Test Better, Teach Better: The Instructional Role of Assessment

by W. James Popham

If you're frustrated by standardized tests that don't give you the information you need to help students achieve in the classroom, then here's a book that explains how to create and use tests to guide everyday teaching practices. A renowned expert on assessment provides you with a "crash course" on the basic principles of How tests can tell you what to teach and how to teach it. What to put on a test and why, including the rules for choosing and writing good test items. The measurement concepts every educator must know in order to design tests that meet tough accountability standards. How to avoid "teaching to the test" and five common mistakes in test writing. Sample test items, tips, and steps guide you in collecting the right testing data, interpreting it, and making sound judgments about whether your instructional practices are achieving the results you want.

Genres: Education, Teaching

175 pages, Paperback

First published January 1, 2003

Reviews

[Melissa Marie · Rating 2 out of 5]

2.5 honestly. I skimmed through it and most of it I already knew or don’t feel I can use. 🤷🏻‍♀️

[Joey · Rating 3 out of 5]

I only continued reading this because I had to, otherwise I would not have trudged on.

[Chrissy · Rating 5 out of 5]

In Test Better, Teach Better: The Instructional Role of Assessment, Popham covers the role of assessment in instruction in very clear, plain language. Although the book is somewhat outdated (still refers to No Child Left Behind), the information that it gives on the use of testing in the classroom to inform and improve instruction as well as the information that it gives on standardized testing is still very current. One thing that Popham is very clear about is that through testing, we cannot truly get an accurate test of any one thing, we can only make an inference from a student's score on a test. Overall, he gives a clear overview of testing in the classroom - including pre- and post-tests, how to construct items, reliability, validity, and how tests can clarify the curriculum. The book is very easy to read and understand as he puts everything into layman's terms and uses lots of examples. The book is an excellent introduction to assessment in the classroom.

[gdota2 · Rating 3 out of 5]

Nice

[Ryan Young · Rating 5 out of 5]

why do we assess students? what are we trying to learn from them? how should they be designed? what should we do with the results? find out!

[Shifting Phases · Rating 4 out of 5]

This is a readable intro to test design. Author W. James Popham is careful to explain the difference between test scores and the judgements we make based on test scores. He also explains the conceptual framework of statistics that is needed to make sense of these ideas (i.e. reliability vs. validity), without getting bogged down in technical jargon or formulas. Lots of his pointers are very concrete, although some seem painfully obvious (don't use double-negatives in test questions; when writing multiple choice questions, don't give away the answer by using the article "an" when only one option starts with a vowel). But many of suggestions were helpful to me, such as how to use tests to determine what to teach/how to teach, or thoughtful examples of how to make a cursory reading lead to a wrong answer without making the questions too complicated. The guidelines for rubrics were quite good as well (can students use it to self-assess? Is the rubric general enough to be recycled for other assignments? Are the criteria actually teachable?). I also really liked the section on portfolio assessment. One suggestion that I will use is to require students to practice assessing themselves and each other using a rubric, and then to submit a filled-in rubric with their portfolio assignments, so that self-assessment can be part of our conversation. There's a good explanation of how to create your own Likert inventories, as well.

Popham advocates for standards-based testing that is reported by standard (rather than in an aggregate). There's a great chapter that briefly explains the politics and economics of standardized tests. He pulls no punches in criticizing inappropriate applications of standardized tests. But he also includes a chapter on how to create your own standardized tests (invite parents to help vet the test; choose a few high-priority standards to assess; report the results by individual standards; consider using a split-and-switch design when comparing post-tests to pre-tests). The book paints a clear picture of how to assemble credible evidence of instructional effectiveness, even when read by a non-specialist like me.