
Data SGP leverages longitudinal student assessment data to produce statistical growth plots (SGPs). SGPs compare a student’s current performance to that of their academic peers nationwide. The data used to generate SGPs are the same as that provided to schools and districts for the state’s accountability reports (STAR). Using these tools, teachers can make comparisons of their students to a national sample of academic peers and see how their students are progressing relative to others.
A SGP is a percentage of the range of scores for students with similar prior test results (their academic peers). The calculations behind a SGP are complex, but the information they provide can be communicated in percentile terms that are familiar to most teachers and parents.
In order to use SGP data, schools or districts must first be a part of the state’s accountability system, which provides access to STAR test data and reports. The next step is to create a spreadsheet that contains a list of all the students in your school or district. For each student, you must provide a unique identifier and grade level. You must also provide at least two prior tests taken within the past 18 months and one of those tests must be from a previous testing window.
Once you have the data, you need to sort and format it. Then, you need to determine what the SGP denominators and factors should be. For example, you need to know how many decimal places to display in the BA column and how many columns to display in each stat category. You should spend at least a minute formatting the cells in these tables so that they will be easier to read.
Using the sgpData and sgptData_LONG data sets to run SGP analyses is relatively straightforward. The first column in the sgpData set, ID, provides a unique student identifier. The other 5 columns, SS_2013, SS_2014, SS_2015, and SS_2016, provide the scale scores associated with each student on the various Star assessments. For students that do not have a full 5 years of test data, the missing values are shown as NA.
It is important to note that the sgptData_LONG set is an anonymized, panel data set with 8 windows (3 windows annually) of assessment data in long format for 3 content areas. This data is primarily used to create SGPs, and there are several additional rules for working with this data. Refer to the SGP data analysis vignette for more information on using wide-format data with SGP analyses. This vignette provides specific instructions on how to prepare the sgptData_LONG for SGP analyses. This is a necessary step in ensuring that the results of SGP analyses are valid. Using sgpData and sgptData_LONG with the SGP package is not as straightforward as using other wide-format data sets. In particular, the variables VALID_CASE and CONTENT_AREA are required to properly calculate SGPs. If these variables are not present in the data, the SGP analyses will fail.