Choosing the Best Method for Testing Knowledge



As a teacher, it’s essential to effectively assess your students’ knowledge. What do they know? Where do they need to improve?

In Designing Effective Instruction, Gary Morrison, Steven Ross, and Jerrold Kemp discuss the various “tests” available to evaluate different types of knowledge, each with its own strengths and weaknesses. The key is to match each test with your instructional objectives along with any practical limitations, like time or manpower.

Here’s a quick look at how to use each test: (a) Multiple Choice, (b) True/False, (c) Matching, (d) Short-Answer, and (e) Essay.

Multiple Choice

When to use it: Multiple choice can be a very versatile testing method, used to evaluate all but the highest levels of Bloom’s taxonomy. Multiple choice questions can be used to assess knowledge, comprehension, application, and analysis. The practical advantage of multiple choice questions is that they are easy and quick to grade.

Disadvantages: It’s very difficult to assess synthesis and evaluation using multiple choice questions. Additionally, these questions test recognition (choosing an answer) rather than recall (constructing an answer), so it’s possible for students to guess correctly without actually knowing the information.

Tips:

  1. Don’t test trivial facts, or try to trick students with cleverly written questions. Focus on testing basic, fundamental knowledge.
  2. It’s best to use 3-5 options. Try not to go over 5, and avoid making one option an absurd throwaway.
  3. Avoid overusing “all of the above” or “none of the above”. This often allows students to eliminate options without really knowing the information.

True/False

When to use it: True/false questions are fairly easy to write and simple to grade. They can be used to test knowledge or comprehension.

Disadvantages: Like multiple choice questions, true/false questions questions can only test recognition rather than recall. Students have a 50% chance of guessing the correct answer, and it’s difficult to test higher level learning objectives.

Tips:

  1. Avoid ambiguous true/false statements.  The statement should be entirely true or entirely false.
  2. Don’t mix multiple ideas into one true/false statement. Students may be able to identify the statement as false while only understanding a portion of the concepts involved.
  3. Use true/false questions sparingly. It’s a low-level testing method prone to student guessing.

Matching

When to use it: Matching is a great alternative to multiple choice that allows you to fit a lot of material into a small space. It’s used most effectively for testing knowledge of definitions, events, dates, people, principles/descriptions, and function/parts. It’s more difficult to guess on matching questions than it is on multiple choice or true/false.

Disadvantages: Matching assesses recognition rather than recall, and can only be used for low levels of learning (knowledge).

Tips:

  1. It’s difficult for students to match more than 7-8 items together.
  2. Keep the description of each item short and to the point. Don’t expect students to match paragraphs of text together…it’s just too much information to process.
  3. Students hate this, but it’s a good idea to add one or two extra items in the match list. Or allow items to be used more than once.  This can limit the chances of guessing and encourage more thoughtful answers.

Short-Answer

When to use it: Short answer tests are the most “objective” of the constructed-response tests, because scoring can be fairly objective. They can also be used to test a large amount of content fairly quickly, and test recall rather than recognition. As an added bonus, it’s difficult for students to guess on short answer questions.

Disadvantages: Short-answer questions can only be used to assess fairly low level learning like knowledge, comprehension, or application. Also, certain types of short-answer questions may be difficult to score quickly or objectively.

Tips:

  1. Short answer questions can range from a single word, to a few words, to a short paragraph. The longer the expected response, the harder it is to maintain objective grading.
  2. Be sure to word your questions so that only one answer is correct. This can help the scoring remain more objective and avoid student complaints.
  3. Programs like SAGrader can be used to ensure objective and fast scoring.

Essays

When to use it: Essay questions are your best option to evaluate higher levels of learning. It’s a great method to use when you’re interested to see how your students can analyze, synthesize or evaluate information.  Essay questions can range from a few sentences to multiple pages. Essays require students to express themselves in writing, which is a valuable life skill, and provide insight into your students’ understanding of the material.

Disadvantages: Though it is a valuable evaluation method, students often don’t like writing, and it can require more time for them to complete essay questions. As a grader, there is always a danger of being swayed by “good writing” and giving higher grades to well-written papers that don’t cover the course material adequately. Essays take longer to score, and scoring can often be unreliable or biased, especially if multiple people (like TAs) are sharing the grading responsibilities.

Tips:

  1. Provide students with your grading criteria ahead of time. How will they be graded? Does spelling count? Do you require a certain format?
  2. Grade essays without knowing the writers’ identities. This can help reduce grading bias, and decrease student complaints.
  3. Keep your writing prompts as focused as possible to avoid off-topic responses.
  4. Programs like SAGrader can be used to ensure objective and fast scoring.  By providing feedback and opportunities for revision, SAGrader can also be used to promote learning through writing, rather than just assessment.

As the authors remind us, whatever test you use needs to be a valid measure of performance. Before choosing your testing method, make sure you understand your learning objectives and rank the level of learning (on Bloom’s taxonomy) you want to assess. Then, select your testing method and write clear questions linked to each learning objective.

 Photo via