welcome to teaching online
Teaching Online Contents
- 1 Classification of Criterion-Referenced Tests
- 2 Types of Criterion-Referenced Tests
- 3 Difference between C.R.T.s and N.R.T.s
- 4 Techniques for Determining Criterion Score
- 5 Application of Criterion-Referenced Tests
- 5.1 Recruitment of Personnel’s
- 5.2 ii. Summative Evaluation
- 5.3 III. Formative Evaluation
- 5.4 Iv. Special Education Teacher
- 5.5 iv. Evaluation of School Programmers
- 5.6 vi. As Predictors
- 5.7 Nature and Meaning of Criterion-Referenced Tests Part-1:- Click Here
- 5.8 Nature and Meaning of Criterion-Referenced Tests Part-2:- Click Here
Classification of Criterion-Referenced Tests
Non-paper and Pencil Tests
Performance batteries of various kinds, not involving paper and pencil tests, may be among the oldest type of criterion referenced tests. CRTs are not restricted to paper and pencil.
Paper and Pencil Tests
This type of tests that has attained a major importance in the field of educational testing. Criterion-referenced tests are most often used in the educational field at end of the year, end of semester, end of courses, end of unit, and end of module achievement test.
Types of Criterion-Referenced Tests
Criterion-Referenced Tests may be used to ascertain an individual’s status with respect to a well defined behavior domain. Behavior domain means a segment of knowledge. Dockrel (1980) has classified C.R.T.s into the following three categories:
- Single Act Test: If the intended learning outcome is a discrete single phenomenon and can be thought of as a measurement, it is termed single act test. For example, the pupil will be able to recall the name of capital city of U.P.
- Closed Domain Tests: Intended learning outcome is more complex and encompasses much more than a discrete single phenomena. However, the domain items are sampled in order to keep the scope restricted to selected attributes of the segment of knowledge. For example, the pupil will be able to choose the adjectives from the given test of words comprising only verbs and adjectives.
- Open Domain Tests: Intended learning outcomes is highly complex encompassing a much larger chunk of the domain or segment of knowledge and demands tests which are open in scope, providing a much broader basis of sampling which remains the underlying construct of such test. For example, the pupil will be able to understand the concept of organic evolution.
Difference between C.R.T.s and N.R.T.s
- Norm-referenced tests help to describe pupil performance in terms of the relative position held in some lmown group. Whereas criterion-referenced tests helps to describe pupil performance in term of a specified domain of instructionally relevant tasks.
- In N .R.T.s the test is typically planned on the basis of general descriptions of subject-matter topics and process skills. In C.R.T.s the test is typically planned in terms of specific, behaviorally stated objectives, each providing the basis, for one or several related test items.
- In N .R.T.s the test items are constructed to produce maximal variability in performance across students. In C.R.T.s the test items are constructed to represent the domain of the objective and to measure proficiency for a specified task.
- All types of test items are used in N.R.T.s All types of test items are used in C.R.T.s.
- Test items are of medium difficulty in N.R.T.s Difficulty of the items depends on the nature of the specific learning tasks to be measured in C.R.T.s.
- Deliberate attempts are made to eliminate very easy and very difficult items to obtain a wide spread of scores in N.R.T. but in C.R.T.s no attempt is made to modify item difficulty, or to eliminate easy items from the test, in order to obtain a range of scores.
- No criterion for mastery is customarily specified in N.R.T. but in C.R.T.s objectives for criterion-referenced test items designate a criterion for mastery or it sometimes can be inferred.
- Percentiles or standard scores are employed for reporting results in N.R.T.s but in C.R.T.s apt to employ ‘percent correct’ scores for reporting results.
- A total score or several sub-scores are computed and a pupil’s relative standing in a group or class is ascertained while in C.R.T.s a pupil’s success or failure on a test item or small group of similar test items is determined and a statement is prepared that describes his performance solely with reference to certain performance objectives.
- Content, construct and criterion-related validity are appropriate in N.R.T.s whereas in C.R.T.s Descriptive, Functional and Domain-Selection validity are appropriate.
- Classical estimates of reliability-test-retest, equivalent form, split- half and Kuder-Richerdson Method are used in N.R.T.s There have been various attempts to develop statistical measures for estimating the reliability of criterion referenced tests, but a satisfactory solution has not yet been achieved.
- Standardized tests are available covering most of the curriculum areas for N.R.T.s Attempts are being made to develop criterion-referenced tests and an increasing available, many of which focus on the basic skills.
Techniques for Determining Criterion Score
There are differences in the views of experts for deciding the mastery level for criterion-referenced tests. One will report that he always sets a standard of two-thirds of the items correct for mastery.
Another expert will report that he holds learners to 70% mastery, and a third advances his 90% standard with an air of tough mindedness and respect for excellence. Therefore, different mastery level has been decided by different experts. Following are the methods for determining criterion score:
1 .Judging Minimal Competence
In this technique the experts study the test and then decide what minimally competent person should score on the test. Therefore scores obtained by a competent person is decided as mastery level.
Performance of others as a Criterion
In this technique criterion levels are established by reference to parameters of existing population of examinees. Hence, the criterion or mastery level on a test may be established as the median test score obtained by persons of a certain type.
3. Decision Theoretic Approach
In decision theoretic approach criterion score is fixed to minimize benefit loss-ratio through classification of examinees by criterion-referenced tests. This is a statistical approach.
Counting Backwards from Hundred Percent
In this techniques it is decided how much concession is to be made to the student to have a short fall from hundred percent. This decision depends on experts idiosyncrasy which is not defensible.
Bootstrapping on Other Criterion Score
In this technique those candidates are identified who are considered to be competent persons in the field. Their scores on the test is considered to be the criterion score for others who take the test in later years. Sometimes a similar approach is adopted by considering the score of competent persons on a similar CRTs.
Operations Research Method
This technique is based on general approach of operations research of maximizing a value commodity by finding an optimum point on a curve or a graph.Guilbert ( 1976) suggest that the criterion score on the test should be equal to the total acceptability indices of all the multiple-choice questions used in the test.
Application of Criterion-Referenced Tests
Recruitment of Personnel’s
Criterion-referenced tests are widely used in the west for the recruitment of personnels.
ii. Summative Evaluation
It’s used for overall evaluation, with respect both to individual students and to the group as a whole, when used for groups (classes, schools etc.) there is generally some normative connotation and when used for individuals they operate in the pass-fail context to determine whether the standard has been met and whether the student may therefore proceed to the next stage.
III. Formative Evaluation
The underlying idea for using CRTs is to discover the adequacy in their learning and help the weaker section of students to reach the level of other students through a regular programmed of remediation . It is assumed that given more time to such students it is quite possible to bring them up to the desired level of intended performance.
Iv. Special Education Teacher
Criterion-referenced tests may be used by special education teachers in diagnosing student learning deficiencies and monitoring the progress of their students.
iv. Evaluation of School Programmers
They are also being used to evaluate various school programmes. These tests are finding some use in higher educational programmes based upon the mastery learning.
vi. As Predictors
Criterion-referenced tests can be set that will be predictive of success in whatever future activity one is trying to predict success, but even in that case it is the criterion of success in the predicted activity that is important not an arbitrary standard of success on the criterion-referenced test itself.