本研究旨在發展一份具信度與效度的問卷，以測量老年人從事不同種類靜態行為的時間。80 位65 歲以上社區老年人填寫問卷兩次，間隔一週，並配戴三軸加速規。透過相依樣本t 檢定及斯皮爾曼相關分析一週再測信度。同時效度先經由斯皮爾曼相關，分析自陳式總靜態時間與加速規測量總靜態時間二者的相關。其次，使用Bland-Altman Plot 比較兩者的一致性。結果顯示：自陳式總靜態時間若未納入小睡片刻（0.64 小時／每天）時間，平均每天7.29（標準差 = 3.60）小時。有關一週再測信度，相依樣本t 檢定顯示：除閱讀時間外，自陳式總靜態時間與各類靜態時間在一週前測與後測上未有顯著差異。斯皮爾曼相關分析指出：以下題項為良好：小睡片刻、用電腦、工作或擔任志工與閱讀；以下題項為可接受：看電視、用餐及興趣嗜好；以下幾類則較差：聊天、開（乘）車，以及其他；總靜態時間則為良好。自陳式總靜態時間（含小睡片刻，共10 種行為）與加速規測量總靜態時間有顯著相關，顯示同時效度良好。根據Bland-Altman Plot，自陳式總靜態時間平均每天低估加速規測量之總靜態時間約0.59 小時，然而，差異並不顯著。總體而言，本研究顯示此問卷測量臺灣老年人靜態行為的信度與效度之初步證據。

This study aimed to develop a reliable and valid questionnaire for measuring the time spent in different types of sedentary behaviors among elderly in their later life. Eighty community-dwelling older adults aged 65 years or above were completed the questionnaire twice over a one-week period, and wore a tri-axial accelerometer during this period. The test-retest reliability over a one-week period was evaluated by paired-sample t-tests and Spearmen’s correlation. Concurrent validity was first examined using Spearman’s correlation by self-reported total sedentary time and accelerometer-assessed sedentary time. Then, the Bland-Altman Plot was utilized to compare the agreement between the self-reported measure and the objective measure. Based on nine types of sedentary behaviors, the mean self-reported total sedentary time was 7.29 ± 3.60 hours/day without napping time (0.64 hours/day). Regarding test-retest reliability, except reading, there were no significant differences in the self-reported total sedentary time and each type of sedentary time between pre- and post-test. Spearmen’s correlation analyses showed that test-retest reliability was satisfactory for taking a nap, computer/internet use, working or volunteering and reading; marginally acceptable for watching TV, eating and hobbies; and poor for chatting, driving/sitting in a car, and others. Total sedentary time (including napping time) on an average day (sum of ten activities) had adequate test-retest reliability. Self-reported total sedentary time was significantly correlated with accelerometer-derived sedentary time, revealing that the concurrent validity was adequate. The Bland-Altman Plot indicated self-reported total sedentary time on average underestimated accelerometer-derived sedentary time by -0.59 hours/day. However, the mean self-reported total sedentary time was not significantly different from mean objective sedentary time. Overall, this study demonstrated the preliminary evidence for the reliability and validity of questionnaire for assessing the sedentary behaviors among Taiwanese older adults.