Table 6

Mechanisms through Which Repeater Peers Influence Nonrepeaters

Descriptiive StatisticsRegression Coefficients
RepeatersNonrepeatersSchool FE with Individual ControlsSchool FE with Individual and Homeroom Teacher ControlsSchool FE with Individual, Homeroom Teacher, and Classroom ControlsN
(1)(2)(3)(4)(5)(6)
Positive student–teacher interaction

2.560

(0.633)

2.745

(0.621)

–0.002

(0.003)

–0.000

(0.003)

–0.001

(0.002)

7,038

Positive student–student interaction

2.996

(0.775)

3.246

(0.734)

–0.010*

(0.005)

–0.004

(0.004)

–0.002

(0.004)

7,016

After-school study time (hours)

1.238

(1.200)

1.334

(1.016)

–0.010***

(0.003)

–0.008*

(0.003)

–0.010*

(0.004)

6,498

Having friends with general disciplinary problems

0.334

(0.472)

0.193

(0.395)

0.001

(0.002)

0.002

(0.002)

0.002

(0.002)

6,943

Having friends who go to internet cafés regularly

0.212

(0.409)

0.087

(0.282)

0.001

(0.002)

0.003+

(0.001)

0.002

(0.001)

6,929

Observations1,3917,061
School fixed effectsYesYesYes
Student characteristicsYesYesYes
Homeroom teacher characteristicsNoYesYes
Classroom characteristicsNoNoYes
  • Notes: Columns 1 and 2 present the raw mean values for repeaters and nonrepeaters with standard deviations in parentheses. Columns 3–5 present estimates based on regressions with different specifications, in which we progressively add student characteristics, homeroom teacher characteristics, and classroom characteristics to the school fixed effects models. Outcome variables in Column 3–5 are standardized among nonrepeaters within each school. Standard errors in parentheses. +p< 0.10, *p<0.05, **p<0.01, ***p<0.001.