大学生心理健康的潜在剖面分析

Latent profile analysis of mental health in college students

目的 考察大学生心理健康状况的潜在分类模式以及人口学变量在其中的作用,为有针对性地提升大学生心理健康水平提供理论依据。方法 采用中国大学生心理健康筛查量表(CSMHSS)对江西省某高校33 250名在校本科生进行心理测查,并对数据进行潜在剖面分析(LPA)和多元Logistic回归分析。结果 大学生心理健康状况可分为严重心理问题组(5.56%),一般心理问题组(13.90%),低自我伤害-高学业就业、网络成瘾组(18.96%),学业就业压力组(27.82%)和心理健康组(33.76%)5个潜在类别。多元Logistic回归分析显示,男生在各心理问题组所占比例更小(OR=0.630,0.980,0.515,0.673,P均<0.01);大一和大二学生在各心理问题组所占比例更大(大一学生:OR=4.896,1.459,5.334,3.024;大二学生:OR=2.303,1.353,2.297,1.461;P均<0.01),大三学生在严重心理问题组、低自我伤害-高学业就业、网络成瘾组、学业就业压力组所占比例更大(OR=1.423,1.540,1.297,P均<0.01);城市大学生在各心理问题组所占比例更小(OR=0.877,0.796,0.633,0.716,P均<0.05),城镇大学生在一般心理问题组、低自我伤害-高学业就业、网络成瘾组、学业就业压力组所占比例更小(OR=0.897,0.809,0.857,P均<0.05)。结论 大学生的心理健康状况存在异质性差异,性别、年级和生源地对各潜在类别均有影响。

Objective To explore the latent classifications of college students'mental health and the role of demographic variables on the heterogeneity of mental health in college students,and to provide a theoretical basis for the targeted promotion of college students'mental health.Methods A total of 33250 undergraduates from a university in Jiangxi were investigated with College Students Mental Health Screening Scale(CSMHSS).Latent Profile Analysis(LPA)was used to explore the classification model of college students'mental health.Logistic regression analysis was used to explore the effect of demographic variables on the heterogeneity of mental health in college students.Results College students'mental health was classified into five latent classifications:serious mental health problems group(5.56%),general mental health problems group(13.90%),low self harm-high academic and employment pressure,internet addiction group(18.96%),academic and employment pressure group(27.82%),and positive mental health group(33.76%).Logistic regression analysis showed that male students had a lower proportion in all groups(OR=0.630,0.980,0.515,0.673,P<0.01).Freshmen and sophomores had a higher proportion in all groups(freshmen:OR=4.896,1.459,5.334,3.024;sophomores:OR=2.303,1.353,2.297,1.461;P<0.01).Juniors had a higher proportion in serious mental health problems group,low self harm-high academic and employment pressure,internet addiction group,and academic and employment pressure group(OR-1.423,1.540,1.297,P<0.01).Urban students had a lower proportion in all groups(OR=0.877,0.796,0.633,0.716,P<0.05).Town students had a lower proportion in general mental health problems group,low self harm-high academic and employment pressure,internet addiction group,and academic and employment pressure group(OR=0.897,0.809,0.857,P<0.05).Conclusion There are heterogeneous differences in college students'mental health.Gender,grades,and source places of college student impact all the five latent classifications.