




已阅读5页,还剩1页未读, 继续免费阅读
版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领
文档简介
Table 4.6 below is the results of paired samples t-test (Naming-Matching). The two variables in the analysis are score of naming test and score of matching test. The figures in the table show that the two tailed significance is 0.764; and the lower level is -4.548 and the upper level is 3.382. Zero is included in the span from lower level to upper level. All these well prove that there is no significant difference between the result of naming test and that of matching test.Table 4.6 Paired Samples Test (NAMING-MATCHING)Paired Samples TestPaired DifferencestdfSig. (2-tailed)MeanStd. DeviationStd. Error Mean95% Confidence Interval of the DifferenceLowerUpperPair 1NAMING - MATCHING-.5839.3901.917-4.5483.382-.30423.764Table 4.7 Paired Samples Test (CHECKLIST-NAMING)Paired Samples TestPaired DifferencestdfSig. (2-tailed)MeanStd. DeviationStd. Error Mean95% Confidence Interval of the DifferenceLowerUpperPair 1CHECKLIST - NAMING4.1676.8251.3931.2857.0492.99123.007 Then we come to the paired samples t-test (Naming-Checklist). The two variables in the analysis are score of naming test and score of checklist test. In table 4.7 above, we can see that the two tailed significance is 0.007, which is statistically significant; and the lower level is 1.285 and the upper level is 7.049. Zero is not included in the span from lower level to upper level, and thus, conclusion comes that the difference between the two variables is significant. That is, the result of naming test is significantly different from that of checklist test.The Paired-item Significant Difference AnalysisThe analysis in the previous section shows that the total scores of naming test and checklist test are significantly different, but the difference between the two tests and matching test is not significant. Further study about the difference between the corresponding items in two tests is conducted by pairing the concerned corresponding items of two tests. Accordingly, there are altogether 60 pairs of items and the only difference for each pair is the test format. Here the author not only pair the items of naming test with checklist test, but also pair naming test items and checklist test items with items of matching test respectively. Paired samples t-tests were conducted to explore the results. Therefore, there are three paired-item analysis altogether. As for the paired item significant difference analysis, the case is a bit different from that of the total score analysis. First, in the paired-item analysis of naming test and checklist test, no significant difference is found in 56.7 percent of the items and 43.3 percent of the items show significant difference according to the 95% confidence interval of difference.Then, in the paired-item analysis of naming test and matching test, no significant difference can be found in 83.3 percent of the 60 items; only scores of 16.7 percent of the items are significantly different from each other.And in the paired-item analysis of checklist test and matching test, the results seem to be more complicated. Although no significant difference is found between the total scores of these two tests, the paired-item analysis indicates that there are significant differences in 33.3 percent of the 60 items between these two tests. Details of these items will be analyzed in the discussion part.The Item DifficultyItem difficulty is an important index of a test. Item difficulty expresses the proportion or percentage of participants who answered the item correctly. The item difficulty index (P) is calculated by P=R/T, in which R is the number of correct answers and T is the total number of participants. Item difficulty can range from 0.0 (none of the participants answered the item correctly) to 1.0 (all of the participants answered the item correctly). Average difficulty index is between 0.4 and 0.6.In the descriptive statistics, we have seen that the mean of checklist test is 38.92; then we can calculate the difficulty of the test by dividing the total score, that is, 60 in the this study, with the mean of the test. So the difficulty of the checklist test is 0.65. And in the same way, we can get the difficulty of matching test and naming test; which is 0.59 and 0.58 respectively. That is to say, difficulties of the three tests are all acceptable.Besides, this research has undergone item difficult study on each test to make it more concrete. The results are shown in table 4.6 below:Table 4.8 The Item Difficult StudyP0.20.2P0.40.4P0.60.6P0.8P0.8CHECKLIST0%6.7%20%23.3%50%MATCHING0%16.7%36.7%30%16.7%NAMING3.3%20%36.7%16.7%23.3%P value in the table means the percentage of the participants who have got the correct answer to the number of the total participants. The table shows that the checklist test is on the easy side for the participants. Half of the items are easy for them and none of the items seem to be very difficult for them. The difficulty of the matching test and naming test is quite acceptable with most of the item difficulty falls between 0.4 and 0.6. Words tested are the same, but item difficulty is quite different. Items in matching test and naming test are comparatively much more difficult than in checklist test. Tentative explanation for this situation is that the participants may have over-estimated their vocabulary knowledge.Besides, by analyzing the paired items of checklist test and naming test, it is found that 50 pairs of items increase in the difficulty from the one in checklist test to the one in naming test in the whole 60 pairs.The Item DiscriminationAn index of item discrimination is the point-biserial correlation, that is to say, how well the item serves to discriminate between students with higher and lower levels of knowledge. The point-biserial correlation reflects the degree of relationship between scores on the item and total test scores. The item discrimination index (D) measures the difference between the percentage of participants who answered the item correctly in the upper group (PU), i.e., the top 27 percent scorers, and the percentage of those in the lower group (PL) who got the correct answer, i.e., the bottom 27 percent scorers; thus D=PU-PL. The index will be positive if better students answered the item correctly more frequently than poorer students did, and negative if the opposite occurred. A negative index is denoted by a minus sign in front of the value.The value of a positive point-biserial discrimination index can range between 0 and1; the closer the value is to 1, the better the discrimination. The value of a negative point-biserial discrimination index can range between -1 and 0, but positive values are desirable. Item discrimination is greatly influenced by item difficulty. Items with a difficulty of either 0 or 1 will always have a discrimination index of 0, and item discrimination is maximized when item difficulty is close to 0.5. As a general rule, discrimination index of 0.2 and above are considered to be desirable.Table 4.7 The discrimination index of the three testsTest FormatDiscrimination IndexCHECKLIST0.47MATCHING0.65NAMING0.43The figures in table 4.7 indicate that the three tests have well discriminated participants of different levels for their discrimination index are higher than 0.2. The analysis of discrimination was made to each item as well. For the checklist test, discrimination index of 28 of the 60 words tested is above 0.2. And discrimination index of 46 of the 60 words tested in naming test is above 0.2. Furthermore, when it comes to compare each concrete paired-item to see if there is any difference in discrimination within each pair, it finds out that 40 of the 60 pairs of items in naming test has better discrimination than the corresponding one in checklist test. When comparing the discrimination index of checklist test and naming test to matching test, the overall discrimination of matching test is the best of the three. And in the paired-item comparison, 44 items of the matching test has a higher discrimination index than the corresponding one in the checklist test; and 32 items of the matching test possess a higher discrimination index than the corresponding one in the naming test.The only difference between these three tests is the testing format; therefore, all these can well prove that the matching format enjoys better discrimination than the checklist format and naming format though the discrimination index of the two tests are acceptable.The criterion-related analysisAs stated in chapter two, criterion-related validity is a predominantly quantitative and a posteriori concept, concerned with the extent to which test scores correlate with a suitable external criterion of performance with established properties. Criterion-related validity divides into two types: concurrent and predictive. Concurrent validity looks for a criterion which we believe is also an indicator of the ability being tested (Bachman 1990:248). Test scores could be correlated with another measure of performance, usually an older, longer, established test, taken at the same time for teachers rankings students, or even student self-assessment.In the present research, a matching test was designed to be the external criterion. Matching test was chosen because it has been proved by many researchers to be a valid test of vocabulary size and has been widely used. The author has verified validity and reliability of the matching test used in this study before the experiment was carried out.Correlation analysis in this study is conducted mainly between the matching test scores and that of the other two tests, using the Pearson Correlations. Pearsons correlation reflects the degree of linear relationship between two variables. On general sense, r is used to present Pearsons correlation. Pearsons r can range from -1 to +1. An r of -1 indicates a perfect negative linear relationship between variables, an r of 0 indicated no linear relationship between variables, and an r of 1 indicates a perfect positive relationship between variables.Table 4.9 shows that the Sig. (2-tailed) is 0.000 for the correlation concerned which means the result is statistically meaningful. The Pearsons r in table 4.9 is 0.810, which means the scores of checklist test are highly correlated with that of matching test.Table 4.9 The correlation between Checklist and MatchingCorrelationsCHECKLISTMATCHINGCHECKLISTPearson Correlation1.810*Sig. (2-tailed).000N2424MATCHINGPearson Correlation.810*1Sig. (2-tailed).000N2424*. Correlation is significant at the 0.01 level (2-tailed). Table 4.10 shows that the Sig (2-tailed) is 0.000 for the correlation concerned. The Pearsons r in table 4.10 is 0.814, which is a little higher than the correlation between Checklist and Matching. That is, naming test are also highly correlated with matching test.Table 4.10 The correlation between Naming and MatchingCorrelationsMATCHINGNAMINGMATCHINGPearson Correlation1.814*Sig. (2-tailed).000N2424NAMINGPea
温馨提示
- 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
- 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
- 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
- 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
- 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
- 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
- 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
最新文档
- 农业科技公司员工保密合同
- 简易手机抵押合同范本5篇
- 购销合同轮胎3篇
- 安全施工现场培训课件
- 瓶子里的小昆虫课件
- 理财培训课件
- 电力工程采购方案(3篇)
- 安全文明施工培训感想
- 贵港港平南港区长岐塘作业区罗洪码头提档升级工程环评报告
- 环卫作业管理知识培训课件
- 2025年中国电信招聘考试行政职业能力测试预测题集
- 静脉治疗知识培训课件
- 学风建设科研诚信宣教课件
- 江西省宜春市2025年上半年事业单位公开遴选试题含答案分析
- 2025繁轩科技发展(天津)有限公司公开招聘工作人员35人备考题库及答案解析
- 2025年度水电项目工程结算与审计服务协议
- 德育副校长在新学期德育工作部署会讲话范文
- (2025年标准)学生玩耍纠纷协议书
- 南昌二中初一数学试卷
- 2025-2026秋季学年第一学期学生国旗下演讲稿(20周):第五周 76载荣光里我们茁壮成长-喜迎国庆
- 2025金融消费权益保护知识竞赛题库(+答案)
评论
0/150
提交评论