苯丙胺戒断后症状评定量表的初步编制和信效度研究

发布时间：2019-02-12 11:47

【摘要】：目的：本研究旨在编制出一个适合中国人群的苯丙胺戒断后症状评定量表,并对其进行心理测量学检验。方法：在湖南省各戒毒所对符合入组条件的苯丙胺依赖者进行资料收集,通过查阅文献、开放性访谈并结合戒毒所专家临床经验初步确定苯丙胺戒断后症状评定量表的条目库,对量表进行小样本预试(n=20),确定量表初稿。量表正式施测后(n=200),通过项目分析和探索性因素分析,按相关标准删除不符合的条目,形成量表的初步理论框架,并对量表进行心理测量学检验,选用分半信度、Cronbach's Alpha系数和Kendall和谐系数检验量表的信度,选用内容效度、专家效度、结构效度和校标效度检验量表的效度。结果：(1)通过项目分析和筛选,确定量表由20个条目组成,探索性因素分析确定量表共包含4个因素,解释总变异的54.895%,4个因素分别命名为：F1：渴求、F2：情绪障碍、F3：躯体症状和F4：精神病性症状。(2)内部一致性分析显示：全量表和各因素分半信度系数分别为：0.883、0.807、0.829、0.796、0.718。总量表和各因子的内部一致性信度Cronbach's Alpha系数为0.846,0.807、0.811、0.793、0.759,量表他评部分Kendall W系数为0.682,P值为0.0310.05。(3)量表内部相关分析显示：各条目与量表总分之间的相关系数在0.380-0.641之间,各因素与总量表总分之间相关系数在0.488-0.762之间；各因素间相关系数为0.166-0.409,相关均具有显著性(p0.01)。(4)量表总分与CGI、HAMD、HAMA和PANSS得分均有显著的正相关(p0.01),相关系数在0.464-0.657之间,且各分量表与相似内容的量表相关度较高,精神病性症状和PANSS相关系数最高(r=0.547),情绪障碍因子与HAMD和HAMA得分相关系数最高,分别为0.625和0.466。(5)量表及其各因子与停用前一月冰毒使用天数和总使用量均有显著的正相关(p0.01),且使用量差异研究结果显示,受试者中高使用量组总量表和各因子得分均显著高于低使用组,差异均具有显著性(p0.01)。结论：(1)本研究初步完成了苯丙胺戒断后症状评定量表的编制工作,经过探索性因素分析初步构建出该量表的合理结构。(2)信度研究表明,该量表的内部一致性信度较好,达到了心理测量学的要求。(3)内容效度和专家效度分析表明,该量表具有较好的内容效度和专家效度。(4)量表的内部一致性和因素分析发现,该量表具有较好的结构效度。(5)量表与CGI、HAMD、HAMA和PANSS4个量表的相关结果表明,该量表具有较好的校标效度；量表与停用前一月冰毒使用天数和总使用量的相关结果表明,该量表具有较好的实证效度；苯丙胺高低使用组的差异分析也表明,该量表具有较高的实证效度。
[Abstract]:Objective: to develop a scale for assessing the symptoms of amphetamine withdrawal in Chinese population and to test it with psychometrics. Methods: the data of amphetamine-dependent patients who met the condition of admission were collected in the rehabilitation centers of Hunan Province. By consulting literature, opening interviews and combining with clinical experience of experts in rehabilitation centers, the item library of the assessment scale for symptoms after withdrawal of amphetamine was preliminarily determined. Small sample pre-test (nn-20) was performed to determine the first draft of the scale. After the formal implementation of the scale (nnm200), item analysis and exploratory factor analysis were used to delete the non-conforming items according to the relevant criteria, to form the preliminary theoretical framework of the scale, and to conduct psychometric tests on the scale, and to select the split-half reliability. Cronbach's Alpha coefficient and Kendall harmony coefficient were used to test the reliability of the scale, content validity, expert validity, structural validity and calibration validity were used to test the validity of the scale. Results: (1) by item analysis and screening, the scale was composed of 20 items, and exploratory factor analysis showed that the scale consisted of 4 factors, 54.895of which explained the total variation, and the four factors were named F1: craving, respectively. F2: emotional disorder, F3: somatic symptom and F4: psychotic symptom. (2) Internal consistency analysis showed that the whole scale and the factor split-half reliability coefficient were 0.8830.8070.8070.7960.796and 0.718 respectively. The Cronbach's Alpha coefficient of internal consistency reliability of total scale and each factor was 0.846 / 0.807 / 0.8110.793/ 0.759, and the Kendall W coefficient of the other part of the scale was 0.682and 0.682respectively. P value was 0.0310.05. (3) the correlation coefficient between each item and the total score was 0.380-0.641, and the correlation coefficient between each factor and the total score was 0.488-0.762; The correlation coefficient between each factor was 0.166-0.409. There was significant positive correlation between the total score of CGI,HAMD,HAMA and PANSS and the total score of CGI,HAMD,HAMA and PANSS (p0.01). (4), the correlation coefficient was 0.464-0.657, and the correlation coefficient was 0.464-0.657. The correlation between each subscale and similar scale was higher, the correlation coefficient between psychotic symptoms and PANSS was the highest (r = 0.547), and the correlation coefficient between emotional disorder factor and HAMD and HAMA scores was the highest. (5) the scale and its factors were positively correlated with the days of meth use and the total usage of methamphetamine in the month before cessation (p0.01). The scores of total amount and factors in the high usage group were significantly higher than those in the low use group (p0.01). Conclusion: (1) this study has preliminarily completed the development of the rating scale for symptoms after withdrawal of amphetamine, and has preliminarily constructed the reasonable structure of the scale by exploratory factor analysis. (2) the reliability study shows that, The internal consistency reliability of the scale is good, which meets the requirements of psychometrics. (3) the analysis of content validity and expert validity shows that, (4) the internal consistency and factor analysis of the scale showed that the scale had good structural validity. (5) the correlation between the scale and CGI,HAMD,HAMA and PANSS4 scales showed that the scale had good structural validity. The scale has good calibration validity. The results of correlation between the scale and the days of meth use and the total usage of methamphetamine before discontinuation showed that the scale had a good empirical validity, and the difference analysis of the high and low use groups of amphetamine also showed that the scale had a high empirical validity.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：R749.64

【参考文献】