张青青

The English test in senior high school entrance examination is a comprehensive consideration of students English language proficiency. Scientific， reasonable， and effective examination paper analysis can help to improve the teaching level of teachers. This article uses SPSS 22.0 to carry out various quantitative analysis of the test scores of the practice test one of English in order to provide a scientific basis for the evaluation of teaching quality and the improvement of teaching effects.

1. Test Description and Data Analysis

1.1  Test Description

In this study， sixty-one students from three classes of a secondary school in Qinhuai District， Nanjing City take part in this examination， hence there are 61 valid test papers in this research. The total score of this test is 90 points， including the objective questions （75 points） and the subjective question （30 points）. Objective questions include single-choice questions （15 points）， cloze questions （10 points）， reading comprehension questions （15 points）， fill-in questions （15 points）， and reading fill-in questions （15 points）， and the subject question only includes writing.

1.2 Basic Data Analysis

The maximum score of the exam is 84 points， Minimum=12 points， Range=72 points， Mean=60.70 points， and Std. Deviation=17.744 points. The standard deviation is too large， which shows that the students in the three classes have large differences in English proficiency， and there exists serious two-level differentiation.

1.3 The Frequency Distribution of Test Scores

As a rule of thumb， values for Skewness and Kurtosis should between -2 and +2. From table 2， we can see that Skewness -0.668 and Kurtosis -0.423 are in this range. With the help of the values of Skewness and Kurtosis and the normal curve， we could draw the conclusion that students scores are in normal distribution. This analysis proves that the papers proposition is reasonable， and the knowledge points reflected in the test paper are comprehensive.

1.4 Difficulty Index Analysis

The difficulty index （marked as P） reflects the degree of difficulty of the test， it is the average score rate for the test. The more people who answer questions correctly， the higher the average score rate will be， thus the higher the value of P will be.

The general difficulty index is in the range of 0.3～0.8. Higher than 0.8 is too easy， and lower than 0.3 is too difficult. Each question type of this paper is basically concentrated between 0.65～0.81， and the overall difficulty index for objective questions is 0.81， indicating that the question type is relatively easy， especially for multiple-choice questions.

2. Conclusions and Suggestions

（1） This test paper is not very challenging， and many students have achieved good results in this test. However， there is a large gap between the individual students in the class and the two levels of differentiation are serious. Teachers should help poorly-scored students to improve their ability and make up for the overall performance.

（2） This papers proposition is reasonable， and the knowledge points reflected in the test paper are comprehensive. Students could basically achieve teaching objectives in terms of knowledge structure and ability development， but teachers should pay attention to the diversity and comprehensiveness of reading comprehension themes and genres.

（3） The difficulty index of each question of this paper is basically concentrated between 0.65～0.81. The general difficulty index of objective questions is 0.81， indicating that the objective questions are relatively easy， especially for multiple-choice questions. It is suggested that this kind of questions should be appropriately increased with some moderately difficult questions.

References

[1]Bachman， L.F. Fundamental Considerations in Language Testing[M]. Shanghai：Shanghai Foreign Language Teach-ing Press， 1999.

[2]Hughes A. Testing for Language Tests[M]. Beijing： Foreign Language Teaching and Researching Press， 2000.

（作者單位：上海理工大学）