Understanding Assessment Formats and Question Types

Assessment is one of the most important components of education as it provides evidence of students' learning outcomes and their growth over time. This process helps not only students but also teachers, families, and other stakeholders in many ways. For example, assessment is a tool to evaluate the effectiveness of teaching and teachers, and it can provide insight into program effectiveness for program developers. According to Brookhart and Nitko (2011), assessments are used to measure, evaluate, and document learning outcomes, proficiency, and educational needs.

Assessment can be formative or summative (Crooks, 2001). Formative assessments are used to monitor day-to-day learning of students, provide feedback to both teachers (i.e., about instructions) and students (i.e., learning process), and make instructional decisions (Tyler, 1967). On the other hand, summative tests are administered after some period of instruction (e.g., after completion of a unit) and are intended to evaluate the effectiveness of overall instruction and students' learning outcomes (Kubiszyn & Borich, 2009). In general, formative assessments aim to give feedback while summative assessments intend to give grades in educational settings.

The questions used in these assessments can be standardized and teacher-made. Standardized tests are administered and graded in a standardized manner. State tests and nationwide tests are examples of standardized tests. Teachers make their quizzes and tests in class that reflect the learning goals and objectives. Additionally, the questions can be objective and subjective. Objective questions usually have only one answer, while subjective questions tend to have several correct answers. The general formats of the questions can be categorized into two groups: (1) close-ended questions, and (2) open-ended questions. Close-ended questions include True/False, Matching, Multiple Choice, and Fill-in-the-Blank, whereas open-ended questions include Short Answers and Essay Questions.

True/False Questions: True/False questions include statements that require test-takers to decide whether the given statement is correct or incorrect. Test-takers can also be asked to write the correct statement if the given statement is incorrect. True/False questions are easy to construct and grade. They are popular because they can cover a wide range of content, and statistical analysis can easily be conducted. True/False questions tend to assess lower levels of skills specified in Bloom's Taxonomy (1957) such as knowledge. However, there are some shortcomings of True/False questions. With every True/False item, regardless of how well or poorly written, the student has a 50% chance of guessing correctly even without reading the item. Secondly, it is not possible to evaluate higher levels of thinking skills using True/False questions.

Test developers should consider some factors when writing True/False questions. First, the statements should be definitely true or definitely false. Second, the statements should be relatively short and kept the same length during the test to avoid misunderstanding because a long and complex statement can cause misunderstanding and reduce the validity of the question. Lastly, the test should include a balanced number of true and false statements.

Example True/False Question:

The product of two positive numbers is positive. True / False

Matching: In matching questions, there is a list of premises and a list of responses that must be paired with each other. Matching is a good way to examine students' understanding of the relationship. In addition to statements, pictures can also be used in matching questions, which might be more useful when developing items for students in lower grades. Statistical analyses can also be conducted for matching questions.

One of the shortcomings of matching questions is the existence of correct guessing. Especially if the number of premises and responses is equal, then the test-taker does not have to match all of the pairs to get a full score because the last premise and response will be automatically paired. Also, matching questions tend to assess lower levels of skills specified in Bloom's Taxonomy (1957) such as knowledge and comprehension.

Test developers should consider some factors when writing matching questions. Firstly, the premises and responses should all be from the same category (i.e., homogeneous). There should be more responses than premises to decrease the probability of correct guessing. Additionally, the premises should be in the left column and responses should be in the right column so that the test-taker will save time when answering the question.

Example Matching Question:

Multiple Choice Questions: A Multiple Choice question involves a statement and more than two responses to choose from (Brookhart & Nitko, 2011). The statement is called a stem, and the incorrect answers are distractors. The stem should be written in direct and non-technical language. The choices should be independent. The distractors should appear correct unless the content knowledge is mastered by the test-takers.

Multiple choice questions can cover a wide variety of learning objectives. Contrary to popular opinion, multiple test questions can measure higher levels of the taxonomy of educational objectives. As there are more than two response options, the chance of correct guessing is lower than True/False questions. Statistical analysis such as item difficulty, item discrimination, reliability can be conducted. In Multiple Choice questions, the probability of correct guessing is lower than True/False questions (i.e., that is the case when a question has more than two response options) but it still exists. Moreover, multiple-choice questions are criticized as students do not create their own answers.

Writing a good multiple-choice question is a difficult process, especially creating plausible distractors. When developing multiple-choice questions, there are some factors to be considered. For instance, the stem should clearly formulate the problem. Second, there should be one and only one correct answer, and distractors should appear correct unless the content knowledge is mastered. Another thing is that the responses should have relatively similar lengths not to give a clue regarding the correct answer. Lastly, the use of negative sentences in the stem should be avoided unless it is necessary.

Example Multiple Choice Question:

Ilker knows the mean of four numbers is 17. If three numbers in this calculation are 14, 19, and 27, what is the last number?

a)8                 b)    10                   c)   12                d) 14

Fill-in the Blank

Fill-in-the-Blank questions are statements with blanks to be filled out. Students are required to fill in numbers, words, or phrases to complete the statements. The primary advantage of fill-in-the-blank assessments is they are easy to construct and grade. Additionally, they do not allow test-takers to guess randomly and get the correct answer. If the item is not written well or if there is more than one answer, then it may confuse the test-taker.

Test developers should consider some factors while creating fill-in-the-blank questions. First, there should be only one blank, and the blank should be located near the end of the sentence. Additionally, if the problem requires a numerical answer, then the unit should be included at the end of the blank. Lastly, test developers should be sure that there is only one correct answer.

Example Fill-in-the-Blank Question:

The product of two positive numbers is _____________.

Short Answer

In Short Answer questions, students are required to answer the questions in a few words or sentences. One of the advantages of short answer items is that they are easy to construct. It also eliminates guessing. On the other hand, short answer questions can be objective or subjective. When there is more than one correct answer, subjectivity is involved in scoring. In that case, a clear rubric should be prepared before the test is administered, and partial credit can also be given. Short answer questions can be used for higher levels of skills specified in Bloom's Taxonomy (1957) such as application, analysis.

Example Short Answer Question:

Prove the identity: sin²(x) + cos²(x) = 1. Please show your work.

Essay

Essay items are open-ended questions where test-takers demonstrate their understanding of the content knowledge with longer written responses (Brookhart & Nitko, 2001). Questions should be written clearly with instructions. The questions should avoid rote memorization of knowledge. Students have to use their own words to write a longer answer to the questions and be able to demonstrate their process of thinking. By using the essay question higher levels of thinking skills as specified in Bloom's (1956) taxonomy such as application, analysis, synthesis, and evaluation.

Essay type questions are easy to construct. Additionally, it eliminates the guessing as the test-takers develop their own answers. On the other hand, there are some disadvantages of essay type questions. For example, they are difficult to score because subjectivity is an important issue in essay type questions. So, the reliability of the grading is always questioned in essay type questions. When creating an essay item, the test developer should prepare a detailed rubric in advance. Additionally, if possible, inclusion of a second grader would increase the reliability of the grades.

Example Essay Question:

As a future math teacher, please identify and discuss the various question formats commonly found in most standardized and teacher-made tests/assessments. Discuss several recommended procedures for writing effective questions for each format discussed and discuss the advantages and disadvantages of each format.