Jones F W, Neuringer C, Patterson T W
J Pers Assess. 1976 Aug;40(4):419-21. doi: 10.1207/s15327752jpa4004_12.
Response inconsistency as an index for invalidating MMPI protocols was investigated by measuring the number of inconsistent responses given to 14 pairs of identical items by brain-damaged and nonbrain-damaged schizophrenic, alcoholic, neurotic and normal patients (n = 22 in each of the 8 groups). In addition the patient's number of inconsistent responses were correlated with the usual MMPI validity measures (?, L, F and K). Although all the response consistency scores of the groups were high it was found that the brain-damaged were more response inconsistent than the nonbrain-damaged patients (regardless of psychiatric diagnoses). The correlations between response inconsistency and the MMPI validity measures were significant but low. Because of the over-all high level of response consistency found for all the groups, it was concluded that MMPI protocol validity should be based on the validity scale measures, rather than upon response inconsistency.
通过测量脑损伤和非脑损伤的精神分裂症患者、酗酒者、神经症患者和正常患者（8组每组各22人）对14对相同项目给出的不一致反应数量，研究了反应不一致作为明尼苏达多相人格调查表（MMPI）无效指标的情况。此外，患者的不一致反应数量与常用的MMPI效度指标（?、L、F和K）相关。尽管所有组的反应一致性得分都很高，但发现脑损伤患者比非脑损伤患者反应更不一致（无论精神科诊断如何）。反应不一致与MMPI效度指标之间的相关性显著但较低。由于所有组的反应一致性总体水平较高，因此得出结论，MMPI协议的效度应基于效度量表指标，而非反应不一致。
