Importing SGP Data Into a Data Management System
A key to a robust data management system is having a strong understanding of the format of data that you work with. Almost all errors that come up during an SGP analysis revert back to issues related to data preparation. This is especially important when working with state data sets, which often contain a mix of formats, lengths and structures that need to be unified into a consistent data set for SGP analyses. The SGP package includes an exemplar data set, sgpData, that models the format of data used by the lower level SGP functions such as studentGrowthPercentiles and studentGrowthProjections. This vignette walks through the process of importing this exemplary data set into your data management system and setting up it for SGP analyses.
sgpData is an anonymized, panel data set containing 5 years of annual, vertically scaled assessment data in WIDE format. This vignette provides an example of how to use sgpData with the SGP package for the purpose of calculating student growth percentiles and projections. The first column in sgpData, ID, provides the unique student identifier. The next five columns, GRADE_2013, GRADE_2014, GRADE_2015, GRADE_2016 and GRADE_2017 provide the grade level associated with each of the students assessment occurrences. In most cases, students do not have 5 years of test data so the missing values (NA) will appear in those rows.
In addition, the sgpData_INSTRUCTOR_NUMBER column in sgpData provides an anonymized lookup table that allows you to associate each students test record with the teacher that administered that particular test. This is particularly useful for determining which teachers might be responsible for students performing poorly.
An important point to note is that sgpData only contains assessment data for students who have taken two or more tests in the same content area during one testing window. It does not include information about held-back students who must take a minimum of two tests in two separate testing windows before having an SGP calculated for them.
Managing SGP analyses with sgpData is, in general, straightforward. However, it is important to understand that there are a lot of details in preparing this type of data and to pay careful attention to all the steps. For more information, please consult the SGP Data Analysis Vignette or the detailed documentation provided by the SGP package. SGP is a very flexible tool and, with careful planning, can be used for all kinds of student performance analyses. SGP also makes it possible to compare students from very different schools and districts in order to identify best practices in a given state or region. This is a major reason why many of the SGP user community have chosen to make sgpData available for public consumption. We hope you find sgpData helpful in your efforts to improve student achievement. Please share your ideas with us on how we can continue to improve this resource. Thank you!