Bruening C C, Wagner W G, Johnson J T
Department of Psychology, Morehead State University, USA.
J Pers Assess. 1997 Jun;68(3):665-77. doi: 10.1207/s15327752jpa6803_12.
Human figure drawings collected from a clinical sample of 20 sexually abused and 20 nonsexually abused girls were randomly assigned to 1 of 2 case descriptions: Actual, in which raters were told the girls' actual abuse status, or Pretend, in which raters were told that drawings were made by girls with the opposite abuse status. Using the Draw-A-Person: Screening Procedure for Emotional Disturbance (DAP:SPED) scoring system developed by Naglieri, McNeish, and Bardos (1991), three raters independently scored 44 randomly ordered protocols, 4 of which were commonly rated as checks for rater accuracy and observer drift. Results revealed no significant effect for girls' abuse status or the case description given to raters, thereby suggesting that the DAP:SPED is sufficiently objective to withstand the confounding influence of varying case descriptions.
从20名遭受性虐待和20名未遭受性虐待的女孩的临床样本中收集的人物画被随机分配到2种病例描述中的一种：实际情况，即评分者被告知女孩的实际受虐状况；或假装情况，即评分者被告知这些画是由具有相反受虐状况的女孩绘制的。使用纳吉里、麦克内什和巴尔多斯（1991年）开发的画人：情绪障碍筛查程序（DAP:SPED）评分系统，三名评分者独立对44个随机排序的记录进行评分，其中4个通常被视为评分者准确性和观察者偏差的检查。结果显示，女孩的受虐状况或给予评分者的病例描述没有显著影响，从而表明DAP:SPED足够客观，能够抵御不同病例描述的混杂影响。
