Objective:To explore the risk factors of undergraduate depression and provide the basis for developing psychological warning mechanism and education.Methods:Two thousand five hundred and ninety-three undergraduates from six universities were randomly selected and required to fill out an uniform general questionnaire(including some basic information such as age,grade,family harmony degree,profession satisfaction,etc.) and the standardized Self-rating Depression Scale.The risk factors were analysed by binary logistic regression procedure.Results:The depression symptoms of 27.70% undergraduates were evident,which included 19.21% mild,7.52% moderate and 1.00% severe depressive symptoms.The mean depression severity was 0.48±0.10.There were a certain relation between gender,family financial situation,family harmony degree,affectional status,profession satisfaction,physical exercise,social activity and interpersonal relationship and depressive symptoms(P<0.05 to P<0.01).The grade,only-child,single-parent family and stay-at-home children were not significantly associated with depression(P>0.05).Conclusions:A variety of factors are involved in the occurrence of undergraduate depression.Paying attention to the risk factors,establishing effective psychological screening and early warning mechanisms and timely tracking and intervention is the foundation of preventing the mental health deteriorating of undergraduate.

Correlation and stepwise selection regression analysis were used to investigate the relationships between sensory quality and precipitation in tobacco growing areas of Shandong Province. The two approaches produced very similar result. The difference of stepwise selection regression equations for aroma quality, after taste, offensive taste, irritancy, combustibility and total score with precipitation were significant at 1% level. There was significance difference in stepwise selection regression equations for impact and ash with precipitation. No variable met the 0.05 significance level for entry into the stepwise selection regression equations for taste concentration and volume of aroma with precipitation. The total score of sensory quality were positively related to precipitation of May and August, whereas the type of aroma and combustibility were negatively related. There were positive correlations between impact of tobacco leaves and precipitation of June. The quality of aroma were negatively related to precipitation of May, yet positively related to June. The after taste and score of offensive taste were positively related to precipitation of May and total precipitation from May to August. There were positive correlations between the score of irritancy and total precipitation from May to August.