When analyzing student growth, it is often necessary to compare the performance of individual students to their academic peers. In order to do this, it is important to be able to access data with a variety of percentile tables. Fortunately, the sgpData package provides a wide variety of these tables. One of the most useful is sgpData_STUDENTS_PERCENTILE_TABLE which contains percentile information for every student in a given cohort. This table can be used to identify students who need additional help and to monitor student progress over time.
Another useful table in sgpData is sgpData_INSTRUCTOR_NUMBER. This table identifies instructors for each test record, making it easy to determine the teacher responsible for most of a student’s learning. This is especially helpful for identifying areas of weakness and for gauging teacher effectiveness.
In addition to the percentile tables, sgpData also provides easy-to-use lookup tables for associating instructors with test records. This is particularly valuable for teachers who are evaluating the effectiveness of their own teaching methods. The first column in sgpData, ID, is the unique student identifier. The next five columns, GRADE_2013, GRADE_2014, GRADE_2015, GRADE_2016 and GRADE_2017, provide the grade level of each student assessment score in each of the past 5 years. In the event that a student does not have 5 years of testing, the system will display a missing value (NA).
For most analyses, it is best to use the LONG format for the sgpData data. This format is used by the lower level functions, such as studentGrowthPercentiles and studentGrowthProjections whereas the higher-level wrapper functions utilize WIDE data formats. The decision to format in either the WIDE or LONG format should be driven by numerous factors, including the ease with which you can prepare and analyze the data, the amount of storage space available, and whether or not the analysis will be run operationally year after year.
The sgpData package has a number of other helpful tools for analyzing SGP data, including the ability to quickly calculate student-level averages, percentiles and trend lines. However, it is important to remember that these tools are only as good as the quality of the underlying SGP data. For this reason, it is important to review the sgpData documentation carefully before using these tools.
If you need more assistance with SGP data analysis, please consult the sgpData analysis vignette. The author, Mark Hauk, is a Senior Data Analyst at the University of California, Berkeley. He has extensive experience with statistical software and is familiar with a variety of data formats. He has worked extensively on the development of the SGP package. He can be reached at haukm@berkeley.edu. He would be happy to answer questions regarding the SGP package or to review a particular dataset. You may also contact the SGP support team through our online help desk. We are available 24/7.