The need for new methods to deal with big data is a common theme in most scientific fields including prevention science. Evaluations of intervention programs often have multiple outcome variables. These are often reported for multiple time points (e.g., pre- and post-intervention) where data are multilevel (e.g., students nested in schools). The task of analyzing such data calls for innovative statistical methods for addressing the new challenges due to the big data. In this talk, we present an overview of variable-oriented and person-oriented statistical methods for big data analytics. Then using data from a provincial-wide mental health prevention study in Manitoba, Canada, we illustrate how these two approaches provide us with different information that can be complementary. Data analyses with variable-oriented approach (multilevel linear regression model) provided us the overall PAX program effects for each outcome variable; the person-oriented approach (latent transition analysis) helps to explore the transition of multiple outcomes across multiple time points and how the intervention program affects this transition differently for students with different risk profiles. The implications of these results and use of the person-oriented statistical approaches for data-driven decision-making and knowledge discovery are discussed.

蒋德鹏教授于2002年获东南大学管理科学与工程学博士学位，先后赴美国霍普金斯大学、加拿大约克大学进行博士后研究。现为加拿大曼尼托巴大学公共卫生学系终身教授，从事生物统计学的教学和科研工作，兼任医学院统计咨询中心主任和曼尼托巴大学健康医疗创新中心统计组负责人。同时，他还是加拿大多伦多大学和约克大学的兼职教授和吉林大学公共卫生学院客座教授。蒋教授在生物统计领域有较深的造诣，对大数据建模、纵向数据分析和混合统计模型有着特别的兴趣和深入的研究。多年来蒋教授为来自许多学科的研究者提供统计咨询，指导硕士和博士研究生，积累了丰富的统计咨询经验。目前已发表论文六十余篇。蒋德鹏教授曾获得中国教育部“春晖计划”特邀专家，与南京大学、东南大学、云南大学、中国药科大学等国内多所高校有合作研究。