Table 2. Accounting for increases in the schooling gap between the top and bottom income quintiles with changes in the levels and effects of income and demographic measures.

	Levels Effects: Last Minus First Six Years in Period (total increase in schooling gap is 0.43 years)			Levels Effects (robustness check): Last Minus Middle Six Years in Period (total increase in schooling gap is 0.55 years)			Value Effects: Last Minus First Six Years in Period (total increase in schooling gap is 0.43 years)
	Change in Gap	Amount of Schooling Gap Accounted for	% of Changes in Attainment Gap Accounted for	Change in Gap	Amount of Gap Accounted for	% of Changes in Attainment Gap Accounted for	Change in Regression Coefficient (Table 1)	Average Gap in First Six Years	% of Changes in Attainment Gap Accounted for
Completed Schooling (years)	0.43			0.55
ln Parent Income	0.50**	0.31	73.1	0.20**	0.13	22.7	−0.041	1.76	−16.8
Mother's Years of Education	−0.14**	−0.03	−7.4	1.47**	0.35	63.9	0.056	4.15	54.0
Number of Siblings	0.75**	−0.08	−19.4	0.02*	0.00	0.0	0.091	−1.52	−32.2
Two-Parent Family	0.14**	0.04	9.5	0.17**	0.07	11.9	0.220	0.49	25.2
Mother Age at Child's Birth	4.45**	0.17	39.3	1.37**	0.05	8.5	0.037	−0.50	−4.30

Notes: “Last Minus First Six Years” gap changes are weighted by the “All Cohorts” regression results shown in the first column of Table 1. “Last Minus Middle Six Years” gap changes are weighted by the “Age 14 in 1982-1999” regression results shown in the third column of Table 1. The p levels of changes in coefficients are shown in Table S3 (Online Resource 1).

^*p < .05;

^**p < .01