中文摘要：

目的分析9张普通话双音节词表应用于听力正常青年人的复测信度，得出其临界差值，以利于对测试结果进行合理分析。方法选取听力言语正常的18～27岁青年受试者32例，以具有等价性的9张普通话双音节词表对其中5名受试者进行预实验，得出正式测试的给声强度，再对27名受试者进行正式测试，得出受试者的言语识别率。7～28天后，再以同样的测试条件进行复测比较各表两次测试言语识别率得分差值。结果复测和初测的言语识别率分别为52．40％±13．82％和50．74％±13．51％，各表复测和初测结果的差异无统计学意义（P〉O．05），复测和初测差值的标准差为12．0％，转化成95％置信度下的临界差值（critical difference，CD）为Z3．6％。结论普通话双音节词表具有较好的复测信度，可初步应用于临床测试和试验研究。以该表进行言语识别率测试时，两次测试差值超过23．6％方可认为两次结果有差异。

英文摘要：

Objective To evaluate the test-retest reliability of 9 equivalent Mandarin disyllabic lists, based on that the critical difference an appropriate conclusion can be drawn in speech assessment. Methods Percentage of correct word recognition was measured for each list by testing 32 Mandarin-speaking people aged from 18 to 27 with normal aural/oral eommunicational abilities. 5 of them joined the pilot study aimed to identify a presentation level that would be used in the formal test. The other 27 subjects participated in the formal two speech perception tests approximately 7 to 28 days apart. Analysis of the test-retest results were administered by SPSS 17.0. Results The 9 lists were found to have aueptable test-retest reliability respectively. The average speech recognition scores for two sessions were 52.40%± 13.82% and 50.74%±13.51%, paired test showed no significant difference （P〉0.05）. The standard difference （SD） of test-retest differences was 12.0% and the critical difference of the test score was 23.6% （in 95% confidence level）. Conclusion The Mandarin disyllabic lists were approved to have good test-retest reliability. The critical difference of 23.6 % is necessary to appropriately assessthe speech recogni- tion ability of Mandarin disyllabic words in clinic and research.