material to a present second language context. They also mention that intralingual transfer can be attributed to the ignorance of rule restriction which according to
Richards and Sampson means “applying rules to contexts to which they do not apply” 1974.
According to Ellis 1994: 59, overgeneralization errors appear when the learner creates a deviant structure on the basis of other structures in the target
languages. This will involve the creation of one deviant structure in place of two target language structures. As an example, a learner says “He will moves” where
English allows “He will move” and “He moves”.
3. Validity and Reliability of the Test
Since this research is a quantitative research, it always depends on the measurement. There will be two essential criteria of the quality of any measuring
instrument: validity and reliability. The following is the further description of the validity and reliability of a test.
a. Validity of the Test
In addition to be reliable, good measuring instruments should be valid Sprinthall et al, 1991: 37. Meaning to say that a test should actually measure what
it purports to measure. They further explain that it is important to remember that instruments are valid only for a particular purpose. A test that has been validated
for one purpose cannot be assumed to be necessary valid for any other use unless it has also been proven to be valid for that purpose. Donal Ary, Lucy Cheser Jacobs,
and Asghar Razavieh argue that “the focus of recent views of validity is not on the PLAGIAT MERUPAKAN TINDAKAN TIDAK TERPUJI
instrument itself but on the interpretation and meaning of the scores derived from the instrument” 2002: 242. This research measures the students’ ability in
mastering the use of English preposition for and to. Therefore, in order to be valid, the test that will be conducted should be concerning the English prepositions for
and to. According to Neuman 2011: 212, there are four types of measurement
validity, namely face validity, content validity, criterion validity, and construct validity. Here are the explanations of each types.
i. Face Validity
Face validity is the most basic and easiest type of validity to achieve. Face validity is a type of measurement validity in which an indicator “makes sense” as a
measure of test in the judgment of others, especially in the specific community. ii.
Content Validity Content validity addresses this question: is the full content of a definition
represented in a measure? In this type of validity, the measurements should represent all ideas or areas in the conceptual space. On other words, content validity
requires a test that represents all aspects of the conceptual definition of a construct. iii. Criterion Validity
This type of measurement ability applies several standards or criteria in order to point out the test accurately. The validity of a test is verified by comparing
it to another measure of the same construct in which a researcher has confidence. This type has two subtypes, namely concurrent validity and predictive validity.
Concurrent validity is measurement validity that relies on a preexisting and already PLAGIAT MERUPAKAN TINDAKAN TIDAK TERPUJI
accepted measure to verify the indicator of a construct. Predictive validity is measurement validity that relies on the occurrence of a future event or behavior that
is logically consistent to verify the indicator of a construct. iv. Construct Validity
Construct validity is for measures with multiple indicators. It address this question: if the measure is valid, do the various indicators operate in a consistent
manner? This measurement validity requires a definition with clearly specified conceptual boundaries. This validity covers two subtypes of validity, namely
convergent validity and discriminant validity. Convergent validity is a type of measurement validity for multiple indicators based on the idea that indicators of
one construct will act alike or converge. Discriminant validity is a type of measurement validity for multiple indicators based on the idea that indicators of
different construct diverge.
b. Reliability of the Test
According to Richard C. Sprinthall, Gregory T. Schmutte, and Lee Sirois, reliability refers to the consistency with which a test or instrument produces result
1991: 34. They further highlight that the basic question addressed by reliability is whether a test given at one time will give the same result if given at another time
under the same conditions. Neuman 2011: 208 also argues that measurement reliability means that the numerical results an indicator produces do not vary
because of characteristics of the measurement process or measurement instrument PLAGIAT MERUPAKAN TINDAKAN TIDAK TERPUJI