Kim-Godwin Yeoun Soo, Maume Michael O, Fox Jane A
J Immigr Minor Health. 2014 Dec;16(6):1217-24. doi: 10.1007/s10903-014-0007-x.
The purpose of the study is to identify the predictors of depression and intimate partner violence (IPV) among Latinos in rural Southeastern North Carolina. A sample of 291 migrant and seasonal farmworkers was interviewed to complete the demographic questionnaire, HITS (intimate violence tendency), Migrant Farmworker Stress Inventory, Center for Epidemiologic Studies Depression Scale (depression), and CAGE/4M (alcohol abuse). OLS regression and structural equation modeling were used to test the hypothesized relations between predictors of IPV and depression. The findings indicated that respondents reporting higher levels of stress also reported higher levels of IPV and depression. The goodness-of-fit statistics for the overall model again indicated a moderate fit of the model to the data (χ2 = 5,612, p < .001; root mean square error for approximation = 0.09; adjusted goodness-of-fit index = 0.44; comparative fit index = 0.52). Although the findings were not robust to estimation in the structural equation models, the OLS regression models indicated direct associations between IPV and depression.
本研究的目的是确定北卡罗来纳州东南部农村地区拉丁裔人群中抑郁症和亲密伴侣暴力（IPV）的预测因素。对291名流动和季节性农场工人进行了访谈，以完成人口统计学问卷、HITS（亲密暴力倾向）、流动农场工人压力量表、流行病学研究中心抑郁量表（抑郁症）和CAGE/4M（酒精滥用）。采用普通最小二乘法回归和结构方程模型来检验IPV预测因素与抑郁症之间的假设关系。研究结果表明，报告压力水平较高的受访者也报告了较高水平的IPV和抑郁症。总体模型的拟合优度统计再次表明该模型与数据拟合程度适中（χ2 = 5,612，p <.001；近似均方根误差 = 0.09；调整后拟合优度指数 = 0.44；比较拟合指数 = 0.52）。尽管研究结果在结构方程模型中的估计并不稳健，但普通最小二乘法回归模型表明IPV与抑郁症之间存在直接关联。
