KIRINYAGA UNIVERSITY

SCHOOL OF BUSINESS AND EDUCATION

EDUCATIONAL MEASUREMENT AND EVALUATION

TAKE AWAY CAT

1. Differentiate between reliability and test validity by highlighting the five categories in
each case.

The main factors to consider while evaluating a test, an object or an item are reliability
and validity. These are two important concepts in the field of research methodology and
assessment.
Reliability refers to the state wherein consistent scores are obtained over repeated
testing. In simple terms it describes the stability and consistency of a test. In other words
it refers to the extent to which the results of a study test or measurement can be replicated
or reproduced under the same conditions.
There are five categories of reliabilities:
a) Test-retest reliability
This refers to the consistency of a test or a measurement when it is administered
at different points in time. It involves testing the same subjects at a later date,
ensuring there is a correlation between the results. It refers to the test’s
consistency when administered under different occasions. Test-retest reliability is
the degree to which scores are obtained over time.
b) Inter-rater reliability
This refers to the consistency of a test or measurement when it is administered by
different raters or evaluators. Here when a single concept is measured using
multiple items, the inter-rater reliability is employed.
c) Parallel/Equivalent reliability
It assesses the extent to which different forms of a test yield a similar results,
indicating the reliability of the test when different versions are used. To determine
parallel forms reliability, a reliability coefficient is calculated on the scores of the
two measures taken by the same sample.
d) Inter-item reliability
It assesses the extent to which different items on the same test measure the same
construct or attribute. Here the values of internal consistency varies from zero to
one. The most common index of reliability is Cronbach’s coefficient alpha which
will be equal to zero.
e) Split-half reliability
It assesses the extent to which different halves of the same test yield similar
scores. This method involves diving the test into two equal halves and then
comparing the scores obtained on each half. The purpose is to evaluate the
 consistency of the test items and determine whether the test is reliable when split
in this way.

Validity on the other hand refers to the extent to which a test or an assessment
tool measures what is supposed to measure. Validity it indicates the degree of
accuracy of the measurement. Here are some categories of test validity:
a) Content validity
Is a measure of all aspects of a construct. This refers to the extent to which a
test or assessment tool cover the relevant content or domain of the construct it
is intended to measure.
b) Criterion-related validity
It points to the presence of a specific trait or behavior which may be of use in
the future. It examines the relationship between a score on a test and a
learning outcome. Criterion-related validity refers to the extent to which a test
or assessment tool is able to predict or correlate with a relevant outcome or
criterion.
c) Construct validity
Is the extent to which a test measures what is supposed to measure. It refers to
the degree to which a test measures what it claims or purports to be
measuring.
d) Concurrent validity
This refers to the extent to which a test or an assessment tool is able to
correlate with an established measure of the same construct. In other hands it
indicates the extent to which the scores on a test correlate with another similar
established test
e) Curricular validity
Is the degree to which the content of the test and the objectives of the
curriculum are in tune with each other. Curricular validity can be the key
determinant on whether a student makes the grade or not.

2. Test moderation is a critical process that ensures the integrity and fairness of Here are
some best practices for effective test moderation:
a) Establish clear guidelines: Develop and communicate clear guidelines for
test moderation, including the criteria for evaluating responses, acceptable
sources, and any specific rules or restrictions.
b) Train moderators: Ensure that moderators are well-trained and
knowledgeable about the subject matter, as well as the assessment criteria and
guidelines. Provide ongoing training and support to ensure that moderators are
up-to-date with best practices.
c) Use standardized rubrics: Use standardized rubrics to evaluate responses,
rather than relying on subjective judgments. This helps to ensure consistency
and fairness in the evaluation process.
 d) Monitor for plagiarism: Use plagiarism detection software to check for
instances of plagiarism or other forms of academic dishonesty. This can help
to ensure the integrity of the assessment process.
e) Monitor for cheating: Monitor the assessment process for any signs of
cheating or other forms of misconduct. This may include monitoring for
unusual patterns of responses or using surveillance cameras in high-stakes
assessments.
f) Provide clear feedback: Provide clear and timely feedback to students on
their performance, including specific areas for improvement and any steps
they can take to improve their performance in the future.
g) Continuously review and improve: Continuously review and improve the
test moderation process to ensure that it is effective and fair. This may include
gathering feedback from students and moderators, analyzing assessment data,
and making changes as needed.
3. Positive discrimination and negative discrimination are terms used in the context of
measurement and evaluation to describe different types of bias that can occur in the
assessment process.
I) Positive discrimination, also known as “affirmative action,” refers to the
practice of giving preferential treatment to certain groups or individuals in
order to promote diversity and inclusion. This may include providing
additional support or resources to underrepresented groups, such as
minority students or students with disabilities. It refers to the practice of
favoring individuals from a particular group. Example include Affirmative
action where policies that give preference to individuals from historically
marginalized groups.
II) Negative discrimination, on the other hand, refers to the practice of
treating individuals from a particular group unfairly or disadvantageously.
Examples include stereotyping that is making assumptions about a
student’s abilities based on their race, gender or other demographic
factors. Biased test items where posing questions to students that may be
culturally insensitive or disadvantage students from certain backgrounds.

ASSIGNMENT
Test construction is an important aspect of a teacher’s role, as it involves creating
assessments that are effective and accurate measures of student learning. Here are
five identifiable test characteristics that are important to consider when
constructing assessments:
1. Alignment with learning objectives: The test should be aligned with the
learning objectives and standards that the teacher is trying to assess. This
ensures that the test is measuring the skills and knowledge that the teacher has
identified as important for student learning.
2. Validity: The test should be valid, meaning that it accurately measures what it
is supposed to measure. This may include using a variety of assessment methods,
such as multiple-choice questions, short answer responses. To establish validity,
teachers should use a variety of methods, such as aligning test items with learning
objectives, consulting with experts, and conducting pilot studies.
2. Reliability:
The test should be reliable, meaning that it produces consistent and stable results
over time. This may include using pilot testing to identify and correct any issues
with the test as well as using statistical analysis to ensure that the test is producing
consistent results. Reliability is essential for ensuring that test scores are not
affected by chance or other extraneous factors. Teachers can enhance reliability
by using clear and unambiguous test items, providing adequate testing conditions,
and administering multiple forms of the test.
3. Fairness:
A test is fair if it does not discriminate against any group of students based on
their race, gender, ethnicity, or socioeconomic status. The test should be fair,
meaning that it provides equal opportunities for all students to demonstrate their
skills and knowledge. Fairness is a critical ethical principle in test construction.
Teachers can promote fairness by avoiding cultural bias in test items, ensuring
that all students have equal access to test preparation materials, and providing
accommodations for students with disabilities.
4. Efficiency:
A test is efficient if it can be administered and scored in a timely manner.
Efficiency is important for maximizing instructional time and minimizing the
burden on both teachers and students. Teachers can improve efficiency by using
well-organized test formats, providing clear instructions, and using appropriate
scoring methods.
5. Authenticity:
A test is authentic if it assesses students' ability to apply their knowledge and
skills in real-world situations. Authenticity is increasingly valued in education as
it helps students develop critical thinking and problem-solving skills. Teachers
can create authentic assessments by using performance-based tasks, simulations,
and projects that require students to integrate their learning.
By carefully considering these five characteristics, teachers can construct tests
that are both informative and valuable for assessing student learning.