【作者】 闫靓； 陈克安；

【Author】 YAN Liang CHEN Ke-an (Department of Environmental Engineering,College of Marine,Northwestern Polytechnical University,Xi’an Shaanxi 710072,China)

【机构】 西北工业大学航海学院环境工程系；

【摘要】 已往采用语义细分法进行声品质评价实验时,其结果统计分析缺乏有效性判据。为此,本文从评价实验设计入手,提出并分析了实验数据中存在的两类误判,即“可允许一致性误判”和“不可允许误判”,确定了无效数据的剔除原则。将两类误判分析法和无效数据剔除原则应用于烦恼度评价实验的原始数据分析中, 其有效性分布检验结果显示：处理数据较好地符合正态分布规律,验证了两类误判分析及无效数据剔除依据的合理性与可靠性。更多还原

【Abstract】 In statistical analysis of semantic differential data in sound quality evaluation,the traditional data processing approaches are devoid of sufficient validity criteria for assessment of judging errors.Based on the design of a laboratory-scale jury test with semantic differential method and an auxiliary“5-points,4-ranks”rating scale,two kinds of misjudging rates,namely“permissible consistency misjudgment”and“un-permissible misjudgment”are proposed.Then the exclusion criteria for invalid raw data are established.The proposed methods and criterion have been applied to the analysis of subjective data of the semantic differential data for unpleasantness evaluation of low-frequency pure tones.Good fitness between analyzed results and normal distribution is acquired, rationality and rightness of proposed criterion has been proved.更多还原