I recently came across a blog post stating something that seems logical and helpful. Before we start working with students to help improve math performance, it noted that we need to test them to know the areas to focus on. A quick search on the internet reinforced this. A number of sites discussed different types of assessment: diagnostic or prior assessment, formative or ongoing assessment, and summative or post assessment. But when looking for examples of each, ongoing formative assessment received less attention compared with more examples each of diagnostic and summative testing.

This reminded me of when I made a presentation before a state curriculum adoption committee. I explained that since my company writes intervention materials designed to support students struggling in math, formative assessment is built into the fabric of each lesson in the program. While there are pre, post, and mid-chapter tests, the emphasis is on the problem solving support for these students of ongoing formative assessment. One member of the committee laughed and said, "Well that requires good teaching!"

It appeared that everyone in the room recognized that good assessment is systematic and ongoing. But several members of the committee saw repeated testing as the means of insuring ongoing assessment.

I won't go into possible reasons for the apparent preference for testing. Rather, in this article, I want to lay out some problems with relying on testing as a primary means of assessment, especially for students challenged by math.

Going back to the opening paragraph, diagnostic or prior assessment is essential. We're flying blind without it. But favoring tests over integrated ongoing assessment may miss part of the potential power and value of assessment and even create some of the very problems our better teaching are trying to address.

Students who are not doing well may not give us the best indication of what they know and can do by taking another test. Continuing to write incorrect answers can reinforce memories of those wrong answers, reinforce faulty procedures leading to the incorrect answers, and reinforce partially understood or misunderstood concepts. Having to leave some answers blank can reinforce a negative self-image, and frustration with and resistance to the study of math. It also means that needed practice on those problems is missed. When the test is graded by the teacher and handed back at a later time, the students are no longer engaged in the problems and many will not retrace their steps in arriving at the wrong answers.

Assessment has multiple uses for teacher and student. It gives the teacher some evidence of student preparation and performance, which informs what is to be taught and how it might best be approached. It also provides students with feedback. While testing is an essential component, should it be the primary form of feedback for the intervention teacher? The teacher may not know from a test, which problems the students labored over or solved quickly whether they got them right or not. The test may not reveal which prior conceptual, procedural, or simple nomenclature gaps led to an incorrect answer. Depending on when a test is given, it may be too late to go back and address certain problems. And because of all of this, it may not be the best form of feedback for students struggling in math either.