臺灣地區中文監獄化量表之編制與初步信效度檢驗
No Thumbnail Available
Date
2008-12-01
Authors
任全鈞
李思賢
Jen, CC.
& Lee, TSH
Journal Title
Journal ISSN
Volume Title
Publisher
中華民國犯罪學學會
Abstract
本研究旨在編制一份中文監獄化量表。依據相關文獻整理出監獄化量表需包含四個向度：對權威的態度、認同非行與反社會態度、出監後的期待與受刑人內規。本研究同時從台灣東部某監獄收取同意填寫問卷受訪者173名，依據資料進行初步信、效度的分析。本研究在建構效度研究結果，由於本研究係依文獻與理論建構命題，驗証式因素分析結果顯示4 個分量表的單一因子在對權威的態度、認同非行與反社會態度、出監後的期待、與受刑人內規的可解釋變異量分別爲30.11%、30.29%、35.76%、與26.08%。在刪除因素負荷值未達.40之項目以確認各因素內的題目數量，整體可解釋中文化監獄量表40.01%的變異量。在內部一致性信度上，刪除與整體關聯性低的題項後，總量表α值爲.89；同時效度上，監獄化與效標疏離相關(r=.20, p<.05)，但與無助感不相關(r=-.09, p>.05)，本研究結果顯示初步建立之監獄化量表有可接受之信度與效度。
The study was designed to develop a Chinese Prisonization Index that can be uniformly used in prisonization research and perhaps as a correctional decision-making tool. Development of the questionnaire, administering the questionnaire to 173 inmates of eastern Taiwan, retesting a portion of the initial sample at least two weeks later, obtaining inmate file data, and data analysis. These concluded as the following. The scale was construct according to theoretical meaningful patterns and related concept. A 62-item prisonization questionnaire was developed which utilized a summative scale model. Its construct validity, the resulting correlation matrix was subjected to a maximum likelihood analysis with varimax rotation. The analysis yielded four components. The four factors were the attitude to authority, affiliation and identification with staff, post-prison expectations and inmate codes. Only loadings >.40 were considered salient. The explanatory powers of each factor was 30.11%、30.29%、35.76%、26.08%. The total amount of variance accounted for by the integrated model is 40.01%. The scale was found to have high internal consistency, a reliability coefficient of. 89, test-retest reliability was high correlation(r=0.739, p<.05) and moderately high correlations with alien scale (r=.20, p<.05), but no correlations with helplessness scale (r=-.090, p>.05). Basically, the scale can serve as a reliable self-reported prisonization assessment.
The study was designed to develop a Chinese Prisonization Index that can be uniformly used in prisonization research and perhaps as a correctional decision-making tool. Development of the questionnaire, administering the questionnaire to 173 inmates of eastern Taiwan, retesting a portion of the initial sample at least two weeks later, obtaining inmate file data, and data analysis. These concluded as the following. The scale was construct according to theoretical meaningful patterns and related concept. A 62-item prisonization questionnaire was developed which utilized a summative scale model. Its construct validity, the resulting correlation matrix was subjected to a maximum likelihood analysis with varimax rotation. The analysis yielded four components. The four factors were the attitude to authority, affiliation and identification with staff, post-prison expectations and inmate codes. Only loadings >.40 were considered salient. The explanatory powers of each factor was 30.11%、30.29%、35.76%、26.08%. The total amount of variance accounted for by the integrated model is 40.01%. The scale was found to have high internal consistency, a reliability coefficient of. 89, test-retest reliability was high correlation(r=0.739, p<.05) and moderately high correlations with alien scale (r=.20, p<.05), but no correlations with helplessness scale (r=-.090, p>.05). Basically, the scale can serve as a reliable self-reported prisonization assessment.