目的：改良技擊反應動作時間目標量測系統及評估其信效度。方法：採用LabVIEW8.0軟體，結合電子硬體，以眼-手水平視覺反應動作時間測量，考驗儀器信效度。信度評估，37名男性大專生連續2天在相同時段完成測量，以內在等級相關係數(Intraclass Correlation Coefficient, ICC)表示再測信度。效度評估，以程度不同之2個族群(24名優秀組及24名一般組空手道男性大專選手)，在1天完成測量，以Independent t-test驗證改良版儀器之已知團體差異的建構效度。結果：信度方面，(慣用手)ICC為0.86，95%信賴區間為0.73~0.93；(非慣用手)ICC為0.75，95%信賴區間為0.51~0.87。效度方面，優秀組(慣用手為378.5±21.3毫秒、非慣用手為387.9±23.6 毫秒)優於一般組(慣用手為439.2±35.8毫秒、非慣用手為457.1±50.9毫秒)，均達統計顯著意義(p < .000)。結論：改良版之技擊反應動作時間目標量測系統可測量技擊選手之眼手水平視覺的反應動作時間，並具有良好之再測信度及已知團體差異的建構效度。實務可用來測量技擊選手的攻擊表現，和訓練技擊選手以提升其眼手反應動作時間。

Purpose: A major factor in fighting arts is the speed of response to a given attack. To determine the reliability and validity of target measurement system of martial attack response time. Methods: A national instruments (NI) data acquisition system Hardware and LabView 8.0 software has been used, the reaction time at eye-hand visual level while attack and defense were selected for trial test in terms of the reliability and validity of the instrument. For reliability, 37 male undergraduate students were recruited and measurements were finished within 2 days. Intraclass Correlation Coefficient (ICCs) was made to summarize the test-retest reliability. For validity, two known groups which expect to have obvious difference in reaction time (male karate contestants and male ordinal undergraduate students, 24 each) were recruited and completed the measurement within one day. Independent t-test was used to compare the response time between the two groups. Results: For reliability, the ICC for dominant hand was 0.86, 95%, CI = 0.73~0.93, p = 0.000; the ICC for non-dominant hand was 0.75, 95%, CI = 0.51~0.87, P = 0.000. For validity, karate contestants had a significantly shorter response time (dominant hand: 378.5±21.3 ms, non-dominant hand: 387.9±23.6 ms) than the ordinal undergraduate students (dominant hand: 439.2±35.8 ms, non-dominant hand: 457.1±50.9 ms) did (p = .000). Conclusions: The modified target measurement system can measure the response time at eye-hand visual level while attack and defense. The system had a good test-retest reliability and known-group difference of construct validity. Our target measurement system can measure the attack response time of martial art contestants, and train contestants to improve their eye-hand response time.