Life chances and class: Estimating inequality of opportunity for children and adolescents in South Africa

Figures & data

Figure 1: Estimated probability of reaching various benchmarks for children from different socio-economic background (age 10 to 14)

Note: HH, household.

Figure 2: HOI and D-Index for access to clean water for children (age 10 to 14)

Source: Author's own estimation based on NIDS 2008.

Figure 3: Education differences for emerging adults by socio-economic background (age 19 to 20)

Figure 4: Life satisfaction of emerging adults from different parental class (age 17 to 20)

Source: Author's own calculation based on NIDS 2008.

Figure 5: Likelihood of a young adult (21 to 25) reaching matric, obtaining some tertiary education and not finding employment

Figure 6: Respondent's future perspective (age 21 to 25)

Source: Author's own calculation based on NIDS 2008.

Figure 7: Median reservation wage for young adults (age 21 to 35)

Source: Author's own calculation based on NIDS 2010.

Table 1: School outcome regressions

	(1)	(2)
Variable	Age 10 to 14, Grade 7	Age 19 to 20, matric
Mother lives in the household	0.0233	0.0184
	(0.0261)	(0.0421)
Mother's years of education	–0.00386	–0.0291**
	(0.00811)	(0.0117)
Mother's years of education^2	0.00171***	0.00337***
	(0.000586)	(0.000938)
Father lives in the household	0.0604**	0.0561
	(0.0273)	(0.0471)
Father's years of education	0.000227	0.0109**
	(0.00275)	(0.00457)
Log of per-capita household income	0.0202	0.0385*
	(0.0128)	(0.0226)
Female household head	0.0115	0.0918**
	(0.0233)	(0.0425)
Number of children	–0.0159**	–0.0148
	(0.00634)	(0.00925)
Coloured	0.0877**	–0.150**
	(0.0414)	(0.0754)
Indian	0.148***	–0.314***
	(0.0505)	(0.110)
White	–0.0298	0.0217
	(0.0611)	(0.0930)
Female	0.180***	0.108***
	(0.0201)	(0.0365)
Tribal	0.0207	–0.0963
	(0.0425)	(0.0694)
Urban formal	0.0535	0.0860
	(0.0442)	(0.0758)
Urban informal	0.0510	0.0233
	(0.0589)	(0.0920)
Constant	0.320***	–0.0938
	(0.107)	(0.177)
Observations	3 305	1 157
R-squared	0.138	0.273

Note: Based on NIDS 2008. Not listed: dummy variables for provinces and if mother and father education observations were missing. Robust standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.

Table 2: Tertiary education regression

	(1)	(2)	(3)
Variable	Some tertiary education	Further education (except university)	University
Mother's years of education	–0.0158*	–0.00747	–0.0196***
	(0.00823)	(0.00774)	(0.00638)
Mother's years of education^2	0.00199***	0.00132*	0.00190***
	(0.000672)	(0.000674)	(0.000533)
Father's years of education	0.0111***	0.00896***	0.00358
	(0.00342)	(0.00312)	(0.00227)
Log of per-capita household income	–0.239***	–0.137*	–0.157***
	(0.0725)	(0.0833)	(0.0562)
Log of per-capita household income^2	0.0245***	0.0151**	0.0147***
	(0.00620)	(0.00726)	(0.00493)
Coloured	–0.0486	–0.0386	–0.0251
	(0.0407)	(0.0384)	(0.0280)
Indian	0.177	0.0541	–0.0479
	(0.144)	(0.174)	(0.0679)
White	–0.121*	–0.149**	0.0651
	(0.0700)	(0.0696)	(0.0618)
Female	0.0239	0.0125	0.0103
	(0.0201)	(0.0199)	(0.0127)
Tribal	0.0337	0.00713	–0.0163
	(0.0402)	(0.0480)	(0.0187)
Urban formal	0.103**	0.0302	0.0469*
	(0.0478)	(0.0560)	(0.0273)
Urban informal	0.124**	0.0635	0.0408
	(0.0536)	(0.0594)	(0.0267)
Constant	0.500**	0.246	0.429***
	(0.217)	(0.237)	(0.161)
Observations	2 315	2 248	2 121
R-squared	0.233	0.142	0.217

Note: Based on NIDS 2008. Not listed: province dummies and whether or not parents are living in household. Robust standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.

Figure 8: Correlation between mother's education/income quintile and reaching Grade 7 on time, derived from regression results (see Table 1)

Note: CI, confidence interval.

Figure 9: Correlation between mother's education/income quintile and reaching some tertiary education, derived from regression results (see Table 2)

Note: CI, confidence interval.

Table 3: Reservation wage regression

	(1)	(2)	(3)
Variable	All	Employed	Unemployed
Best age in years	0.0102**	0.0184***	0.00393
	(0.00399)	(0.00666)	(0.00547)
Years of education completed	–0.0311**	–0.00614	–0.0703***
	(0.0158)	(0.0167)	(0.0246)
Years of education completed^2	0.00644***	0.00462***	0.00879***
	(0.00102)	(0.00109)	(0.00162)
Mother's years of education	0.0127**	0.0210***	0.00572
	(0.00499)	(0.00797)	(0.00606)
Father's years of education	0.00306	–0.00706	0.0104*
	(0.00504)	(0.00822)	(0.00603)
Log of per-capita household income	0.121***	0.201***	0.0720**
	(0.0187)	(0.0312)	(0.0303)
Coloured	0.269***	0.250***	0.292***
	(0.0606)	(0.0727)	(0.1000)
Indian	0.150	0.235*	0.0753
	(0.0950)	(0.121)	(0.138)
White	0.360***	0.348***	0.272**
	(0.0861)	(0.105)	(0.133)
Urban formal	0.0876*	0.0626	0.148**
	(0.0528)	(0.0791)	(0.0650)
Urban informal	–0.0232	–0.0273	–0.0101
	(0.0610)	(0.0923)	(0.0793)
Tribal	0.0782	0.0996	0.0656
	(0.0498)	(0.0782)	(0.0634)
Female	–0.187***	–0.182***	–0.186***
	(0.0337)	(0.0523)	(0.0411)
Constant	6.172***	5.302***	6.784***
	(0.182)	(0.275)	(0.313)
Observations	5 323	1 984	3 345
R-squared	0.240	0.276	0.184

Note: Based on NIDS 2010. Outcome variable log (reservation wage). Not listed: dummies for mother and father education missing. Robust standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.

Figure A1: HOI for access to selected key basic goods for children age 10 to 14

Source: NIDS 2008.

Table A1: Logit school regression

	(1)	(2)
Variable	Age 10 to 14, Grade 7	Age 19 to 20, matric
Mother lives in the household	0.0289	0.0348
	(0.0258)	(0.0517)
Mother's years of education	–0.0220**	–0.0370**
	(0.00894)	(0.0164)
Mother's years of education2	0.00326***	0.00409***
	(0.000721)	(0.00126)
Father lives in the household	0.0571**	0.0607
	(0.0283)	(0.0619)
Father's years of education	0.000249	0.0124**
	(0.00271)	(0.00502)
Log of per-capita household income	0.0294**	0.0495*
	(0.0140)	(0.0292)
Household has female head	0.00623	0.116**
	(0.0235)	(0.0531)
Number of children in household	–0.0143**	–0.0225
	(0.00608)	(0.0138)
Coloured	0.113***	–0.152**
	(0.0396)	(0.0606)
Indian	0.243***	–0.251***
	(0.0234)	(0.0451)
White	0.0159	–0.0154
	(0.0992)	(0.108)
Female	0.190***	0.144***
	(0.0205)	(0.0446)
Tribal	0.0342	–0.102
	(0.0407)	(0.0879)
Urban formal	0.0573	0.138
	(0.0435)	(0.0995)
Urban informal	0.0514	0.0847
	(0.0515)	(0.130)
Constant	–0.0199***	–0.00554***
	(0.6314)	(0.5087)
Observations	3 305	1 157

Note: Not listed: dummies for mother and father education if missing and provinces. Robust standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1. Note that this is a logit model but marginal effects are reported.

Table A2: Panel regression

	Ordinary least squares
	(1)	(2)
Variable	Reach matric	Get into tertiary education
Mother's years of education	0.00712	0.00779**
	(0.00474)	(0.00376)
Father's years of education	0.00699	0.00404
	(0.00485)	(0.00366)
Log of per-capita household income	0.0435**	0.0732***
	(0.0201)	(0.0165)
Household has female head	0.0242	–0.0313
	(0.0433)	(0.0334)
Number of children in household	–0.0103	–0.00527
	(0.00946)	(0.00602)
Coloured	0.0495	–0.144**
	(0.0880)	(0.0689)
Indian	0.141**	0.0629
	(0.0717)	(0.236)
White	–0.174	–0.172
	(0.137)	(0.124)
Female	0.0273	–0.0292
	(0.0345)	(0.0277)
Tribal	–0.0853	0.000384
	(0.0789)	(0.0491)
Urban formal	0.00447	0.124**
	(0.0837)	(0.0591)
Urban informal	0.0762	0.146*
	(0.0999)	(0.0748)
Constant	0.0630	–0.512***
	(0.187)	(0.138)
Observations	1 467	1 467
R-squared	0.076	0.131

Note: Based on NIDS 2008 and 2010. Not listed: dummies for mother and father education if missing and province dummies. Robust standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.