True False and Short Answer Questions

True/false questions (TFQ) are one form of binary choice question and may be considered a variation on the multiple-choice question (MCQ) format, a version with only two response options (Parkes & Zimmaro, 2016). This type of question can be very useful in situations where there can be a true binary choice (e.g., true/false, yes/no, fact/opinion) and, like MCQ, can assess simple to more complex learning outcomes (e.g., knowledge, application, analysis), can sample a large number of content domains, and can be easily scored (Gronlund, 2006).

As with MCQ, developing good TFQ, particularly ones that assess higher order learning, can be difficult and time consuming. Other limitations of TFQ are that scores can be highly impacted by guessing, incorrect responses provide little diagnostic information, and keying the answer as false does not provide any indication that the students know the correct answer (Gronlund, 2006).

Well-written TFQ are short and simple declarative statements that include only one primary concept and can be judged as categorically true or false. When writing these questions, Gronlund (2006) recommends (1) avoiding negative statements (e.g., including “no” or “not”) as well as double negatives to reduce possible confusion; (2) citing a source for opinion-based questions as opinions are not inherently true or false; and (3) avoiding using words that may signal the correct answers. For example, items with the words “all” or “always” are generally false and those with “may” and “sometimes” are generally true.

As with MCQ, TFQ can assess higher order learning outcomes (e.g., application, analysis) by using context-dependent item sets and vignette-based TFQ (Parkes & Zimmaro, 2016). Context-dependent item sets are a series of TFQ that require students to answer a set of questions about a stimulus such as a table, graph, picture, screen shot, video, or simulation. For vignette-based TFQ, the stem of the item includes a scenario or short case that reflects a real world situation consistent with the course learning outcomes. The stem is followed by one or more TFQ addressing the scenario (see the section on higher order MCQ for more information about these two forms of questions).

As indicated above, guessing is a major concern with TFQ. Methods to discourage guessing such as Negative Marking (e.g., a correct response receives 1 mark, skipping the questions receives 0 marks, an incorrect response receives -1 mark) are debated in the literature. Criticisms of Negative Marking include that negative marking may add stress to the testing situation for students and that students’ test-taking strategies and tendency to take risks may play a role in how they answer, not just their learning of the content (Parkes & Zimmaro, 2016; cf. Burton, 2005).

Short Answer Questions

Short answer (SA) questions require the respondent to provide one or more words, numbers, or other symbol to reply to a question or complete a sentence. These items are generally easy to write, reduce guessing relative to MCQ and TFQ, and can be used for computational problems. On the flipside, it can be difficult to write good SA questions with only one possible answer, scoring can take time, student spelling can be an issue in grading answers, and it is difficult to write SA questions that assess higher order learning (Gronlund, 2006).

Well-written SA questions are direct questions or clear incomplete sentences for which only one short correct answer, which addresses the main point of the question, is possible (Gronlund, 2006). Gronlund (2006) also recommends that, for fill in the blank SA questions, that each question only have one blank and that it appears at the end of the statement as this can reduce possible confusion in reading and interpreting the item. Also, as with all assessment questions, avoid unintended hints to the answers (e.g., longer blanks for longer correct responses, an indefinite article such as “a” suggesting the correct response starts with a consonant). For answers requiring a number, it is important to include the degree of precision required in the response (e.g., respond to two decimal places) and the required units for the responses (e.g., kilometres; Gronlund, 2006).

References

Burton, R. F. (2005). >Multiple‐choice and true/false tests: myths and misapprehension. Assessment & Evaluation in Higher Education, 30(1), 65-72.

Gronlund, N. E. (2006). Assessment of student achievement (8^th ed.). Pearson.

Parkes, J., & Zimmaro, D. (2016). Learning and assessing with multiple-choice questions in college classrooms.Routledge.