The Development of Mental Health Scale for Military

碩士 === 國防大學政治作戰學院 === 心理碩士班 === 102 === The purpose of the present study was to develop a scale which can predict the voluntary soldiers of Republic of China Armed Forces who have suicidal ideation. Based on the relevant literatures and expert advices, the researcher categorized this scale (with a t...

Full description

Bibliographic Details
Main Authors: Lin, Chen-Kuang, 林辰光
Other Authors: 邱發忠
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/66358725867296951413
Description
Summary:碩士 === 國防大學政治作戰學院 === 心理碩士班 === 102 === The purpose of the present study was to develop a scale which can predict the voluntary soldiers of Republic of China Armed Forces who have suicidal ideation. Based on the relevant literatures and expert advices, the researcher categorized this scale (with a total of 73 questions) into 7 subscales: "depression," "anxiety," "hopelessness," "social isolation," "impulsive," "adaptability," and "life stress." To overcome the limitations of classical test theory, we use item response theory models to analysis items. In this study, 1,371 voluntary soldiers were randomly selected form Republic of China Armed Forces, and 35 questions were selected to measure the "internal consistency reliability," "construct validity," and "criterion-related validity." The results indicated that the Cronbach's alphas of the seven subscales were between .71 and .92. Regarding the construct validity, except λ_27,6, λ_28,6, x^2/df, GFI, and AGFI did not meet the standard, the SRMR, RMSEA, NFI, NNFI, CFI, IFI, PGFI, PNFI, composite reliability and average variance extracted met the criteria. Regarding the criterion-related validity, we used four scales (ie., BAI, BSS, BDI-II, and BHS) with high reliability and validity as the indexes. The results indicated that the correlation between the four indexes and the sub-scales were between .32 and .83. In Study 2, 117 voluntary soldiers who were randomly selected form army and air-force and were categorized into the high-risk and the non-risk group based on the results of structured interviews. Receiver Operating Characteristic curve was applied to assesse the cut-off point of their scores. The results indicated that the area under the ROC curve was .86 with hig sensitivity. Next, we investigate the cut-off point by using total scores of 75.5, 76.5, 77.5, 78.5, 79.5, 82, and 84.5 for the sub-scales. Moreover, in terms of the discriminative validity, the total score of the Mental Health Scale of Military and the cores of the sub-scales of the high-risk group were significantly higher than those of the non-risk group.