家庭系統測量的評估——以晚期青少年的家庭為例

The Evaluation of Family System Measurement for Late Adolescents and Families

研究目的：台灣地區深受工業化與都市化的影響，家庭結構與功能均面臨相當的衝擊。複雜多變的家庭互動，使得系統的研究觀點不斷地被討論。家庭系統的測量工具雖然在近二十幾年來已逐漸發展，但由於家庭系統的複雜性，使得測量工具在實際操作的層面卻未必能確切捕捉及描繪其真正內涵。研究目的在針對國內普遍延用的FACES家庭系統測量工具，系統地評估Olson等人（1979）的編製理念及其在測量之間的落差，更進而以國內青少年與其家人為例，評估此工具對國人家庭的適用性。研究方法：以量性設計為主，質性設計為輔的二種研究程序共同探討以美國社會所發展出的家庭系統量表如何測量本地的青少年與其家人對家庭系統的知覺。研究結果：先整理出家庭系統量表的常模及向度，再整理出家庭系統向度中的內涵。接著，探討家庭系統的知覺與父母親互動，及與背景因素間的關連。研究結論：對於本樣本與Olson等人之研究結果之差異提出文化觀點的思考。

Purpose: The research is to evaluate how family systems for late adolescents and their families were measured by FACES-Ⅱ (a very popularly used inventory by researchers and practitioners). Due to the drastic social change process, traditional family structures have been facing great challenges, which make family interactions more varied and hard to be measured.
Method: In order to evaluate how well the inventory measured, both methods of quantitative and qualitative procedures were designed. Based on the quantitative research design, 392 college students were surveyed, the FACES-Ⅱ and the Multi-Individuation Scale devised by the researcher as a validity indicator were administered. Based on the qualitative research design, 36 focus group members were recruited into 6 groups to collect qualitative reactions to FACES-Ⅱ.
Results: The research results were presented to answer four questions: (1) What were families reflected from the inventory. (2) What aspect of dimensions were found from the inventory. (3) What theoretical constructs were in dimensions. (4) How did adolescents' perception of family system correlate to their individuation process and their demographic backgrounds. Quantitative data were analyzed by Factor Analysis method to examine major dimensions and factor components within dimensions. And, qualitative data collected from focus groups were analyzed by open coding procedures, and measurement difficulties for some factor components such as ''family boundary'', ''coalitions'', ''family role sharing'', and ''family rules'' were presented. Finally, relations among FACES-Ⅱ, multi-individuation process and background variables were analyzed by correlation method.
Conclusions: In general, Taiwanese family types tend to be more unbalances than Olson, et al's (1979) sample, whereas the cohesion dimension was underestimated and the adjustment dimension was overestimated. In conclusion, differences between this research results and Olson, et al.'s (1979) are also discussed with culture perspectives further. At last, the findings will have implications for researchers and practitioners who work with families in Taiwan, and limitations of this research are also discussed.