1、Criteria of Tests,測試的標準,Validity 效度 Reliability 信度 Power/Difficulty 難度 Discrimination 區(qū)分度 Practicality 實用性 Backwash effects 后效作用,Criteria of tests,Validity,The validity of a test is the extent to which it measures what it is supposed to measure and nothing else. 效度是指一套測試所考的是否就是設(shè)計人想 要考的內(nèi)容，或者說，在多大程度上考

2、了想 要考的。,Discuss on the following items:,“Is photography an art or a science?” Discuss. “The mind is in its own place, and itself can make a Heaven of Hell, a Hell of a Heaven.” (Milton) Discuss.,Use the following words in sentences: courageous, choosy, acceptable, complicated, etc. A. John is a very

3、 courageous boy. B. John, the captain of our team, is courageous. C. I have a courageous father.,Factors of validity,Face validity 表面效度 Content validity 內(nèi)容效度 Construct validity 結(jié)構(gòu)效度 Empirical validity 實驗效度 Concurrent validity 共時效度 Predictive validity 預測效度,Face validity,If a test item looks right to

4、other testers, teachers, moderators, and testees, it can be described as having at least face validity. 表面效度指考試表面的可信度或公眾的可接受程度。 鄒申：一個考試看上去具有了擬定的技能或能力測試。（測語音語調(diào)用筆頭考試來測則表面效度低。）,Content validity,A test is said to have content validity if its content constitutes a representative sample of the language sk

5、ill, structures, etc. with which it is meant to be concerned. 內(nèi)容效度指測試是否考了考試大綱規(guī)定要考的，或者說考試的題目在多大程度上能代表它所要測量的目標。,Is the content of a test related to the objective or purpose of it? Are the test items representative? Is the content appropriate or suitable for the testees?,Construct validity,If a test ha

6、s construct validity, it is capable of measuring certain specific characteristics in accordance with a theory of language behavior and learning. 結(jié)構(gòu)（構(gòu)卷）效度指測試是否以有效的語言觀（包括語言學習觀和語言運用觀）為依據(jù)。這里的結(jié)構(gòu)并不是指試卷的結(jié)構(gòu)或題目的編排，而是指整個考試的理論基礎(chǔ)。,Empirical validity,This validity is obtained as a result of comparing the results

7、 of the test with the results of some criterion measure. 實驗（統(tǒng)計）效度是將考試結(jié)果與其它測量結(jié)果相比較而得來的。它又可分為共時效度和預測效度。,Concurrent validity,If the results of the test are compared with the results of some criterion measure such as: an existing test, known or believed to be valid and given; or the teachers ratings or

8、any other such form of independent assessment given at the same time, then results obtained by either of the above two methods are measures of the tests concurrent validity in respect of the particular criterion used.,In other words, concurrent validity is established when the test and the criterion

9、 are administered at about the same time. 共時效度是將一次測試的結(jié)果同另一次同時或時間相近的測試的結(jié)果相比較，或同教師對學生的評估相比較而得出的系數(shù)。例如拿期末考試成績與剛剛結(jié)束的四級考試成績相比，假若得分情況相似，則說明期末測試有較高的共時效度。 （前提：四級考試效度很高。）,Predicative validity,If the results of the test are compared with the results of some criterion measure such as: the subsequent performance

10、 of the testees on a certain task measured by some valid test; or the teachers ratings or any other such form of independent assessment given later, then results obtained by either of these two methods are measures of the tests predicative validity in respect of the particular criterion used.,In oth

11、er words, predicative validity concerns the degree to which a test can predict the testers future performance or success. 預測效度涉及測試的預測能力，即測試結(jié)果到底在多大程度上能夠預測出某些將來會發(fā)生的可能性，或者說考試是否具有預測學生未來表現(xiàn)或成績的功能。,A Test is said to be reliable if it is consistent in its measurements. 信度是指考試結(jié)果的可靠性和穩(wěn)定性。例如 拿一份卷子對同一組學生實施兩次或多次

12、測 試，如果結(jié)果很一致，則說明該測試的信度 較高。,Reliability,驗證測試信度的方法,考后復考法 (test/retest method) 試題分半法 (split-half method) 平行試題法 (parallel forms method),test/retest method,This method is to re-administer the same test after a lapse of time. It is often impracticable since certain students will benefit more than others by

13、 a familiarity with the type and format of the test. Moreover, in addition to changes in performance resulting from the memory factor, personal factors such as motivation and differential maturation will also account for differences in the performances of certain students.,split-half method,This met

14、hod estimates a different kind of reliability from that estimated by test/re-test procedure. It is based on the principle that, if an accurate measuring instrument were broken into two equal parts, the measurements obtained with one part would correspond exactly to those obtained with the other.,par

15、allel forms method,This method is to administer parallel forms of the test to the same group. This assumes that two similar versions of a particular test can be constructed: such tests must be identical in the nature of their sampling, difficulty, length, rubrics, etc. only after a full statistical

16、analysis of the tests and all the items contained in them can the tests safely be regarded as parallel. If the correlation between the two tests is high, then the tests can be termed reliable.,影響考試信度的因素,題量 題目性質(zhì) 題目區(qū)分度 成績分布 題目難度 評分是否客觀 考試的時間,Power/Difficulty,難度是指一套試題中每個題目的難易程度。分析一套試卷的質(zhì)量如何，除了看其信度和效度這兩個

17、重要指標之外，還要研究試題的難度指數(shù)(index of difficulty/facility value)，即試題的難易度。,難度值的計算公式,題目的難度通常用P來表示，P值實際上指的是答對題目的比率。假設(shè)有10名考生，某道題有8人答對，那么該題的難度值為：,適用于主觀性試題的公式,假設(shè)某寫作題的滿分為20分，所有考生在這道題上的得分的平均分為16分，則該題的難度值為：,正態(tài)分布圖,Discrimination,Discrimination of a test is its capability to discriminate among the different candidates a

18、nd to reflect the differences in the performance of the individuals in the group. 區(qū)分度指一個題目區(qū)分考生能力的程度。,計算題目區(qū)分度的方法,公式法 點雙列相關(guān)系數(shù)法 雙列相關(guān)系數(shù)法,Practicality,A good test is practical. It is within the means of financial limitations, time constraints, ease of administration, and scoring and interpretation. 實用性是指試題是否便于使用以及實施 起來是否可行。,Factors affecting practicality,the length of time available for the administration of the test the answer sheet and the stationery used the test situation the necessary eq