1.
Validity
A
test is considered as valid when it measures what it intends to measure
·
There
are different types of validiy
1. Face validity
Looks as if it measures what it is supposed to
measure.
2. Operational validity
A test will have
operational validity if the tasks required by the test are sufficient to
evaluate the definite activities or qualities
It has discriminating
power, that is: its
power to discriminate between the upper and lower groups who took the test. The test should contain
different difficulty level of questions
3. Content validity
If the items in the
test constitute a representative sample of the total course content to be
tested, the test can be said to have content validity
4. Construct validity
The test from should fit the aspect to test.
5. Concurrent validity
Looks as if it measures what it is supposed to
measure.
6. Predictive validity
A test has
predictive validity if the scores on it predict future performance
2.
Reliability
§
A
test is considered reliable if it is taken again by the same students under the
same circumtances and the score average is almost the constant/consistent,
taking into consideration that the time between the test and the retest is of a
reasonable length.
§
Test
scores said is said to have stable or
consistent scores when the test scores demonstrate consistency or stability no
matter who administer the test.
§
The
scores are also consistent no matter when or where the test is administered.
§
Valid
test is also a reliable test, but reliable test may not be valid one.
§ Objectivity of a test can be increased by using more objective type of
test items and the answers are scored according to model answers provided. Objectivity
Different method for determining reliability
- Test-retest method
·
This
method utilized the same test twice to
the same test twice to the same group of subjects on different testing
occasions.
·
Repitition
on the use of the same instrument and the involvement of the same subjects, the
different is in the time for test administration.
·
Advantage
·
Only
has one set of test to be constructed
·
Disadvantages
·
It
is not known exactly what the best time
interval for conducting the second test administration is so as to assure a
tolerable change on the test takers’ attributes.
Too short time→’history’ (the test taker may
still rememberr the contents of the test)
Far too long time→may make test takers’ attribute
change that will affect the test takers’ performance on the second
administration.
·
Involving
the same subject in the test administration twice also is likely to endanger
both the mental and physical states of the subjects such as boredom and
ailment.
- Parallel form method
1. Used two different versions of
the same test
2. Administered to the same group
close together in time
3. It used form A or B and can be
given on the same day or the next day
4. The difference of the two is
how they worded or written
5. It should measure the same
skills and errors are significantly controlled
6. If higher the correlation, the
more reliablity.
·
Disadvantages
·
Making
tests that are equally similar in all aspects is not an easy task
·
It
is not easy to keep the test takers’ mental condition the same when they
responded to two sets of tests administered almost at the same time.
- Internal consistency
·
Split-half Procedure
·
Equally difficult questions are given in two or more
question forms. They are set in two halves. But actually they are the same.
·
Inter - item Estimation
·
Equally difficult parts of questions are given in a
question.
3. Practicality
Practicality
of the test depends on:
•
Administrative
ease. Like the example of bringing elephant to the class.
It is not practical at all.
•
Scoring
ease. Using scoring rubric if it is subjective test or
scoring guide if it is one-correct-answer (objective) test.
•
Interpretative
ease. Test
should be written in a clear, correct and simple as possible while still
testing skill that is intended to test.
4. Economy
Time, money and energy efficiency in making the test
item, administering the test, scoring procedure of the test
0 comments:
Post a Comment