透過您的圖書館登入
IP:34.230.84.106
  • 學位論文

台灣中老年人Goldberg’s IPIP五大因素人格量表簡式版本的發展

The Developmental of Short Form of Goldberg’s International Personality Item Pool 50-items among Middle-Aged and Older People in Taiwan

指導教授 : 李仁豪

摘要


研究旨在根據Goldberg所發展之IPIP五大因素人格量表進行簡式版本的發展,並加以驗證以適用於我國使用之版本。研究屬於橫斷面問卷調查,研究樣本為臺灣地區(含澎湖)的中老年人,第一次回收之有效樣本共455人,間隔一年所收集之第二次有效樣本為534人。研究工具包含「Goldberg’s IPIP 五大因素人格量表-五十題版本」(IPIP-50,中文版取自Zheng等人,2008)、「五大因素人格問卷」(BFI-44,本研究翻譯自John與Srivastava, 1999)、「心理幸福感量表」(PWB-18,中文版取自Li, 2014)、以及「自尊量表」(SES,Rosenberg, 1965;中文版取自楊中芳,1997)。第一次樣本收集IPIP-50、SES、及基本資料中的性別、年齡、教育程度、職業,第二次樣本收集簡版IPIP五大因素人格量表、BFI-44、PWB-18、及基本資料中的年齡變項。   第一次回收樣本使用統計軟體隨機分組為223人和232人,使用223人作為預試樣本以進行IPIP五大因素人格量表簡式版本的選題,並使用另一樣本(232人)與總樣本(455人)進行驗證性因素分析,以及以總樣本進行IPIP五大因素人格量表簡式版本和SES效標關聯效度的檢驗。且將第二次所收集的不重複樣本(279人)進行IPIP五大因素人格量表簡式版本與BFI-44、IPIP-50的聚合效度和心理幸福感效標關聯效度的檢驗。   在發展IPIP五大因素人格量表簡式版本的部分,先進行探索性因素分析,採用主軸因素萃取法限制萃取出五因素,並以直接斜交方式轉軸,再根據三個標準刪除不適合的試題:1. 試題在所跨因素間的負荷量差距小於.25者; 2. 每一因素應至少包含三題題項;3. 因素負荷量低於.50者。根據探索性因素分析刪題所剩的27題因跨因素的情形仍十分嚴重,接著使用驗證性因素分析進行選題,選題策略如下:1.利用MI(modification index)指令獲得原先預設為0的參數若放寬限制後,估計模型適配可減少的卡方值,包括試題跨因素可以減少的適配卡方值,以及試題的測量誤差兩兩相關時可以減少的適配卡方值;2.試題在其設定因素上的負荷量高低;3.質性考量,包含試題表達的清晰性、試題對特定母群的適當性、與其它試題內容的語意重疊性、知覺到的試題侵入性以及試題的表面效度等。結果共選出15題。   資料以描述性統計、t檢定、變異數分析、皮爾森積差相關與結構方程模式進行統計分析及檢定。研究結果整理如下: (一) 簡版IPIP五大因素人格量表共15題,每一因子包含3題,以總樣本進行驗證性因素分析的結果顯示卡方值 = 256.16、自由度 = 80、GFI = 0.93、AGFI = 0.90、CFI = 0.93、RMSEA = 0.07、SRMR = 0.04。其它的心理計量特性為:Cronbach’s α內部一致性考驗介於.58-.79;組合信度係數介於 0.59-0.80;潛在變異平均萃取量介於0.33-0.57;因素負荷量介於0.45-0.84;區辨效度亦顯示兩因素間相關之平方皆低於這兩個因素的潛在變異平均萃取量。而與IPIP-50各向度的聚合效度分別為外向性r = .864**、友善性r = .823**、審慎性r = .854**、情緒穩定性r = .901**、智性/想像力r = .876**;與五大因素人格問卷各向度的聚合效度外向性r = .669**、友善性r = .569**、審慎性r = .538**、情緒穩定性r = -.688**、智性/想像力r = .670**。 (二) PWB-18與簡版IPIP五大因素人格量表中的友善性、審慎性、及智性/想像力呈顯著中度相關,而與外向性和情緒穩定性呈顯著低相關。 (三) SES與簡版IPIP五大因素人格量表中的情緒穩定性呈顯著中度相關,而與外向性、友善性、智性/想像力呈顯著低相關。 (四) 不同背景變項之中老年人在「簡版IPIP五大因素人格量表」的差異: (1) 性別:男性在外向性、情緒穩定性、智性/想像力三個人格特質的分數較女性高,而女性則在友善性和審慎性的分數較男性高。 (2) 教育程度:基本上在友善性、審慎性、情緒穩定性、智性/想像力面向上,國中/初中以下學歷者會明顯低於高中/高職/專科肄業以上的學歷者。 (3) 職業:基本上在友善性、審慎性、智性/想像力面向上,工廠工人、傭工、女傭等職業明顯低於中小學校長、中小學教師、社工等職業。 (4) 年齡:老年人在五個人格特質的分數皆較中年人低,而在友善性、審慎性、智性/想像力部分達到顯著性的差異,其他向度如外向性和情緒穩定性則不因年齡而有所差異。 最後,提出研究限制和未來的研究方向。

並列摘要


The purpose of the research was to develop a short form of Goldberg’s International Personality Item Pool 50-items among middle-aged and older people in Taiwan. This was a cross-sectional questionnaire survey study in which two sample were recruited with one year interval. The first sample included 455 participants, and the other 534 participants, which were both from Taiwan and Penghu county. The instruments applied in the first sample were Goldberg’s International Personality Item Pool 50-items (IPIP-50), and Self-Esteem Scale (SES), as well as demographic information consisting of gender, age, education degree and occupation status, and in the second sample were administered the short form of IPIP-50, Big Five Inventory (BFI-44), Psychological Well-Being Scale (PWB-18), and age data. The first sample was randomly assigned to two groups with 223 and 232 participants respectively by statistical package for the social science software (SPSS 17.0). The 223 participants were used to develop the short form of IPIP-50 first with exploratory factor analysis (EFA) and the confirmatory factor analysis (CFA) with structural equation modelling (SEM) methodology. The other 232 participants and total sample (455 participants) were used to confirm the selected items again with CFA. Besides, the total sample (455 participants) was also used to analyze criterion-related validity of the short form of IPIP-50 with SES. Furthermore, in regard to the two samples (455 and 534 participants), non-overlaying 279 participants come from the sample of 534 participants were used to analyze the convergent validity of the short form of IPIP-50 with BFI-44 and IPIP-50 respectively, and to analyze criterion-related validity of the short form of IPIP-50 with PWB. To develop the short form of IPIP-50, EFA was set up to extract five factors using principal axis factoring extraction with direct oblimin rotating, and then inadequate items were considered deleting according to three criterions: 1. The item loaded on two factors with a loading difference less than 0.25; 2. Every factor should include at least three items; 3. Factor loadings should not be less than 0.50. Given that items with loadings crossing factors were still severe in the remaining 27 items, CFA was used to go on item selection. The CFA selection strategies were listed below: 1. The items were deleted when modification (MI) indicated higher chi-square values for the item loading crossing factors and for correlation between item measurement errors; 2. Based on a size of factor loading; 3. Qualitative consideration of item content. Finally, fifteen items were selected as the short form of IPIP-50. The collected data were analyzed by descriptive statistics, t-test, one-way ANOVA, Pearson’s product-moment correlation and SEM. The results were as follows: (1) The short form of Goldberg’s IPIP-50 had fifteen items. Each factor consisted of three items. The results of confirmatory factor analysis indicated that χ2 = 256.16, df = 80, GFI = 0.93, AGFI = 0.90, CFI = 0.93, RMSEA = 0.07, SRMR = 0.04. The factor loadings ranged 0.45-0.84, Cronbach’s α ranged .58-.79, composite reliability ranged 0.59-0.80, average variance extracted ranged 0.33-0.57, and discriminant validity also met acceptable standards. The results of convergent validity between the short form of IPIP-50 and IPIP-50 indicated that Extraversion: r = .864**, Agreeableness: r = .823**, Conscientiousness: r = .854**, Emotional stability: r = .901**, Intellect/Imaginary: r = .876**; The results of convergent validity between the short form of IPIP-50 and BFI-44 indicated that Extraversion: r = .669**, Agreeableness: r = .569**, Conscientiousness: r = .538**, Emotional stability: r = -.668**, Intellect/Imaginary: r = .670**. (2) Agreeableness, Conscientiousness, and Intellect/Imaginary of the short form of Goldberg’s IPIP-50 have significant medium correlations with PWB, and Extraversion and Emotional Stability have significant low correlations with PWB. (3) Emotional Stability of the short form of Goldberg’s IPIP-50 had a significant medium correlation with SES, and Extraversion, Agreeableness, and Intellect/Imaginary had significant low correlations with SES. (4) Difference testing for the demographic variables of gender, age, education degree, and occupation status in personality of the short form of Goldberg’s IPIP-50: (a) Gender: men had higher of Extraversion, Emotional stability, and Intellect/Imaginary than did women, but women had higher Agreeableness and Conscientiousness than did men. (b) Education: basically, people with junior high school degree and below had significantly lower Agreeableness, Conscientiousness, Emotional stability, and Intellect/Imaginary than those of senior high school degree and above. (c) Occupation: basically, factory workers, laborers, and maid, etc. had significantly lower Agreeableness, Conscientiousness, and Intellect/Imaginary than did president of schools, teachers, and social workers, etc. (d) Age: older people had lower five personalities than did middle-aged people, but only in Agreeableness, Conscientiousness, and Intellect/Imaginary did they have significant difference. Finally, the researcher proposed some limitations and suggestions for future studies.

參考文獻


王秀美、李長燦(2011)。五大人格特質量表中文版之信效度研究。社會服務與
張春興(1989)。張氏心理學辭典。臺北市:東華。
林意雪(2013)。閱讀:多元文化教育的議題。教育部性別平等教育季刊,63,
Barrick, M. R., Mount, M. K., & Gupta, R. (2006). Meta‐analysis of the relationship
Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP

被引用紀錄


鄭傑仁(2016)。擇前審思對五大人格與幸福感關係的調節作用之探索〔碩士論文,中山醫學大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0003-2112201615181800

延伸閱讀