Data SGP software enables us to track and assess student growth over time. This enables us to better understand our educational system’s strengths and weaknesses and make informed decisions on how we can enhance it, as well as develop strategies to assist those not meeting academic standards – for instance using student growth percentiles we can compare one student’s progress against that of similar prior test score students (their academic peers), providing us with a way of fairly comparing students who enter school at various levels.
Data SGP can also help us monitor the effects of new programs and interventions on student learning, particularly when we evaluate programs that have yet to be implemented. An initial pilot can often produce inaccurate or even misleading results if we fail to carefully examine all data sources to detect any discrepancies – therefore it is essential that we use high quality data SGP available.
The SGP package contains several functions to simplify data preparation and analysis processes associated with SGP analyses. These functions include prepareSGP which takes longitudinal student assessment data in WIDE or LONG format from either WIDE or LONG files (sgpData_LONG or sgpData_WIDE), converts them to Demonstration_SGP or sgpData_INSTRUCTOR_NUMBER objects with all pertinent meta data embedded; createSGP automatically generates new files by automatically creating them through prepareSGP automatically creating new files eliminating most tedious data preparation involved with SGP analyses
analyzeSGP is designed to conduct SGP analyses for every year and content area, including calculations such as student growth percentiles, baseline student growth projections and lagged student growth projections. Furthermore, an SGP matrix displays distribution of student growth projections across all years and content areas.
combineSGP – This step combines the results from analyzeSGP into the master longitudinal record Demonstration_SGP, consolidating student growth percentiles and projections into one file per year.
SGPs may be affected by various factors, including design of the baseline cohort, teacher or school characteristics and duration needed for creating a baseline-referenced SGP. For optimal results, SGP analysis should be applied across multiple years and standardised by measuring scale used to report results. To do this, student assessments from multiple years should undergo SGP analyses. As part of their educational experience, all baseline cohort students should have similar instructors throughout their education experience in order to reduce any false correlations due to teacher or school characteristics. Due to this reason, the SGP package recommends using LONG formatted data for operational SGP analyses as it provides more flexibility and ease of administration than WIDE formatted data. Higher level SGP functions have also been tailored to work efficiently with long formatted data files, taking full advantage of embedded SGPstateData meta-data in sgpData_LONG files to facilitate easier use and expansion of existing analyses with additional years of data.