What good is high school testing that does not count towards students’ grades, cannot be used for college admission or placement, and is barely related to things students are learning in their current classes?
Dr. Scott Marion, Executive Director of the Center for Assessment and a national leader in educational assessment has been asking that question for a long time. In this guest blog, he explains the purpose of high school testing, identifies types of tests, and explores how we can make testing more equitable and meaningful for students, educators, and schools.
What is the purpose of testing and school accountability?
Every school is required by federal law to administer assessments of English language arts and mathematics each year in grades 3 through 8 and once in high school. In New Hampshire, districts use the NH Statewide Assessment System for the elementary and middle school testing and the SAT for the test in high school.
Dr. Marion explains:
Simply saying the purpose is school accountability does not offer enough information to guide test design. What is the purpose of the accountability system?
The Elementary and Secondary Education Act, of which the Every Student Succeeds Act (ESSA) is the latest version, has a clear focus on evaluating and enhancing equality of educational opportunity. Therefore, an assessment system in support of such an accountability system must provide information about the extent to which students have an opportunity to demonstrate learning of the intended curricular goals (e.g., standards).
An accountability system focused on prioritizing excellence would likely lead to a different sort of assessment system. Many states are trying to promote both excellence and equity, an admirable endeavor, but this means that state leaders have to be exceptionally thoughtful about the assessments they employ…
There are two basic types of high school testing: Single grade (e.g., 11th grade) survey tests or end-of-course tests. As the name implies, end-of-course tests are tied to specific high school courses (e.g., American Literature, Life Science, Algebra) where only those students participating in the course sit for the exam. A survey test is administered to all students in a given grade designed to broadly cover the grade level or grade span content standards in that subject area. The use of a “nationally-recognized college entrance exam,” when used as the achievement indicator, is a special case of the survey test…
What about the SAT as a statewide test?
Dr. Marion writes that the Every Student Succeeds Act (ESSA), the federal education law that replaced No Child Left Behind (NCLB), allows states to use a college readiness exam like the SAT or ACT as the statewide test for high schoolers.
Marion is concerned that they are not aligned to statewide standards, which may give districts an inaccurate picture of student performance because they’re not being tested on the benchmarks that the district has agreed that all students should know:
In spite of claims made by the companies, the few independent alignment studies conducted to evaluate the relationship between the ACT/SAT and state content standards have questioned the match between the tests and the standards students are expected to learn. Assuming these alignment studies are generally accurate, this would mean that only part of the standards would be tested and many worry, justifiably, about the narrowing of the curriculum because of the “what gets tested, gets taught” phenomena. This is a legitimate concern and violates the promissory note discussed earlier. Further, under a standards-based approach, the use of non-aligned tests may challenge the validity of school accountability inferences.
However, choosing a test that over half of the students would likely take anyway is a good alternative to giving a test that doesn’t count towards their grades or college admissions:
Having received first-hand many of the complaints about high school testing such as “students are just drawing Christmas trees on the answer sheets,” I argue that, in many cases, states are making a rational decision to use the ACT or SAT as the achievement indicator. While many state policy makers do not understand or do not care about the alignment concerns, they are happy to reduce some testing and provide a visible benefit to many of their constituents.
Further, there is some evidence and many anecdotes supporting “diamond in the rough” stories where a few students with poor grades and lack of school performance score surprisingly well on the college entrance exams. School and district leaders are more aware of the alignment issues, but if they are going to be held accountable for the achievement of their students, they would rather have the 50-70% of students considering attending college take the test seriously than not.
What can we do to make testing more equitable?
According to Marion, high school testing should go beyond the SAT or ACT. These tests were designed to measure college readiness and aren’t always aligned with state standards, so they often exacerbate gaps in achievement. Not all students will attend college, and state standards are written to prepare students for whatever they choose to do after high school–college, career, service, and more–so what students actually learn can be different than what’s tested on the SAT or ACT. There are also concerns about test accommodations for students with disabilities and English Language Learners with those specific tests.
Using a single test (a survey test) that’s administered to all students at the end of the year in 9th or 10th grade that is tied to state math and English standards can help alleviate concerns with equity because they can provide a better common measure of what students should have learned. They also provide feedback sooner than a student’s junior year, allowing educators to address learning gaps early on.
There are challenges to these approaches, Marion warns. Specifically, with math, students often have different tracks–some take algebra in 9th grade, others begin at pre-calculus or a fundamental math course–making one, uniform test for all students difficult.
Measuring student progress and growth should go beyond one or two assessments, writes Marion:
[T]o move toward more meaningful and deeper learning opportunities, students should be expected to engage in opportunities such as senior exhibitions, pursue internships or other extended learning opportunities, and/or complete rich performance tasks characteristic of New Hampshire’s Performance Assessment of Competency Education initiative and Wyoming’s former Body of Evidence System. Of course, there is nothing stopping states and districts from pursuing such approaches now, but if such activities are not included in the ways in which high schools are held accountable, it is easy for such ambitious efforts to fall by the wayside.
Dr. Scott Marion, Executive Director of the Center for Assessment, is a national leader in designing innovative and balanced assessment systems. He has served on multiple National Research Council (NRC) committees, including several to identify best practices in state assessment systems. His work has been featured in dozens of peer-reviewed journals and national conferences. He also serves on the Rye School Board. Contact: firstname.lastname@example.org @ScottFMarion