汉语单音节测听表在北京听力正常人群中的复测信度评估  

Mandarin Monosyllable Short Lists:Test-retest Reliability Study on Normal-hearing Subjects Living in Beijing

作　　者：郗昕[1] 赵乌兰[2] 冀飞[1] 陈艾婷[1] 赵阳[1] 王永华[2] 韩东一[1] 

机构地区：[1]解放军总医院耳鼻咽喉-头颈外科,北京100853 [2]浙江中医药大学听力与言语科学学院

出　　处：《听力学及言语疾病杂志》2009年第2期95-99,共5页Journal of Audiology and Speech Pathology

基　　金：科技部十一五支撑计划(2007BAI18B12)

摘　　要：目的评价22张等价的单音节测听表在正常人群中的复测信度,以利于采用该言语测听材料时得出合理可靠的结论。方法招募北京22名18～30岁耳科正常人,在10 dB HLSpeech强度下,以拉丁方实验设计方案,依次进行22张单音节表的识别率测试。间隔6～35天,在完全相同的条件下重新进行第二轮次的识别率测试。结果两轮测试的识别率分别为74.0%±12.2%和72.4%±10.3%,经配对t检验显示P=0.000152。逐表汇总两轮测试得分的差值,统计该差值在22名受试者中的标准差为9.3%,转化成95%置信度下的临界差值(crit-ical difference)为18.3%。两轮测试得分的相关系数为0.684。结论正常人使用该套单音节测听表进行有关言语工程的评价研究时,同一受试者前后两次识别率之差只有高于18.3%时,方可认为处置有效。Objective To evaluate the test--retest reliability of a new set of 22 equivalent phonemic--balanced Mandarin monosyllable lists by the normal--hearing subjects living in Beijing. The test--retest reliability is a practical index before the efficient conclusion can be drawn appropriately. Methods Twenty--two oral/aural normal participants aged from 18-30 were recruited to assess the monosyllable perception scores at 10 dB HLSpeech for 22 lists following the Latin--Square design. Tests were administered twice with the same experimental procedure in the exactly same situation, with 6-35 days intervals for all participants. The differences in scores over the two test sessions among 22 lists were calculated for 22 participants. Results The average speech recognition scores for two sessions were 74.0%±12.2% and 72.4%±10.3% individually, paired t test showed significant difference （p= 0. 000152） between two sessions. The random measurement error calculated by the SD of intra--participant test--retest differences was 9.3%. The grand critical difference（CD） for test score improvement was delineated as SD ＊ 1.96 = 18.3 % （in 95% confidence level）. The test--retest correlation coefficient of 0.684 was also analyzed via a Pearson product--moment correlation. Conclusion Critical difference of 18.3% is necessary for researcher to appropriately understand the efficiency of the speech engineering whenever the new set of Mandarin monosyllable short lists was used as an assessment tool.

关 键 词：言语测听 复测信度 单音节 

分 类 号：R764.04[医药卫生—耳鼻咽喉科]