動態人樹房子電腦化計分系統的適用性與心理計量之研究–以台灣精神疾病患者為例

The Usefulness and Psychometric Study of the Kinetic-House-Tree-Person Computerized Scoring System for Persons with Mental Illness in Taiwan

李致瑩(Chih-Ying Li)

指導教授 ： 潘璦琬

國立臺灣大學/醫學院/職能治療研究所/碩士(2008年)

https://doi.org/10.6342/NTU.2008.03019

摘要

研究背景：

職能治療師在精神健康領域使用創造性藝術活動作為治療媒介，已有很長一段時間。根據2000年一項全國調查結果顯示，「動態人樹房子繪畫測驗」(The Kinetic-House-Tree-Person drawing test; KHTP)是台灣精神疾病職能治療領域最常使用的評估工具之一。由於繪畫不僅能幫助精神疾病患者統合內在與外在的世界，也有助於讓個案在一個安全和支持的環境下表達其情感和思想。因此，「動態人樹房子繪畫測驗」特別適合口語表達能力受限的精神疾病患者使用。
然而，「動態人樹房子繪畫測驗」雖然在臨床上有很高的使用率(77.8%)，但仍無足夠的文獻證據能支持「動態人樹房子繪畫測驗」之信效度與適用性。因此，本研究目的欲建立中文版電腦化「動態人樹房子繪畫測驗」之評分系統(KHTP scoring system-Chinese Computerized version; KHTP-CC)，並探討此評分系統之心理計量特質；以期提供「動態人樹房子繪畫測驗」於臨床使用之科學證據。

方法：

本研究共66位精神疾病患者(26位個案為精神分裂症患者、20位個案為憂鬱症患者、20位個案為其他精神疾病患者)與50位正常大學生參與。每位受測者均進行「動態人樹房子繪畫測驗」，此外，66位精神疾病患者需完成貝氏焦慮量表、貝氏憂鬱量表、貝氏無望感量表、職能自我評估問卷、自我掌控感量表、社會支持量表、意志量表、簡明版世界衛生組織生活品質問卷與質性訪談。其中，33位精神疾病患者在兩週內完成再測，並由2位評分者獨立進行圖畫之評分。評分系統共有54題，其中35題為焦慮與憂鬱繪畫特徵、28題為精神症狀繪畫特徵，共9題重複。這些題目都使用Visual Basic程式設計軟體建構至電腦中，以便讓評分者使用電腦進行圖畫之評分。

資料分析：

本研究使用SPSS(11.5版本)與Winsteps(3.56版本)進行信效度檢驗。KHTP-CC分數均經羅序分析轉換成等距分數再進行統計分析。以ICC值作為檢驗施測者間信度之指標，以Kappa值作為檢驗再測信度之依據。並以羅序分析探討KHTP-CC之建構效度。此外，使用斯皮爾曼相關係數分析以檢驗KHTP-CC之同時效度。並使用t-test和卡方檢定檢驗精神疾病患者組與大學生組在KHTP-CC分數和其他表現上之差異。最後，亦整理質性訪談結果。

結果：

在心理計量特質上，KHTP-CC具有良好的施測者間信度及可接受的再測信度。KHTP-CC的項目均符合羅序模式之單一建構效度。同時效度的結果較差，可能是由於用以比較之測驗的內容不同等原因所導致。此外，質性訪談發現，精神疾病患者少有繪畫經驗，然而是否學過繪畫，並不影響KHTP-CC之分數。喜歡繪畫的精神疾病患者，通常圖畫內容較為豐富。
在精神疾病患者組與大學生組的比較上，兩組的分數具有統計上顯著差異。兩組的年紀、教育程度與喜不喜歡繪畫、對繪畫的興趣、學過繪畫與否、覺得繪畫的功用及完成測驗所需的時間長短，具有統計上顯著差異。精神疾病患者組的圖畫相較於大學生組，內容較單調、細膩度低，且人、樹、房子間的互動與故事性較少、較少呈現歡樂氣氛、臉部表情淡漠或奇特、連接房子的道路往往不會與人相接。
在臨床適用性上，質性訪談結果發現，KHTP-CC讓精神疾病患者有機會表達自已，並且促進治療師與精神疾病患者之溝通。繪畫特徵中，房子的特徵所代表的意涵與繪畫者描述的特質之差異最大，可能與文化有關。因此，施測者對於繪畫特徵的詮釋，應考量受測者的文化背景之影響。完成KHTP-CC約需15分鐘，因此，相當適合臨床使用。

關鍵字

職能治療 ； 動態人樹房子繪畫測驗 ； 電腦化評分系統 ； 羅序測量 ； 信度 ； 效度

並列摘要

Background:
Occupational therapists have used creative art activities as a treatment medium in the mental health domain for a long time. According to a national survey conducted in 2000, the Kinetic-House-Tree-Person Drawing Test (KHTP) was one of the most frequently used evaluative tools in mental health occupational therapy in Taiwan. Drawing not only could help the person with mental illness to integrate the inner self, but express emotion under a safe and supportive environment. Therefore, the KHTP was especially useful for the persons with mental illness with limit of verbal expression.
Although KHTP is commonly used (77.8%), there were few studies to support the validity, reliability and usefulness of the KHTP. Therefore, the purpose of this study is to develop a KHTP Chinese computerized scoring system (KHTP-CC) and to examine its psychometric properties.
Methods:
Sixty-six persons with mental illness and 50 college students were recruited. Each one completed the KHTP, and the persons with mental illness also completed the Beck Anxiety Inventory (BAI), the Beck Depression Inventory (BDI-II), the Beck Hopelessness Scale (BHS), the Occupational Self Assessment (OSA), the mastery scale, the Inventory of Socially Supportive Behaviors (PSI) and the World Health Organization Quality of Life (WHOQOL-BREF). The Volitional Questionnaire (VQ) was assessed by the researchers. A qualitative interview was also conducted.
Thirty-three persons with mental illness completed the KHTP again within 2 weeks. Two raters independently scored the drawings. The KHTP-CC contains 54 items including 35 items of anxiety/depression characteristics and 28 items of psychotic characteristics (9 of them are in both subscales). These items were programmed into the computer using Visual Basic software to let the raters score with the computer.
Data Analysis:
SPSS (11.5 version) and Winsteps (3.56 version) were used. The ICC value was used to examine the inter-rater reliability. The Kappa value was used to examine the test-retest reliability. Each drawing score was transformed into Rasch measure for subsequent analysis. The unidimensionality and item difficulty was examined by using Rasch analysis. The Spearman correlation coefficient was used to examine the concurrent validity. The t test and chi-square was used to examine if there was statistical differences of the KHTP-CC scores between the group of persons with mental illness and college studnets. Finally, the results of the qualitative interviews were analyzed.
Results:
The KHTP-CC showed good inter-rater reliability and acceptable test-retest reliability. It was proved to encompass unidimensional construct. The concurrent validity was poor. This could be due to different content pertaining to each measure. Persons with mental illness seldom had drawing experiences compared with the college students. People who liked to draw usually demonstrated richer contents than those who don’t. It also showed that the experiences of drawing did not relate to the KHTP-CC score.
There was statistically significant difference in the KHTP-CC Rasch transformed score, age, educational level, likeness of drawing, interest about drawing, learning experience of drawing, perceived functions of drawing and time consumption between the group of persons with mental illness and college studnets. The KHTP drawn by persons with mental illness was monotonic, lack of interactions or story telling, unhappy atomosphere, poor facial expression, and the lack of connectedness of the road to the person.
The KHTP could be a useful tool to accelerate self-expression and communication between the persons with mental illness and the therapists. There might be cultural differences in symbolic meanings of house between western countries and Taiwan. It was important to notice the context of the persons with mental illness when interpret their drawings. Finally, it took about 15 minutes for the persons with mental illness to complete the KHTP and for the therapist to score. Therefore, the KHTP-CC was easy to use in the clinical practice.

並列關鍵字

occupational therapy ； Kinetic House-Tree-Person Drawing Test (KHTP) ； computerized scoring system ； Rasch measurement ； reliability ； validity

參考文獻

姚開屏 (2002). 台灣版世界衛生組織生活品質問卷手冊. Taipei, 台灣大學心理系.

蕭小菁、潘璦琬、鍾麗英、呂淑貞 (2000). "台灣精神科職能治療評估工具的現況調查." 職能治療學會雜誌 18: 19-32.

鄧珍珠、羅鈞令、黃曼聰、施杏如 (1994). "台灣與美國精神病患繪圖內容及色彩意義之比較." 職能治療學會雜誌 12: 41-49.

楊明山 (2007). Psychometric Study of Chinese Version of the Volitional Questionnaire for Clients with Depressive Disorders in Taiwan. School od Occupational Therapy. Taipei, National Taiwan Unversity. Master: 91.

Pan, A. W., Chen, T. J., & Li, C. Y. (2006). Validation of a computerized kinetic house tree person test. Taiwan Occupational Therapy Association Sixth Scholarship Workshop. No.17 Xu-Zhou Road, Taipei, 10020 Taiwan R.O.C.