29 мая на факультете социологии пройдет международный семинар на тему: "How Student Ties Make Peer Effects Work". Авторы проекта: Олег Полдин, Диляра Валеева и Мария Юдкевич.
"How Student Ties Make Peer Effects Work"
Estimating peer effects fundamentally requires a correct identification of an actor's relevant peers. In this study, we explore how peer characteristics and achievements university influence university students' academic performance through social networks. The analysis uses data from two directed networks: one of friends and another of study partners among third-year students at a top-tier Russian university. We also use data on network ties from randomly formed student groups to address both endogeneity and to disentangle the influence of peers’ achievements from the effect of these peers' background characteristics. We show that both peer grade point average (GPA) and ability measures are significant in our models. A one-point increase in average peer GPA yields an increase in an individual student’s own GPA by approximately one-fourth of a point. Our models indicate that the peer effects we observe have slightly greater explanatory power among the study partner network than the friendship network. In the model that assumes interactions occur between groupmates, we find no peer effect stemming from students' classmates.
Начало в 18:00. Ауд. 317.
Рабочий язык семинара – английский. Приглашаются все желающие.
Estimating peer effects fundamentally requires a correct identification of an actor's relevant peers. In this study, we explore how peer characteristics and achievements university influence university students' academic performance through social networks. The analysis uses data from two directed networks: one of friends and another of study partners among third-year students at a top-tier Russian university. We also use data on network ties from randomly formed student groups to address both endogeneity and to disentangle the influence of peers’ achievements from the effect of these peers' background characteristics. We show that both peer grade point average (GPA) and ability measures are significant in our models. A one-point increase in average peer GPA yields an increase in an individual student’s own GPA by approximately one-fourth of a point. Our models indicate that the peer effects we observe have slightly greater explanatory power among the study partner network than the friendship network. In the model that assumes interactions occur between groupmates, we find no peer effect stemming from students' classmates.
Начало в 18:00. Ауд. 317.
Рабочий язык семинара – английский. Приглашаются все желающие.
Дата
27 мая
2013
В статье упомянуты