You are on page 1of 4

Republic of the Philippines

Tarlac State University


COLLEGE OF TEACHER EDUCATION
CENTER OF DEVELOPMENT
Lucinda Campus, Tarlac City
re

6.2: VALIDATION
A WRITTEN REPORT IN EDUC 202

Prepared by:
GAMIS, NESTILYN E.
BACHELOR OF ELEMENTARY EDUCATION 2-2

Submitted to:
Mr. Paulo Xavier Samson
Course Instructor
Objectives:
At the end of this Report, you should be able to:

• Explain the meaning of Validation


• Explain the three main types of evidence

Introduction
The instrument must be validated when the item analysis is completed and the items that require
revision have been revised. Determining the attributes of the entire test, specifically its validity and
reliability, is the goal of validation. The process of gathering and evaluating data to bolster the test's
utility and meaninglessness is known as validation.

Validity
Validity is a test can be defined as its ability to measure what it is supposed to measure or as the
appropriateness, correctness, significance, and utility of the judgments and educator makes in
response to the test results. The distinction between these two definitions of validity is that the
first one to speaks of the teacher’s decision that are based on the test results.

Tree Essential Main Types of Evidence that may be Collected:


Content-related evidence of validity refers to the instruments and content. How suitable is the
material? How thorough? Does it make sense to arrive at the desired variable? To what extent
does the sample of questions or items accurately reflect the topic that needs to be evaluated?

Criterion-related evidence of validity describes the connection between the results of the
instrument and the results of one or more additional test (commonly referred to as a criterion)
How solid of a bond is this? To what extent do these scores accurately reflect current or forecast
performance of a particular kind?

Construct-related evidence of validity relates to the type of psychological concept or trait that
the test is measuring. To what extent can a measure of the construct account for individual
variances in behavior or task performances?

1
The usual procedure for determining content validity may be described as
follows:
The test objectives are drafted by the teacher using the specifications table. The test and
these are then given to a minimum of two experts, along with a summary of the exam
participants. The specialists review the test's objectives and items, marking any that they believe
do not measure the objectives with a checkmark. In front of every goal that isn't evaluated by a
test item, they furthermore place a checkmark. Writing new items to cover objectives not covered
by the current test, or having an item verified and submitted to experts, is up to the teacher. This
continues until the specialists give their approval to every item and until all components are
approved by specialists, as well as until they concur that the test adequately covers each of the
purposes.

The teacher typically compares test results with those of another independent criterion test
that is assumed to have high validity in order to find indications of criterion-related validity. We
say we have strong criterion-related evidence of validity, for instance, if a test intended to assess
pupils' mathematical proficiency has a strong correlation with an external criterion—a
standardized mathematics achievement test. Its concurrent validity refers specifically to this kind
of criterion-related validity. Predictive validity is a different kind of criterion-related validity in
which there is a correlation between the test scores on the instrument with the students' results on
a subsequent performance (criterion measure). For instance, there might be a correlation between
the students' results on the teacher-created mathematics ability test and their subsequent
performance in a Division-wide math exam accomplishment evaluation.

In addition to the correlation coefficient, Gronlund proposed the use of an expectancy table
to assess the validity of criteria related to the criteria. The test (predictor) categories are listed on
the left side of the table, while the criterion categories are listed horizontally at the top of the
chart. This table is simple to put together. Assume, for illustration purposes, that an
accomplishment test in mathematics is created, and that the results are classified as high,
average, and low. The final average high school grades of the students—Very Good, Good, and
Needs Improvement—are employed as the criterion measure.

2
The two-way table lists down the number of students falling under each of the
possible pairs of (test, grade)

Test Score Very Good Good Needs


Improvement
High 20 10 5
Average 10 25 5
Low 1 10 14

Grade Point Average

According to the expectation table, twenty students received high test scores and were later
rated as excellent; twenty-five students received average scores and were subsequently rated as
good; and fourteen students received low test scores and were later rated as needing
improvement. For this specific test, the evidence seems to suggest that students who score well
would be considered great, those who score averagely would be rated good subsequently, and
those who score poorly would be regarded as needing work later.

You might also like