TEST DESIGN AND

IMPROVEMENT
STRATEGIES

Prepared by: Rosita L. Lacea, Ed.D.

 Test Types
When categorizing test types, it's essential to understand that different types of tests serve various purposes in
assessing students' knowledge, skills, and abilities. A well-designed test should align with the intended learning
outcomes and the level of cognitive skills required.
Here are the main categories of test types:
 Objective Tests: These are tests with a clear, correct answer, such as multiple-choice, true/false, or matching
questions. They are useful for assessing basic recall and comprehension (lower-level thinking skills).
 Subjective Tests: These types of tests require students to produce their own responses. Examples include
essays, short answer questions, and problem-solving tasks. They assess higher-order cognitive skills like
application, analysis, synthesis, and evaluation.
 Performance-based Assessments: These involve students demonstrating a skill or applying knowledge in a
practical context. Examples include projects, portfolios, and presentations. They assess real-world
application and integration of knowledge.
Discussion Points:
 How can we decide which type of test to use based on the subject matter or learning objective?
 What are the benefits and drawbacks of each test type in terms of validity and reliability?
 Relationship of Test Types to Levels of Learning Outcomes
The relationship between test types and levels of learning outcomes is grounded in Bloom’s Taxonomy, which
classifies cognitive skills from simple recall to more complex skills like analysis and evaluation.
 Knowledge & Comprehension (Lower-Level Thinking):
o Test types: Multiple-choice, True/False, Fill-in-the-blank.
o These tests focus on recalling facts and basic understanding of concepts.
 Application & Analysis (Mid-Level Thinking):
o Test types: Short answer, Problem-solving, Case studies.
o These tests require students to apply knowledge to new situations and break down information.
 Synthesis & Evaluation (Higher-Level Thinking):
o Test types: Essays, Research papers, Projects, Presentations.
o These tests assess students’ ability to integrate information and make judgments based on critical thinking.
Discussion Points:
 How do we ensure that the test types we use assess the correct level of learning?
 How can a single assessment combine different levels of Bloom’s Taxonomy?
 Various Item Types
Constructing different item types requires careful thought to ensure the test measures what it intends to. Here are
three main item types:
 Selected Response Items: These include multiple-choice and true/false questions. They are efficient for
assessing factual recall and some basic application.
o Best Practice: Keep options relevant, and avoid trivial distractors. Limit the number of options (3-5) for
clarity.
 Supply Response Items: These require students to generate an answer. Examples include short-answer and
essay questions.
o Best Practice: Be specific with prompts and provide clear guidelines for grading to maintain consistency.
 Performance-based Items: These assess practical application of knowledge, often through tasks or projects.
o Best Practice: Provide clear rubrics for grading to assess skills objectively.
Discussion Points:
 How can we balance different item types within a test to ensure comprehensive assessment of all learning
objectives?
 What are the challenges of grading subjective items like essays or performance tasks?
 Prepare a Table of Specifications

A Table of Specifications (TOS) is a tool used to ensure that a test is valid, reliable, and aligned with
learning objectives. It maps out the test’s content coverage and cognitive levels, allowing for a balanced
assessment.
Components of a TOS:
 Content Areas: List the key concepts or skills to be assessed (e.g., specific topics or units in the
syllabus).
 Cognitive Levels: Identify which cognitive levels from Bloom’s Taxonomy are targeted (e.g.,
knowledge, application, synthesis).
 Number of Items: Allocate a certain number of items to each content area and cognitive level, ensuring
proportional representation.
 Example:
Discussion Points:
•How can we use a TOS to prevent bias or overemphasis on certain topics?
•What challenges might we face when trying to balance cognitive levels and
content areas in a test?

Content Area Cognitive Level Number of Items

Photosynthesis Knowledge 5

Cell Respiration Application 4

Genetic Variation Analysis 3

Ecology Synthesis & Evaluation 3

 Incorporate Gender and Inclusivity Considerations into Test Planning
Ensuring gender fairness and inclusivity in assessments is essential to avoid biases
that could disadvantage certain groups. Here are key strategies for achieving this:
 Language: Avoid gendered language (e.g., "fireman" vs. "firefighter"), and ensure
that examples or case studies are diverse in terms of gender, ethnicity, and culture.
 Representation: Make sure that examples, images, and contexts in test items
reflect a range of backgrounds, abilities, and experiences.
 Accessibility: Provide accommodations for students with disabilities, such as
extended time or alternative formats for visually impaired students.
Discussion Points:
 How can we ensure that test content reflects the diversity of students'
backgrounds?
 What challenges arise in making tests universally accessible and inclusive?
 Incorporate Gender and Inclusivity Considerations into Test Planning
Ensuring gender fairness and inclusivity in assessments is essential to avoid biases that could
disadvantage certain groups. Here are key strategies for achieving this:
 Language: Avoid gendered language (e.g., "fireman" vs. "firefighter"), and ensure that
examples or case studies are diverse in terms of gender, ethnicity, and culture.
 Representation: Make sure that examples, images, and contexts in test items reflect a range
of backgrounds, abilities, and experiences.
 Accessibility: Provide accommodations for students with disabilities, such as extended time
or alternative formats for visually impaired students.
Discussion Points:
 How can we ensure that test content reflects the diversity of students' backgrounds?
 What challenges arise in making tests universally accessible and inclusive?
 Improvement of Tests
6. Explain How to Improve a Test by Using the Judgmental Approach
The judgmental approach involves gathering expert opinions to assess and refine test items. This can be
done through processes like:
 Peer Review: Involving colleagues or content experts to review test items for clarity, accuracy, and
alignment with learning objectives.
 Focus Groups: Using small groups of students to review the test and provide feedback on its clarity and
difficulty.
 Item Analysis: Reviewing how students perform on each test item to identify problematic questions,
such as those that are too easy or difficult.
Discussion Points:
 How can peer review help identify bias or ambiguities in test items?
 How can student feedback be used to improve future assessments?
 Conduct Empirically-Based Procedures in Improving a Test
Empirical methods involve using data to make decisions about test improvement. Item analysis and
statistical methods are common tools for improving a test.
 Classical Test Theory (CTT): Involves analyzing the difficulty level (how many students get the item
correct) and the discrimination index (how well the item differentiates between high and low
performers).
 Item Response Theory (IRT): A more advanced approach that provides a statistical model for
analyzing item difficulty and how well items assess varying levels of ability.
 Reliability and Validity Testing: Analyzing the consistency and accuracy of a test through measures
like Cronbach’s alpha for reliability and content or construct validity for ensuring the test measures
what it intends to.
Discussion Points:
 How can item analysis help identify test items that may be too easy or difficult?
 What are the benefits of using statistical methods like IRT in test improvement?
CONCLUSION
Designing and improving assessments is a dynamic process that involves careful planning,
thoughtful construction of test items, and continuous improvement based on feedback and empirical
data. By incorporating inclusive practices, utilizing a well-prepared Table of Specifications, and
using both judgmental and empirical approaches, educators can create assessments that are fair,
comprehensive, and aligned with learning outcomes.

Questions for Discussion:

 What are some of the most common challenges you face when designing assessments?

 How can technology assist in improving the fairness and accessibility of tests?