2. Categories of Good Test in Summative Test
Actually, the summative test has to represent all the materials which have been taught by the teacher. So, the teacher should have some criteria of good
summative test. Those are validity, reliability, and practicality.
a. Validity
According to Wilmar Tinambunan, “Validity refers to the extent to which the
results of an evaluation procedure serve the particular uses for which they are intended. Thus, the validity of the test is the extent to which the test measure what is
intended to measure.”
7
It can be concluded that the test will be useful if the test is able to measure what is intended to measure to know the quality of the test. It means that the test
items in the summative test, the test maker must give the materials which are covered as a category of good test.
b. Reliability
The second criteria of good test are reliability. It is measured by a correlation between the scores of the same set of students on two consecutive administrations of
the test.
8
It can be supporte d by Herbert and William that, “reliability refers to the
degree to which the measurements yielded by a test are consistent or stable.”
9
It means that, if the test has been tested more than once in the same student in the different time but the score does not really different or change drastically with the
score before it means the test can be called reliable. Another opinion, result of the score of the test not only stable but also dependable means show of the readiness of a
test, and predictable that is the test is able to predict the next the result. Heaton states that there are five factors affecting reliability of the test. The
first is the extent of the sample of material selected for testing, the second is the
7
Wilmar Tinambunan, op.cit., p. 11.
8
Robert Lado, Language Testing, New York: McGraw-Hill Book Company, 1964, p.31.
9
Herbert J. Klausmeier and William Goodwin, Learning and Human Abilities, 2
nd
Ed., New York: Harper Row, Publisher, 1961, p. 585.
administration of the test, the third is the instruction, the fourth is personal factors such as motivation and illness, the last is about the scoring the test.
10
c. Practicality
Practicality, the third criteria of a good test can be called as usability. In this case the teacher or the test maker in making a test should be consideration some
practical such as economy, scoring, and interpretation. As Douglas Brown said in his book, “a good test is practical. It is within the means of financial limitations, time
constraints, ease of administratio n, and scoring and interpretation.”
11
It means that there are some factors which are considered before make a test.
According to Wilmar said, “before administering a test, some factors about the administration and the test itself must be carefully conside
red.”
12
It can be concluded that before the test is used, the test maker has think of the some
consequences of the usability the test itself, such as scoring procedure in order to easier the teacher to calculate the result of the test, in giving the test item in a test, and
so on.
3. Types of Test Item in Summative Test
Besides the categories of the summative test, in that test there are some kinds of the types of the test items. Those are:
a. Objective Test
This test can be called objective if the answers of the test have only one the correct answer as a key of the test item. In this test the students have choose one
correct answer from some choices which provided by the teacher. In giving the scoring of objective items requires much less time than the scoring of essay items.
10
J.B.Heaton, Writing English Language Tests, New York: Longman, 1988, p. 162-163.
11
H. Douglas Brown, Teaching by Principles, New York: Addison Wesley Longman, Inc., 2001, p. 386.
12
Wilmar Tinambunan, op. cit., p. 23.
With its objectivity, objective items can be accurately scored with little if any dispute about the correctness of response.
According to Zainal Arifin’s book, “There are many varieties of there new
test, but four kinds are in most common use, true-false, multiple-choice, completion, matching.”
13
1 True-False
Based on Jum C. Nunnally, “The popularity of the true-false item is probably due to the ease with which such items can be composed. It is usually easy to make up
many such items in a relatively short period of time.”
14
Besides that, James Dean Brown and Thom Hudson, in their book “requires student to respond to the language
by selecting one of two choices, for instance, between and true and false or between correct and incorrect.”
15
To sum up, in true-false the students are able to answer the statement with true or false by short of time. Then the function of this test is to measure the
competence of the student to differentiate between the fact with the opinion. In addition, the teacher not only provides question or statement in this test but also
possible to give the picture, diagram, or table.
2 Multiple-Choice
The multiple-choice is the most popular in types of test because of the multiple-choice often is used in many kinds of objective test. To support the
statement above, William states in his book, that “by far the most popular type of
13
Zainal Arifin, op. cit., p. 135.
14
Jum C. Nunnally, Educational Measurement and Evaluation, New York: McGraw-Hill, Inc., 1964, p. 160.
15
James Dean Brown and Thom Hudson, Criterion-Referenced Language Testing, Cambridge: Cambridge University Press, 2002, p. 66.
Kategori