Standards-Based Grading Only Solves Half The Problem

The debate dividing the education community is over how to best grade students. While grading standards are important, they can only solve half of the problem. The other half comes down to giving feedback that will motivate learners and improve their future performance in school or elsewhere throughout life.

Standards-based grading is a method of grading that has been used for years. It allows teachers to grade students on their mastery of a standard, rather than how well they did on a test. However, there are limitations to this system that make it difficult to use in many cases.

Standards-based assessment only solves half the problem

Grant Wiggins, Authentic Education

Editor’s comment: Grant Wiggins is at 26. Passed away in May 2015. Grant has had a major impact on TeachThought’s approach to education, and we welcome his contribution to our website. From time to time we will go back and share his most memorable messages. This is one of those positions. Fortunately, his company, Authentic Education, is continuing and expanding Grant’s work.

This article was first published in 2014 and updated in March 2020.

Over the past few months, I’ve worked with a number of middle and high schools where assessment and evaluation practices just don’t work in a world of standards. Schools do not make on-site assessment rigorous enough, fearing that pupils with poor grades will be demoralised. The solution is simple: Don’t translate the points into notes indiscriminately.

The problem is that schools have to meet the standards and local assessments have to prepare kids for the standards that PARCC and SB are reviewing. But the new tests are more complex and rigorous than most local tests. Therefore, the values must be low. (Unfortunately, anyone who follows NAEP results has known this for years). This seems to be part of a long tradition of assessment that we don’t want to punish children for getting bad grades (similar to the outrage at the results of schools that failed this year).

Yet there seems to be no alternative: Increasing local performance standards means a significant reduction in pupils’ grades. Or we could maintain the current average grade of B for students at the local level, but then we would be less rigorous than necessary in preparing children for tests – and in predicting test scores (which local assessments should certainly do if they are valid and useful).

It should be noted that so-called standards-based assessment alone does not solve this problem. Moving to standards-based assessment does not mean that the assessment will be rigorous. In fact, if you look at schools that use standards-based assessment, it is very rare for students to get grades that differ significantly from the range of grades (i.e., alternatives to letter grades) in those schools for. i.e., we do standards-based assessment in the standard! The local mistake was to assume that the standard assessment was sufficient to ensure rigour. But it’s not enough, it can’t work on its own.

What is seriousness?

The rigor of the teaching was not noticed. So it is not determined by comparing education to standards. Strictness is determined by our expectations: how we assess and grade students’ work. This means that accuracy is determined by means of three different evaluation elements:

Complexity of the task or questions

Complexity of the criteria defined by the headings

Expected level of performance, determined by anchor points or success rates.

Currently, many districts and schools do not even meet the first criterion. When my colleagues and I review local tests, the tests generally appear to be much easier than tests reviewed from outside – even in districts of good quality. The usual explanation? The problem of fair assessment.

It also follows from these three elements that even complicated rubrics about homework and quality are not sufficient to ensure rigor. The task may be challenging and the criteria demanding, but if the expectations for the student’s products or outcomes are very low (as defined by specific models or local standards), the assessment is not rigorous. That’s why a score of 40 or 50 on state tests is a terrible solution – if the goal is to report standards-based results, not to find a way for most kids to succeed.

Think of the high jump or pole vault in athletics: You can compete in a difficult discipline and be judged on real criteria, but if the height you have to climb is absurdly low, then the judging is not rigorous – even if it is standards-based testing and scoring.

Troubleshooting

One of the solutions? Avoid unnecessary calculations based on false equivalents. The solution lies in athletics: We don’t need to calculate a score for an athlete by mechanically converting his jump height into a score based on an arbitrary but easy-to-use formula, nor do we need to. This will significantly reduce scores and strongly encourage less qualified athletes.

Instead, we evaluate progress and results based on the level of initial jumps and aim for appropriate growth based on growth efforts and gains. (I blogged about it extensively here and here.) However, expectations for all jumpers are high and still rising.

The same solution is needed at the local level in academia, if real standards are to be applied to make students aware of their situation without pushing them away. (This is the idea behind OLMS and CIS in many states). Therefore, their work should be reviewed several times a year against external standards (defined by quality tests and examples of student work). But we have to enter grades into our online grade book throughout the year! I know. But instead of turning their results into mind-numbing formulaic evaluation, we use our wisdom and judgment to account for equity, growth, and effort on a consistent basis.

For example, suppose that in testing writing against national standards we link the testing to national samples of published tests. Next, let’s assume that a 6-point heading is used. Now suppose that on the first assessment, say in October, almost all students get a 1 or a 2 (these are the lowest scores on the scale). This is what we can say before the results are released to students and turned into grades:

Boys, I’m giving you points for the best essay in the whole state. That way, your first course of the fall will reflect an accurate assessment of your current situation. A 1 would be a 10. A grade of 2 would be a B+. Any grade above 2 is an A – for the first semester.

The next semester, in the winter, you will have to go higher to get the same grades. And in the spring, you have to go up two numbers to make those numbers.

This is already happening, of course, in AP and IB classes. So it should be relatively easy to do all the topics. Here’s how we solved the problem: Assessments became fair, standards became clear, and incentives emerged to get better over time.

This article was updated in March 2020

This article is an adaptation of a post that first appeared on Grant’s personal blog; Grant can be found on Twitter here; Standards-based assessment only solves half the problem.

A common practice in education is to grade students on a letter scale. This method is flawed, as it only solves half the problem. Standards-based grading would be a much better option. Reference: standards-based grading to letter grade.

Frequently Asked Questions

What is wrong with standard based grading?

A: Standard based grading is a system that automatically grades students on what they know rather than the effort made. This can be an issue when students are rewarded for memorizing content, and not understanding it in its entirety.

Is standards based grading effective?

What is the purpose of standards based grading?

A: Standards based grading is a method of evaluating students or other individuals performance in which the standard used to evaluate performance varies.