傳統研究指出測驗散布度(Intersubtest scatter)是腦功能异常的可能指標。本研究統整相關文獻,分析四種魏氏分測驗布度指標:「全距」、「剖面變异度」、「與平均數達顯著差异分測驗數目」及「全距除以平均數」,探討指標之再測信度及相關。研究發現「全距除以平均數」有相對較高的穩定性,但與其他指標關聯較低。「與平均數達顯著差异分測驗數目」之再測信度相對最低,與其他指標相關亦不高;而「剖面變异度」表現則居中。本研究同時呈現臺灣WISC-Ⅲ常模在各指標所展現的多種基本率決斷值。文中對研究應用性有所討論,并對後續研究發展提出建議。
It is believed for a long time that the intersubtest scatter might be an index for possible brain dysfunction. The major purpose of this study was to investigate the property of the four often referred Wechsler intersubtest indices:「Range」, 「Profile Variability Index (PVI)」, 「Selz and Reitans range divided by mean index」, and 「The number of subtests which deviate significantly from the mean」. Each of these indexes was reported and described in details. Study of the test-retest reliability shows that「Selz and Reitan’s range divided by mean index」 is the one with the greatest stability, while 「the number of subtests deviates significantly from the mean」 is the least stable one. While compared to 「range」 and 「variance」, these two indexes correlate relatively lower with others. Also reported were the base rate tables for each index, various cut scores were provided for proper clinical usage. Applications and future developments were also discussed.