Table 3. Accounting for increases in gaps in college attendance and graduation between the top and bottom income quintiles with gap changes in income and demographic measures between 1968-1973 and 1994-1999.

	College Attendance Gap (total increase is 15.3 percentage points)				College Graduation Gap (total increase is 24.9 percentage points)
	Marginal Effects and Standard Errors From Logistic Regression	Change in Gap Last–First 6 Years	Amount of Schooling Gap Accounted for	% of Gap Accounted for	Marginal Effects and Standard Errors From Logistic Regression	Change in Gap Last–First 6 Years	Amount of Schooling Gap Accounted for	% of Gap Accounted for
ln Parent Income	0.148**(0.015)	0.50	0.07	48.5	0.107**(0.014)	0.50	0.05	21.6
Mother's Years of Education	0.053**(0.004)	−0.14	−0.01	−4.9	0.040**(0.004)	−0.14	−0.01	2.3
Number of Siblings	−0.024**(0.006)	0.75	−0.02	−12.0	−0.024**(0.005)	0.75	−0.02	−7.2
Two-Parent Family	0.038†(0.022)	0.14	0.01	3.6	0.086**(0.022)	0.14	0.01	4.9
Mother Age at Child's Birth	0.009**(0.002)	4.45	0.04	25.7	0.007**(0.002)	4.45	0.03	11.0

Notes: The two logistic regressions include controls for race/ethnicity, sex, firstborn status, and age of mother at first birth. Regressions are weighted using the PSID attrition-adjusted weight. Family-cluster-adjusted standard errors are shown in parentheses.

^†p < .10;

^*p < .05;

^**p < .01