Proficiency Test Summative Test

a. Content Validity A test can be said to have content validity if it is built with a representative sample of the language skills, structures, etc. which it is meant to be concerned. 32 In line with that, Anthony J. Nitko writes that, “content validity is the extent the items on a test are representative of the domain or universe that they are supposed to represent.” 33 Thus, the degree of content validity in a test relates to how well the the test measures the content of subject matter that students studied before. Therefore, it is important to make sure that the test covers all the areas of material that are supposed to be assessed. For example, a grammar test should be made up of items relating to the knowledge of grammar. b. Construct Validity This type of validity relates to any underlying ability that is formulated in a theory of language ability. Construct validity is “the extent that a test measures the trait, attribute, or mental process it should measure, and whether descriptions of persons in terms of such constructs can follow using the scores from that test.” 34 Moreover, Arthur Hughe s writes that, “it is a matter of empirical research to establish whether or not such a distinct ability exsists, can be measured, and is indeed measured in that test.” 35 In other words, it can be said that a test has construct validity if it is able to measure certain specific characteristics agreeable with a theory of language and behavior in learning. c. Criterion-Related Validity Criterion-related validity relates to the extent how agreeable the results of the test with the results come from the another independent 32 Arthur Hughes, Testing for Language ..., p. 26. 33 Anthony J. Nitko, Educational Test ..., p. 413 34 Anthony J. Nitko, Educational Test ..., p. 413. 35 Arthur Hughes, Testing for Language ..., p. 31. and trustworthy assessment of student’s competence. 36 In addition, in his book, Educational Tests and Measurement, An Introduction, Anthony J. Nitko states that, “criterion-related validity questions concern the extent to which scores on a test permit inferences about examinees’ likely standing on another measure called a criterion.” 37 This type of validity can be divided into two parts; namely, concurrent validity and predictive validity. 1. Concurrent Validity According to J. Stanley Ahmann and Marvin D. Glock, this validity is “designed to estimate present status with respect to a c haracteristic different from the test.” 38 In other words, it tries to determine a student’s present standing indirectly. Concurrent validation is carried out by comparing an individual ’s test scores with his other assessment taken at about the same time. 2. Predictive Validity Predictive validity is intended to predict how well someone will perform in the future. It is supported by a quote, “predictive validity concerns the degree to which a test can predict candidates’ future performance.” 39 To do this validition, the earlier test scores from individual students are correlated with grades made at the end of the first semester.

2. Reliability

Consistent measurement is a necessary condition for high quality educational testing. This consistency of a test is called as reliability. 36 Arthur Hughes, Testing for Language ..., p. 27. 37 Anthony J. Nitko, Educational Test ..., p. 422. 38 J. Stanley Ahmann and Marvin D. Glock, Educating Pupil ..., p. 288. 39 Arthur Hughes, Testing for Language Teacher..., p. 29.