Does early tracking affect learning inequalities? Revisiting difference-in-differences modeling strategies with international assessments

Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 7 Tracking variable: number of years since tracking

	Inequality measure
	Family background inequalities		Overall inequalities
	F-gap		SD
	(1)	(2)	(3)	(4)
Constant	76.71***	82.05***	63.32***	63.60***
No years since tracking	4.00**	− 3.86	0.848	0.494
Inequality measure at t = 1	0.538***	0.473***	0.438***	0.434**
No years since tracking inequality measure at t = 1		0.089		0.004
N countries	24	24	24	24
R²	0.503	0.540	0.286	0.287

No years since tracking is defined as (15 − age of tracking) if tracking occurs up to age 15, is equal to −1 if tracking occurs after age 15 (not yet occurred). Under the heading SD we report results relative to the effects of tracking on the country standard deviation
Under the heading F-gap we report results relative to the effects of tracking on the difference between tertiary educated parents with the largest number of books and no tertiary educated parents with the lowest number of books: F-gap = [ln(500) × \(c_{1}\) + \(c_{2}\)] − ln(5) × \(c_{1}\), where \(c_{1}\) and \(c_{2}\) are the coefficients of the family background variables: ln(books) and tertiary education
Columns (1) and (3) refer to the model with no interaction; columns (2) and (4) to models with interaction between the tracking indicator and inequality at t = 1
*0.05 < p value < 0.10, **0.01 < p value < 0.05, ***p value < 0.01