The Use of Student Growth Percentiles (SGP) As an Indicator of Student Achievement
The use of student growth percentiles (SGP) as an indicator of student achievement allows educators to communicate student academic progress in terms familiar to teachers and parents. In addition, SGPs provide a framework for discussing student learning that goes beyond just test scores. For example, a low SGP might reflect poor instruction, while a high SGP might indicate effective teaching. This allows administrators and teachers to identify instructional challenges in a way that is not biased by the impact of student background or outside influences such as family income or after-school programs.
SGPs are more accurate than value-added models (VAM) in predicting student achievement and fit well into accountability systems that emphasize student growth over test score measures. Moreover, they are easier to interpret than averages and do not require the assumptions that go along with the calculation of median and mean measures.
A key feature of SGPs is that they are calculated for each individual student, allowing districts and schools to make comparisons among students who started the year with similar knowledge, skills, abilities, and grades. These analyses can help determine the effectiveness of educational programs and identify students that need special attention.
However, SGPs do not represent the full picture of student learning and may be influenced by factors that are out of the control of school personnel. For this reason, they do not fully measure academic performance. In addition, they may not always be related to underlying causes of student achievement such as poverty or family background or by school policies such as teacher quality, curriculum, or school culture.
Nevertheless, SGPs have several strengths over other measures of student performance and should continue to play a role in school improvement efforts. Moreover, they can be used to evaluate the effects of various school interventions and policies on student outcomes, such as providing quality instruction or after-school activities.
In order to conduct SGP analyses, schools must have access to a long dataset of student assessments with raw and percentile scores. The minimum dataset required is sgptData_LONG, which provides 8 windows (3 annually) of assessment data in long format from the Early Literacy, Mathematics, and Reading content areas. In addition, the state level student aggregates needed for SGP analyses are available through sgpstateData.
A common method for aggregating SGPs is to use median measures, which represent the middle value of each group. However, this can lead to misleading conclusions about a program’s effectiveness. For example, accelerated programs might have a high median SGP because they push many students to keep up with the rest of their class. However, these students might not have gained as much as other students who were taught by less effective instructors. This makes the median SGP misleading and does not align with the Department’s guiding philosophy that all students contribute to accountability results. Therefore, the Department is changing from median SGPs to means as the primary summary measure for SGP analysis. These new summary measures will be introduced in the MCAS 2.0 assessment system starting this fall.