Assessment of young English language learners

Standardized Tests used with Young English Language Learners. When students are newly admitted into schools, their parents or caregivers fill out a home language survey or a home language questionnaire. These evaluative instruments, used for preliminary screening, contain questions about a language or languages that children and their caregivers speak at home. Overall, this type of preliminary screening works fairly reliably for the identification of English language learners. However, there is evidence that due to immigrant parents’ lack of English language proficiency, or concerns about their legal status in the United States, or reluctance to have their children identified as limited English proficient, language surveys and questionnaires are not always filled out correctly. [1]

If the survey or questionnaire does reveal that a child speaks a second language at home, the school must administer a federally mandated standardized language test to determine whether or not a child is proficient in English and whether or not she would benefit from placement in a second language program. Usually, the test is administered again at the end of the school year to measure the child’s second language attainment and academic language development. Once standardized test demonstrates that a child has achieved a level of proficiency that enables him or her to function in the mainstream classroom, the student exits a second language program.

The question considered in this article is as follows: What features does the language test need to possess in order to be useful?

To answer the question above, let us compare a language test to another measurement device such as a thermometer, a scale, or a speedometer. While the thermometer measures temperature, the scale measures weight, and the speedometer measures speed, the language test is meant to measure a child’s command of language or demonstrate to the public the rate of academic progress of English language learners.

In order to be reliable, a measuring device needs to be well made, that is, it needs to be well calibrated and well constructed. The same holds true of a language test. What kind of a language test can be described as well made? Some of the criteria that determine reliability of assessment instruments meant for young second language learners are as follows:

❖ The language test has clear directions. It tells language learners what exactly they need to do in simple, clear, and unambiguous language.

• ❖    If the test has an open-ended task, it explicitly tells test takers the extent of the desired response.

• ❖    The test is made of tasks that are free of ambiguity and easy to interpret.

• ❖    The print and graphic materials are user-friendly; its pages do not look overcrowded; its graphics are clear, simple, and attractive.

• ❖    The language test is neither too short nor too long.

• ❖    The well-made assessment tool also has provisions for reliable scoring. By providing scoring rubrics and unambiguous scoring guidelines, it enables the

test reader to interpret test results accurately and to assign correct scores to test takers.

• ❖    The test site staff and physical environment provide a supportive atmosphere. Learners’ state of mind at the time they take a language test and the physical environment at the test site are all part of test reliability. If the learner is overly nervous or distracted at the time of testing, the test will fail to provide accurate results. [2]

We can find out whether a language test is reliable or not by performing some relatively simple procedures. One of the reliability evaluation procedures is known as a split half test. When this procedure is implemented, a group of students complete all even numbered tasks contained in a test and then do all odd numbered tasks. If both times students get the same (or almost the same) score, the test is reliable. Another method used to evaluate reliability of a language test is the so-called test–retest procedure. When this method is used, two different versions of the same test are taken by the same group of individuals within a short time period. If subjects of the test–retest procedure get a low score one time and a high score a second time, the test is clearly unreliable. Conversely, when each individual who participates in a test– retest experiment earns the scores that fall within the close range of each other, the test is pretty reliable.

Reliability is not all there is to a good language test. A good test is also valid. To understand the concept of validity, let us again consider a scale analogy. There are all kinds of scales out there. Some scales are used for weighing human bodies, others are for weighing huge containers, and still others are for weighing miniscule amounts of chemicals. It is not enough that these measuring tools should be well made and reliable. It is also important that they be able to provide the kind of information that one needs to collect. You would not get very far if you tried to weigh electronic particles with a bathroom scale no matter how well that bathroom scale is designed. It is not enough that a measuring instrument be well made or reliable. For any measuring procedure, you need to be assured that the device you are using can provide the kind of data that you seek to obtain. Language tests used in schools are no exception. These assessment tools should also be appropriate for educators’ intentions and have the capacity to provide the type of data which educators are interested in gathering. [3]

Unlike test reliability, test validity is not quantifiable. There is no simple number that can express the test’s capability to provide users with the kind of information that they are seeking to obtain. The job of assessing test validity is performed by teams of capable experts. Qualified, experienced language educators alone can say whether tasks contained in a test can really measure language proficiency or language attainment of language learners.

One of the most commonly made recommendations is taking classroom teachers’ appraisal into account when evaluating language development and academic attainment of language learners. “Whose judgment counts?” ask second language educators dismayed by the fact that their perspective is not taken into consideration in the evaluation of language learners. [4]

Creating a Comprehensive Assessment System for Young Children While it sometimes feels like we have made little progress, it is important to look at where we are today. There is far more integrity in our approach to assessment and how what we learn can be best used to support families and children whose home language is not English. Perhaps one of the most important realizations is that in order for children to make progress in educational settings, we must approach assessment from the position of strengths and that the students must be valued, which includes embracing their language and culture in our classrooms. [5] The National Association for the Education of Young Children (NAEYC) and the National Association of Early Childhood Specialists in State Departments of Education (NAECS / SDE) have written a powerful position statement that addresses the need for appropriate and responsive assessment of children, curricula, and programs to assure that the needs of all young children are met. As the professional organization for early childhood, NAEYC has published several statements that guide the assessment practices for diverse groups of young children [6]. Whereas in the past there were relatively few screening tools, there are now choices, and many of them address needs of English learners (ELs).

Perhaps the greatest gain is the understanding of the importance of using a comprehensive assessment system that includes various tools and strategies to gain a holistic picture of a child. We understand the importance of observation as a primary tool for data collection, whether using standardized assessments or authentic assessments within an environment that includes play and learning centers; however, we also embrace the fact that we need to use research-based tools, goals, and benchmarks to assess progress and appropriate support for development and learning across domains. We collect work samples, use screeners and checklists, have conversations, document experiences with pictures and detailed data, and most importantly, we open ourselves up as learners in the process who examine our own assumptions. Parents are welcomed into the process and seen as partners in making sense of our data. We never hesitate to have conversations with children to gain their perspectives. The assessment process is recursive – it never ends. [7]

Researchers also recommend that the context of student learning and students’ educational backgrounds should be factored in during the assessment process. According to this scenario, language learners’ home background, previous educational experiences, and the students’ current classroom environment should be taken into consideration in the assessment process.

Given that classroom teachers observe their students speak, read, and write English on a daily basis and in a natural communication context, researchers and classroom practitioners argue that including classroom generated data in the evaluation of language learners can render language learners’ assessment both more accurate and more valid.

It is important that teacher-made assessment instruments be reliable and valid, second language educators know that the quality of a teacher-made assessment tool cannot be reduced to its reliability or validity. Nor is the most innovative assessment tool inherently effective persent. The usefulness of assessments administered in the language classroom has everything to do with the quality of instruction. If instruction challenges and stimulates young language learners, assessment tools are likely to be conceived in the same intellectually challenging and stimulating spirit. [8]

In summary, good assessment instruments used with young language learners need to be reliable (accurate) and valid (produce the kind of data that educators seek to obtain), second language educators question reliability and validity of standardized proficiency and high-stakes tests used with young language learners. Suggestions have been made to take teacher judgment into account when evaluating language proficiency and academic attainment of young language learners. Some of the innovative assessment instruments used by second language teachers include journals, observations, and portfolios.