Language Proficiency Test Review - Essay Example

Language Proficiency Tests Legally speaking, the purpose of language proficiency tests is to provide a means of allowing schools to assess the levels of English proficiency held by students entering school from homes where English is not the spoken language. Standards for what constitutes fair, accurate and professional testing varies from state to state, but is regulated and kept generally uniform by federal mandates. All of the tests are designed to test the four general modalities of using the English language, which are identified as reading, writing, speaking and listening. This proficiency is measured in terms of the student’s grasp of several areas of the English language, but testing regarding the student’s ability to recognize the interaction of the parts differs depending upon the test used and the model it was based upon. There are several gray areas involved in each step of the way in developing these models. The first problem is determining a specific definition and evaluation level as being language proficient or sufficiently language deficient to cause concern regarding the student’s success in the English-speaking classroom or ability to function fully in an English-speaking society, warranting inclusion in ESL or other specialized instruction. Another sticking point is whether to measure language skills as a group of separate but related parts or to consider whether the student has a grasp of English as an integrated language, with full ability to readily understand those aspects of language such as idioms that defy literal translations. Because of these issues, the testing is typically designed to closely mimic the skills and abilities that would be necessary for an average student participating in a daily English-speaking class to determine whether the LEP student would be able to successfully navigate such an environment. The results of the tests are then used to help determine the best course of instruction for the child in order to provide him/her with the greatest chance for success and greatest English language integration. There are five standardized tests discussed in the study. These include the Basic Inventory of Natural Language (BINL), The Bilingual Syntax Measure (BSM) I and II, the Idea Proficiency Tests (IPT), the Language Assessment Scales (LAS) and the Woodcock-Muñoz Language Survey. The BINL test consists of showing a series of large photographs to the student and allowing the student to make free, unsolicited comments regarding the picture. These can be descriptions, stories about someone the picture reminds him/her of, events, discussions of aesthetic quality or whatever the student chooses to discuss. The important element is recording the language the student uses to convey these thoughts to the tester. The language sample is then evaluated to determine his/her fluency, level of complexity and average sentence length as compared to average sentence length in the native language. Because of the flexibility of this test, it can be used to test students from a wide variety of language backgrounds. It is intended to test primarily the student’s oral language proficiency. Comparing scores from the first five sentences to the second five sentences in the same test showed a correlation of .925, but some discrepancies were shown when students were retested, often showing a higher score on the retest but still reflecting a high degree of reliability. Validity is based upon correlation between 116 holistically scored writing samples from a southern California high school, teacher oral proficiency judgments of 263 students and three separate studies using reading passages of varying complexity, all subsequently evaluated using BINL standards. Like the BINL, the BSM is designed to test oral language proficiency, but is meant for an older student, at least having reached third grade. This is because the test is comprised of a series of cartoon drawings shown to the child with a series of prepared questions posed to the child by the tester. Also like the BINL, the BSM relies on a recorded language sample being analyzed, but goes beyond the BINL in that it is able to also evaluate whether the child provides appropriate responses to the questions asked as well as whether or not they are able to answer in accurate grammatical structure. However, because of its greater specificity, the test is not as easily adapted to other languages. The reliability of the test was tested in several ways. It was first run through several series of a test-retest reliability procedure and scorers were also tested for consistency and reliability in their evaluations of the scores. Both sets of tests showed a high degree of reliability. Validity is supported by “the body of psycholinguistic theory and research supporting the concept of natural sequences of acquisition; the statistical procedures and research which have resulted in the production of the acquisition hierarchy and of the scoring system which is based on the hierarchy; and evidence that the BSM classifications reflect the relationships expected to be found among bilingual children” (Del Vecchio & Guerrero, 1995). The IPT goes further than either the BINL or the BSM in that it is designed to not only test oral ability in students, but also written and reading ability in students as young as kindergarten all the way through adult. The testing is done in segments, with oral testing being conducted individually while reading and writing tests can be given in a small group atmosphere. The tests involved in this procedure measure students in their vocabulary comprehension, syntax usage and reading for understanding. Following the discrete point model, these tests generally tend to keep the various aspects of the language separate, for example testing syntax separately from sound recognition separate from vocabulary. Again, because of the large degree of specification and prepared materials, these tests are limited in the range of languages in which they are available. Reliability for this particular series of tests is rated as extremely high in the oral capacity, but a test-retest procedure showed less spectacular results for the reading and writing segments of the test. While still moderate to high, the reliability for these sections was not as strong as the oral results. Validity for the tests covered a great deal of discussion into the content, construct and criterion. The LAS is similar to the IPT in that it works to test oral proficiency, reading and writing abilities in students aged kindergarten through adulthood. It also must be delivered in two parts, with the oral section of the test administered individually and the written and reading portions administered to groups if desired. Also like the IPT tests, the LAS is based more upon the discrete point model. However, where the LAS differs from the IPT is in its inclusion of several aspects of English, such as vocabulary, minimal pairs, listening comprehension and story retelling. Reliability for this test shows a wide range of results, typically scoring from moderate to high results, but with a swing of approximately .20 points on the written and reading portions of the test while oral sections tested somewhat higher and with a smaller degree of swing. The reliability is increased with a standard training requirement for scorers. Validity for the test starts with its basis on the assumptions that linguistic subsystems can be assessed independently of one another and later combined to obtain a single score. It is also based upon statistical evidence that supports the oral portion of the test. Measuring content such as vocabulary, verbal analogies and letter-word identification, the Woodcock-Munoz Language Survey is the only test developed to test students as young as four years old. Much more complicated in nature than the IPT or LAS, all parts of this test must be administered individually. This test is administered through the easel technique, in which the examiner is provided with a view of both the question and the answer for each item while the easel itself shields this information from the student, presenting him/her with only the question. The study reports there is extensive evidence available to prove the reliability and validity of the survey. Reliability scores were consistently in the .90s and above, scoring much higher than all of the other tests presented. Validity is proven with studies comparing survey results with the results of other tests. Construct validity is established through several means, all of which favor this test highly. The BINL and the BSM pave the way to allow testers to evaluate a student’s overall language proficiency while lessening the pressure of the testing environment and allowing the student to draw on any interrelated language skills in testing. However, because of the way in which the BINL is scored, it leaves room for much oversight or possible cultural bias. This is because the test is often scored by an impartial machine when large numbers of tests have been given or personally by individual scorers who have not been tested for reliability. BSM tests have studied the reliability of scorer interpretations, but this does not eliminate the possibility of scorer bias. The strength of the IPT test includes its great deal of research support and its high level of oral reliability, but, like the LAS and Woodcock-Munoz Language Survey, it lacks the ability to test the student’s grasp of language on a contextual basis. The LAS requires training for its examiners, making results somewhat more reliable and consistent. Although it is also based on a discrete model of testing English proficiency by testing each construct independently, the Woodcock-Munoz Language Survey is shown to be the strongest of the five testing methods available. References Del Vecchio, Ann & Guerrero, Michael. (December 1995). Handbook of English Language Proficiency Tests. Albuquerque, NM: EAC-West, New Mexico Highlands University. Read More

The language sample is then evaluated to determine his/her fluency, level of complexity and average sentence length as compared to average sentence length in the native language. Because of the flexibility of this test, it can be used to test students from a wide variety of language backgrounds. It is intended to test primarily the student’s oral language proficiency. Comparing scores from the first five sentences to the second five sentences in the same test showed a correlation of .925, but some discrepancies were shown when students were retested, often showing a higher score on the retest but still reflecting a high degree of reliability.

Validity is based upon correlation between 116 holistically scored writing samples from a southern California high school, teacher oral proficiency judgments of 263 students and three separate studies using reading passages of varying complexity, all subsequently evaluated using BINL standards. Like the BINL, the BSM is designed to test oral language proficiency, but is meant for an older student, at least having reached third grade. This is because the test is comprised of a series of cartoon drawings shown to the child with a series of prepared questions posed to the child by the tester.

Also like the BINL, the BSM relies on a recorded language sample being analyzed, but goes beyond the BINL in that it is able to also evaluate whether the child provides appropriate responses to the questions asked as well as whether or not they are able to answer in accurate grammatical structure. However, because of its greater specificity, the test is not as easily adapted to other languages. The reliability of the test was tested in several ways. It was first run through several series of a test-retest reliability procedure and scorers were also tested for consistency and reliability in their evaluations of the scores.

Both sets of tests showed a high degree of reliability. Validity is supported by “the body of psycholinguistic theory and research supporting the concept of natural sequences of acquisition; the statistical procedures and research which have resulted in the production of the acquisition hierarchy and of the scoring system which is based on the hierarchy; and evidence that the BSM classifications reflect the relationships expected to be found among bilingual children” (Del Vecchio & Guerrero, 1995).

The IPT goes further than either the BINL or the BSM in that it is designed to not only test oral ability in students, but also written and reading ability in students as young as kindergarten all the way through adult. The testing is done in segments, with oral testing being conducted individually while reading and writing tests can be given in a small group atmosphere. The tests involved in this procedure measure students in their vocabulary comprehension, syntax usage and reading for understanding.

Following the discrete point model, these tests generally tend to keep the various aspects of the language separate, for example testing syntax separately from sound recognition separate from vocabulary. Again, because of the large degree of specification and prepared materials, these tests are limited in the range of languages in which they are available. Reliability for this particular series of tests is rated as extremely high in the oral capacity, but a test-retest procedure showed less spectacular results for the reading and writing segments of the test.

While still moderate to high, the reliability for these sections was not as strong as the oral results. Validity for the tests covered a great deal of discussion into the content, construct and criterion. The LAS is similar to the IPT in that it works to test oral proficiency, reading and writing abilities in students aged kindergarten through adulthood. It also must be delivered in two parts, with the oral section of the test administered individually and the written and reading portions administered to groups if desired.

Read More