When we watch the Olympics, we see athletes performing at breathtaking levels of mastery — and yet the performances seem effortless. And even if we spectators are ignorant of the intricacies of a sport, we’re still able to be awed by a virtuoso performance and unimpressed by a clunky one.
Similarly, when we review test items, we don’t need to know every intricacy of item writing to be able to identify the good items. But unlike our experience of watching the Olympics, we most likely view the good items without awe — without even a comment. All we notice is that this particular item does what it’s supposed to do.
What we notice, what stops us in our item-reviewing tracks, is that occasional clunky performance: the fall on the ice. The item might trip up through imprecise or redundant language, awkward syntax or nonstandard usage; the focus might be fuzzy, or the item might inadvertently muddle into assessing knowledge other than what is required by the designated standard. The language used in the item might be wildly sophisticated or far too simple for the grade level. There are various roads to ruin when it comes to item development. After that long list of negative possibilities, let’s look at the positive. What makes a good item?
In order to describe a good item, we must first identify the purpose of the item: to measure either knowledge of a specific concept or mastery of a skill. A good item provides the test-taker with the opportunity to show what the test-taker knows and can do and accomplishes this purpose without putting any obstacles in the test-taker’s path.
A good item has these traits:
- Content validity: The item measures exactly what it is intended to measure.
- Appropriateness for the grade level: The item is suitable for test-takers at the intended grade level in all aspects: content, readability, syntax, vocabulary, and demands on the test-taker in terms of developmental abilities and cognitive processes.
- A solid framework: The ideas in the item are organized; the item format is intentional and supports the content.
- A sound premise: The item addresses worthwhile content in a manner that is accurate and logical.
- A clear focus: The item presents a task that is readily understood by the test-taker, regardless of whether the test-taker possesses the targeted skill or knowledge.
- Precise language: The language used in the item is clear, brief and specific.
- Free from error: An item with errors will lead to confusion and will not measure what you want to measure.
A reviewer might not notice if an item has these traits, but the same reviewer will definitely notice if an item does not have them.