Average Student Test Score (PI9024)

This metric is intended to capture the average percent of total possible points students earned on tests during the reporting period. For example, if Student A earned 75/100 points and Student B earned 30/100 points, the Average Student Test Score for these two students would be 52.5/100 points, or 52.5%. Students who were not tested should not be included in this calculation.

In many cases, test-score data are collected by schools (public or private) or another third party.

In educational contexts, Client Individuals: Total (PI4060) can refer to students and may be used in this calculation in place of School Enrollment (PI2389). Organizations may also disaggregate data by grade level as needed to identify underlying patterns.

In all contexts, tests should reflect learners' social, environmental, and learning norms and needs. Organizations are encouraged to footnote details on how and by whom tests have been developed, as well as how they are fit to context.

Average Student Test Score (PI9024), combined with factors included under "HOW is change happening?," gives a directional indicator of student learning outcomes. In some contexts, however, Student Promotion Rate (PI4924)—which measures the number of students proceeding to the next grade—may be a more appropriate outcome metric. In these cases, Student Promotion Rate (PI4924) should be substituted for Average Student Test Score (PI9024) under WHAT and HOW MUCH. For more details on calculation, see usage guidance for each specific metric in the catalog.

In some contexts, this metric can serve as an indicator of whether the outcome being sought by an investor or organization is occurring (the WHAT dimension of impact). For more on the alignment of IRIS metrics to the five dimensions of impact, see IRIS+ and the Five Dimensions of Impact (https://iris.thegiin.org/document/iris-and-the-five-dimensions/). No single metric is sufficient to understand an impact; rather, metrics are selected as a set across all dimensions of impact. When possible, the selection of metrics to measure and describe the five dimensions should be based on best practice and evidence.