语言测试short answer_爱的五种语言的测试题

语言测试short answer由刀豆文库小编整理，希望给你工作、学习、生活带来方便，猜你可能喜欢“爱的五种语言的测试题”。

Shortanswer(6×5`)

(Requirements: Answer the following questions with brief and exact expreions in your own words)

已选：

讨论3

1.What is the difference between validity and validation?

Validity is often regarded as involving the question of whether a test measures what it intends to measure, which is an abstract concept, and validation is the proce of operationalizing it.Validation is a long-term proce of accumulating, evaluating, interpreting, conveying, and refining evidence from multiple sources to fulfill particular purposes.Validity is the most fundamental consideration in developing and evaluating language tests, while at the practical level, the validity of a test can only be established through a proce of validation.(The concept of validity guides validation practices.)

讨论4

2.How can we provide some feedback which given more information in details to the test-takers in scoring reporting?

Scores should be accurately reported in a timely manner.the information should be clearly presented in non-technical terms that are interpretable by laypersons.the information should be detailed enough to allow test-takers and their teachers to plan a course of action for improving the learners' target language skills.Clearly,further investigation into the effects of score reporting is needed.讨论 8

3.What strategies would emerge when raters using rating scale of writing?

1，The first strategy is aigning a global score to a script, usually afterthe first reading.2.The overall, holistic type rating often results in a halo effect, where a rater awards the same score for a number of categories on the scale.3.Some raters seem to disregardthe descriptors and override them with their own general impreion of a script.Chapter 5

4.What are the widely used ways to mutilate a text? And try to illustrate each of them.the first is known as rational deletion, in which the test developer deletes words on the basis of some rational decision.For example, i want to test my students' knowledge of verb tenses, i might delete only verbs.The second method is known as fixed ratio deletion or nth word deletion, which means that , regardle of its part of speech or the semantic load it bears within the text, every nth word is omitted and a blank line is inserted in its place.For example, every five word or every seventh word is omitted.Chapter6

5.Why do people use indirect measures at all?

It’s partly because many indirect tests are discrete-point in focus, and for many years

discrete-point testing was thought to be the fairest, most objective, and most precise way to evaluate language skills.Chapter 8

6.Why is it important to make distinctions between types of data in language aement?

These distinctions are very important in language aement because they will determine, in part, the nature of the claims that we can make on the basis of our data.The type of data we are working with will also determine, to a certain extent, the types of data we are working with also determine, to a certain extent, the types of statistical analyses we can legitimately conduct.备用：

Chapter 1

Please list four criteria by which tests are normally evaluated?

1.Validity

2.Reliability

3.Practicality

4.Positive washback

Chapter 2

Please list four components of a test according to standard dictations and Wesche’s Framework.1.Stimulus material

2.Task posed to the learner

3.Learner’s response

4.Scoring criteria

Chapter 3

What is Norm-referenced test? Can you set some examples of norm-referenced testing?

Norm-referenced tests are those aociated with the familiar bell-shaped curve, which is referred to in the phrase “grading on a curve”.In this philosophy, grades or scores are based on a comparison of the test-takers to one another.Chapter 7

What is the main reason for language teachers’ being familiar with the statistics used in language aement？

Because the data language teachers glean by computing and interpreting some statistical procedures provide us with the information they need to insure that their aement devices are valid and reliable.This is important because students’ lives may be affected by the outcomes of the language tests they use.If the tests are invalid and unreliable, the students may suffer and their judgments may be called into question.讨论 1

How many stages are there in test development conceptually?

Test development conceptually into 3 stages: design, operationalization and administration.讨论 5

How to calculate item discrimination?

1.subtracting the low-scoring group' proportion correct from that of high-scoring group.2.2.looking at the point biserial which is the correlation between individual item responses and total test score.Generally,point biserial correlations of 0.25 and better are acceptable.讨论6

What did Bachman suggest in order to ‘maximize’ the reliability of test scores and the validity of the test use?

He suggested that we should provide clear and unambiguous theoretical definitions of the abilities we want to measure and specify precisely the conditions, or operations that we will follow in eliciting and observing performance.讨论7

How many item types are there in New TOFEL reading section and what are they?

There are ten item types representing five Basic comprehension tasks that include vocabulary, pronoun reference, sentence simplification,factual information and negative fact;three Inferencing tasks that include inferencing, rhetorical purpose and insert text;and two Reading to learn tasks that include prose summary and schematic table.