Data sgp is an easy-to-use online tool for organizing longitudinal (time dependent) student assessment data into statistical growth plots. It includes two common formats: WIDE, where each case/row represents a single student and time dependent variables are associated with them at different times; and LONG, where time dependent variables are spread across multiple rows per student. The tool also provides a number of examples of WIDE and LONG formatted data sets (sgpData_WIDE and sgpData_LONG) to assist users in setting up their own data sets.
The data sgp package also facilitates the creation of individual student growth percentile reports. These are useful for highlighting student performance relative to their academic peers and can be used in a variety of ways to improve instruction. To generate these reports, the sgpData_PERCENTILE function is used with a specific data set/function combination (sgpData_PERCENTILE_DATA_SET and sgpData_PERCENTILE_FUNCTION) and a grade level range from 2013 – 2017 (sgpData_GRADE_RANK). The results are displayed on screen or printed on a report.
In addition to growth percentiles, sgpData also provides several other statistical indicators of student achievement and progress. For example, the sgpData_CONTENTS table provides information about how much of a content area was assessed on each assessment, which can be helpful for curriculum coordinators in identifying gaps in coverage. It also displays the mean and standard deviation scores for assessments, which can be used to detect patterns of missing or struggling students.
This analysis is an important first step in establishing achievement standards for students that are both realistic and attainable. This step will also help identify the factors that must be addressed in order to achieve these goals, such as the importance of addressing absenteeism and the need for improved instructional practices.
An important issue in interpreting SGPs is that they are error-prone measures of a student’s latent achievement traits, which are defined by their prior and current test scores. The errors from both the prior and current test scores, as well as the correlations between the prior and current test scores of a student, result in SGP estimates that are noisy. This variability makes it difficult to interpret estimated SGPs aggregated to the teacher or school levels.
Further, research has shown that these noise sources in SGPs are related to student background characteristics. Thus, aggregating estimated SGPs to the teacher level may obscure true student growth and mask important differences in student achievement. This is an especially important concern for students from low-income families, who are likely to have more difficulty achieving academic growth due to the lack of resources and educational opportunities they experience in their lives.