Having worked both in education as a teacher, and in industry as a developer of certifications, I understand the prevalence of multiple-choice tests. They are easy to administer and grade without the specter of subjectivity. But it is not simple trying to make a question of this type assess anything other than memorized information. So often I have seen standardized tests in which “higher thinking skills” were supposed to be determined by multiple choice questions.
I recall a standardized grade school reading test with an example of many “higher level” questions that failed to hit the mark. In one case students were suppose to infer that the main character in a writing selection thought the vacant lot without a building was ugly. The rationale was the author had described the street as looking “like a child with a gapped-tooth smile.” However, the students being tested were of the age in which children lose their baby teeth. Were they supposed to assume that everyone their age was ugly? The adults who wrote the test forgot what it was like to be a child and came up with an invalid question in an attempt to assess “higher thinking” skills.
Because test creators know more than the students for whom they are creating the test (at least most of the time) they try to make questions harder using subtle differences between correct and incorrect answers. Vague wording with more than one interpretation, or using unusual phrasing which is harder to read end up turning questions on math or science into reading comprehension questions which are not really valid indicators of either math or science ability. Two errors that occur most frequently in these kind of questions are two right answers or no right answer.
Of course, with all these possible problems, the companies creating these standardized assessments need a way to determine the validity of the questions. How do they do this? By putting field test questions in the standardized tests currently given to students. Some standardized tests contain many questions that don’t even matter as far as the score. They are included as samples to get reliability and validity statistics. Then, that particular type of question can be used in the future. I discovered that removing the field test questions from the TAKS was one of the modifications used for learning disabled students in Texas.
So field test questions do affect student performance. Think about it. You are a students are already stressed about the all-important standardized multiple-choice test and you run into questions that makes no sense. You have no way of knowing if it is your deficiency or one of those field test questions. Now, this same method of “testing the test” is being included in the Common Core assessments.[1] Do we really want our students to be guinea pigs?
Photo by Tukka (Public Domain)
Write about what? 