本研究的目的是透過Rasch 模式分析，驗證自編之工具性日常生活活動量表 用來測量精神分裂症患者之Rasch 模式適合度。本研究之樣本來自於四所精神科 機構，包含慢性、日間病房、以及社區復健中心之精神分裂症患者，共226 位。 其中男性129 位 (57.1%)，平均年齡43.8 歲（標準差= 9.21）。本研究採用自編之 工具性日常生活活動量表，共有13 項分量表43 題，該量表預試分析結果顯示具 有良好之信效度。研究資料以WINSTEPS 3.75.1 與SPSS 18.0 軟體進行統計分析。 根據WINSTEPS 的分析結果顯示，第1、2、3、6、7、8、10、12、13、15、39、 40、41、42、43 題，這15 題的適合度不夠理想，刪除這15 題後，共11 項分量表 28 題，成為正式版之工具性日常生活活動量表。正式版的所有題目內、外適合度 均方均小於1.50。工具性日常生活活動量表的Rasch 模式適合度良好，其中正式 版的模式適合度優於施測版。因此，為了臨床上達到快速準確的診斷結果，可以 選擇使用正式版之工具性日常生活活動量表。

The purpose of this study was to check the model fit of a self-developed assessment of instrumental activities of daily living (IADL) in schizophrenia with Rasch modeling. The subjects for the study were 226 patients with schizophrenia from four institutions, including two chronic wards, a day care facility, and a community rehabilitation center. There were 129 (57.1%) males and the mean age of the subjects was 43.8 years (SD=9.21). The self-developed IADL assessment used in the study had 43 items. The pilot study of the assessment tool showed good initial reliability and validity. Statistical analysis was performed in WINSTEPS 3.75.1 and SPSS 18.0. According to the results of WINSTEPS analysis, 15 of the items (1, 2, 3, 6, 7, 8, 10, 12, 13, 15, 39, 40, 41, 42, and 43) demonstrated unsatisfactory model fit. Without these 15 items, the measure becomes a formal IADL assessment (28 items). The InMnsq and OutMnsq of all items in the formal version were smaller than 1.50. The Rasch model fit of IADL assessment appears to be good, and the formal version is superior to the tested version. Thus, for a quick and precise measurement outcome, the formal IADL assessment could be used in clinical settings.