traits than math traits, and that a greater proportion of students with low-achieving characteristics have ELA traits than those with high-achieving characteristics. The relationship between a student’s math and ELA traits could be explained by the fact that they are taught by different teachers, or that they have different backgrounds.
SGP estimates are based on historical growth trajectories of Star examinees and project what they will need to reach proficiency in each subject. The projections are then compared to a set of benchmarks that determine whether a student is at or above proficient.
The sgpData dataset provides the raw data used to calculate these projections. The first row of the dataset provides a unique ID for each student, and the next five rows provide their assessments from grades 2013, 2014, 2015, 2016, and 2017. Each row includes an indicator of whether or not the student was identified as having English learning disabilities. The last five columns give the scale score associated with each of these assessments. These scale scores are used to create a student growth percentile, which is a measure of how far the student has come on MCAS compared to other students with similar performance histories. The higher the number, the more advanced a student is.