After a series of formative assessments that provide opportunities for students to receive quality feedback, they need a chance to bring multiple concepts and skills together through a summative assessment. This assessment is a chance for students to apply and prove their knowledge, not a chance for teachers to catch students off guard. Teachers should consider the validity, reliability, and security of the assessment when writing the assessment and when reviewing students’ results.
Valid assessments are those that target the content by using multiple forms of questioning and at different levels of difficulty. Teachers can use Depth of Knowledge (DoK) to gauge the difficulty level of the questions. It describes how many levels of thinking a student must perform in order to answer a question, and by matching questions to a DoK level, a teacher can ensure that the test questions vary in complexity. Below are questions that assess the student’s knowledge of Greek and Latin word parts. Notice how DoK 1 questions require the student to recall information. The DoK 2 question requires the student to apply the knowledge of “sub” to the reading passage and the word “order.” Then in question 11, the student must know the definition of “less,” and use it to describe a baleen whale as it is defined in the reading passage.
Tests must also provide reliable results that are fairly graded and weighted appropriately to match the test’s level of difficult and effort. Test questions can be weighted to make questions that involve more analysis worth more, and rubrics can be used for projects and writing assignments to ensure fair grading.
The last consideration is the security of the summative assessment. In a physical classroom, teachers can separate desks, use privacy screens, and monitor the class room to prevent cheating, but in a virtual classroom, these options are not available. The major concern when administering a test online is preventing students from looking up answers during the test and/or sharing answers with the class. To prevent students from looking up answers during the test, the teacher should set a time limit that is strict and short so that students do not have the time to look up answers. To prevent students from sharing answers, teachers can create multiple versions of a test drawing questions from a large question bank, and some software does this automatically. Finally, students should only be able to complete the test once. After the test, the teacher can look for standard deviations, and converse with students individually if there are too many deviations.
Below is one example of a summative assessment that test students’ knowledge of text organization, word parts, and sentence structure. A variety of question types gives students multiple exposures to each skill, and the test includes a variety of DoK 1, 2, and 3 questions to assess whether students have mastered the content.