Failing to notice? Uneven teachers’ attention to boys and girls in the classroom

Table 10 Correlation between gender bias and SIMCE scores, by socioeconomic status of the school

Variable	Reading		Math		Science
	(1)	(2)	(3)	(4)	(5)	(6)
	Boys	Girls	Boys	Girls	Boys	Girls
Low socioeconomic status
Teacher-initiated action
Coefficient	− 7.804	− 2.007	− 7.006	− 32.140**	− 4.079	2.088
Standard error	(5.263)	(5.795)	(9.875)	(12.709)	(4.520)	(4.766)
Normalized coeff.	− 0.153	− 0.040	− 0.155	− 0.678	− 0.087	− 0.046
R squared	0.262	0.297	0.302	0.409	0.284	0.308
Student-initiated action
Coefficient	− 7.322	− 6.735	4.521	− 8.381**	− 1.118	− 2.140
Standard error	(5.630)	(5.981)	(3.034)	(3.611)	(5.203)	(5.316)
Normalized coeff.	− 0.109	− 0.103	0.137	− 0.166	− 0.018	− 0.036
R squared	0.262	0.298	0.302	0.409	0.284	0.308
Observations	1035	972	497	476	1028	963
High socioeconomic status
Teacher-initiated action
Coefficient	5.981	− 8.382			14.671***	− 6.990
Standard error	(5.759)	(5.297)			(5.316)	(4.810)
Normalized coeff.	0.069	− 0.131			0.185	− 0.119
R squared	0.200	0.179			0.277	0.234
Student-initiated action
Coefficient	2.915	− 12.558**			16.293**	− 7.508
Standard error	(6.976)	(6.253)			(6.515)	(6.071)
Normalized coeff.	0.027	0.143			0.163	− 0.094
R squared	0.197	0.184			0.272	0.233
Observations	264	226			260	225

Source: Primary data and data from the Sistema de Medición de la Calidad de la Educación (SIMCE)
Note: Normalized coeff. reports the association between gender bias and SIMCE scores in units of a standard deviation of the test score of the subject for the corresponding gender. Robust standard errors are in parentheses. All specifications include covariates: the ratio of boys to girls in the classroom, characteristics of students (family income, school performance, household structure, and parents’ education), characteristics of the school (type of administration, income decile, and experience and tenure the school of principal), and characteristics of the classroom teacher (total experience, tenure in the school, and whether he or she has or is pursuing a graduate degree) and the Classroom Assessment Scoring System (CLASS) score. All specifications include school fixed effects. The association between student- and teacher-initiated actions and math scores cannot be estimated in the model with the full set of controls and fixed effects because of insufficient observations
Significance level: * = 10%, ** = 5%, *** = 1%