ASSESSMENT STANDARDS

    MAKING VALID INFERENCES

    Quality assessment leads to valid inferences about student learning.

    Quality assessment leads to valid inferences and a clear and accurate picture of learning. An inference about learning is a conclusion about a student's cognitive processes that cannot be observed directly. The conclusion is based instead on the student's performance. Many potential sources of performance are available. Mathematics and science assessments include evidence from observations, interviews, open-ended tasks, extended problems and investigations, portfolios and exhibitions, as well as more traditional paper and pencil forms of assessment, such as multiple-choice and short-answer tests.

    Valid inferences about learning are promoted when:

    • the purpose of assessment is clear and known to students,
    • the learning outcome is clearly defined in expectations about essential knowledge, capabilities, habits of mind and values,
    • multiple forms of assessment are used as sources of evidence about learning and are appropriate for the purpose of the assessment,
    • there is a sufficient sample of student work to provide quality evidence that is adequate and relevant to the outcomes,
    • teachers are knowledgeable about assessment and their professional judgment about the quality of student work is based on well-defined criteria,
    • biases that threaten validity have been eliminated or minimized,
    • students have alternative ways for communicating their learning.

    A quality assessment process informs students of the expectations for success in mathematics and science­as defined in the Pacific Standards for Excellence and local performance standards­and ways in which they are expected to demonstrate their learning.

    Because single assessments often yield an incomplete or inaccurate picture, quality assessment utilizes multiple samples of student work to get a valid picture of student learning. Multiple assessments enable students to show their understanding through a variety of learning styles and engage more than a single intelligence. This variety allows for weaknesses in one assessment to be compensated by strengths in others, resulting in a more detailed and complete picture of mathematical and scientific literacy and leading to valid inferences about student learning. The existence of biases in the assessment process masks learning and leads to inaccurate portrayals of students and classrooms. Quality assessments are carefully designed so that context, language, culture, gender, preconceptions on the part of the teacher, and other factors are not allowed to interfere with fair and accurate judgments about student learning. Such designs ensure that assessment leads to fair and valid inferences about learning. In the Pacific region, language often inhibits response. Quality assessment is carried out in the language of instruction at a level matching the language development of the students and is designed so that language is not a barrier to displaying learning.

    New forms of assessment require increased attention to procedures for making valid inferences about the mathematics and science that students know, can do, and care about. Assessments based on a framework of important mathematics and science, draw on multiple sources of evidence, minimize bias, and support student learning in providing the evidence needed for valid inferences.