中華心理學刊 民 82，35 卷，1 期，19-31
Chinese Journal of Psychology 1993, Vol.35, No.1, 19-31

張素凰（國立台灣大學心理學系）；柯永河（國立台灣大學心理學系）

摘要

本研究旨在探討國內沿用多年的台大注意力測驗之心理計量特性，根據其計分方式進行注意力測驗反應的成份分析，提出不同樣本的再測信度報告，並且依不同的診斷組別建立大樣本常摸，再據所得結果提出一套電腦化注意力測驗的構想。經由五個樣本，研究一顯示台大注意力測驗具有穩定的再測信度，可以進一步檢驗其臨床效度。研究二的樣本來自精神科病人：藉由將730位病人分成不同的診斷組群，分別針對A式、B式與C式注意力測驗的各反應指標（反應總時間、漏劃數、錯劃數與單位時間正確反應數）進行組間差異比較，本研究發現在大多數的反應指標上可得到組間顯著差異。本文對此結果提出可能的解釋，並據此提出經由將認知心理學領域與心理測驗領域結合而建立一套適合於電腦作業、計分並解釋的電腦化注意力測驗的構想。

TAI -TA ATTENTION TEST: ITS PSYCHOMETRIC PROPERTIES AND ITS COMPUTERIZATION

Sue-Hwang Chang（Department of Psychology, National Taiwan University）；Yung-Ho Ko（Department of Psychology, National Taiwan University）

Abstract

The purpose of the present study was to investigate the psychometric properties of the Tai-Ta Attention Test and to establish some norm reference data of the psychiatric patients with different diagnostic types regarding to the following components of the attentional response: total response time, number of omit response, number of error response, and number of correct response per second. Via five different samples, clinical or nonclinical, Study 1 showed that Tai- Ta Attention Test had high test- retest reliability and appeared to be a promising instrument in assessing attentional function. Study 2 consisted of 730 patients. By subgrouping the subjects according to different criteria, the results revealed that, even after IQ, sex, education and age variables had been partialled out, several of the attentional indexes still showed significant difference between different subgroups. The implications of the findings are discussed and a model of an automated attention test concerning the intention to combine systematically the fields of computer, cognitive psychology and psychometry and its implications for the future research possibilities are postulated.