The effects of system type and characteristics on skills inequalities during upper secondary education: a quasi-cohort analysis of OECD data

Figures & data

Table 1. Countries included in the sample.

Country	Country code	PISA wave	SAS round	System Type
Australia	AU	2006	2011/12	4
Austria	AT	2006	2011/12	3
Canada	CA	2006		2a
Canada (English)	CA-en		2011/12	2a
Canada (French)	CA-fr		2011/12	2a
Chile	CL	2009	2014/15	4
Czech Republic	CZ	2006	2011/12	1
Denmark	DK	2006	2011/12	1
England (UK)	UK-en	2006	2011/12	4
Estonia	EE	2006	2011/12	1
Finland	FI	2006	2011/12	1
Flanders (Belgium)	BE-fl	2006	2011/12	1
France	FR	2006	2011/12	1
Germany	DE	2006	2011/12	3
Greece	GR	2009	2014/15	1
Ireland	IE	2006	2011/12	4
Israel	IL	2009	2014/15	4
Italy	IT	2006	2011/12	1
Japan	JP	2006	2011/12	1
Korea	KR	2006	2011/12	1
Lithuania	LT	2009	2014/15	1
Netherlands	NL	2006	2011/12	1
New Zealand	NZ	2009	2014/15	4
Northern Ireland (UK)	UK-ni	2006	2011/12	4
Norway	NO	2006	2011/12	2b
Poland	PL	2006	2011/12	1
Russian Federation	RU	2006	2011/12	1
Singapore	SG	2009	2014/15	1
Slovak Republic	SK	2006	2011/12	1
Slovenia	SI	2009	2014/15	1
Spain	ES	2006	2011/12	4
Sweden	SE	2006	2011/12	2b
United States	US	2006	2011/12	2a

Figure 1. Gini coefficients of inequality for scores in reading/literacy and maths/numeracy by country/territory (PISA – 15-year-olds and SAS – 18-20-year-olds).

Figure 2. Gini coefficients of inequality for scores in reading/literacy and maths/numeracy by system type (PISA – 15-year-olds and SAS – 18-20-year-olds).

Figure 3. Main system characteristics – duration of upper secondary, curriculum standardisation, stage at first selection, vocational prevalence, vocational specificity and teaching hours per week – by country/territory and system type.

Table 2. OLS and DID estimates for the effect of system type on inequality in literacy/numeracy.

	Inequality in literacy (Gini coefficient)				Inequality in numeracy (Gini coefficient)
	OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2	OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2
System type (ref: Type 1)		.48		.26		.59		.31
Type 2a	1.70* (0.63)		1.87* (0.90)		1.64** (0.46)		2.00* (0.79)
Type 2b	0.01 (0.62)		−0.27 (0.98)		0.17 (0.54)		0.13 (0.95)
Type 3	0.34 (0.65)		−0.48 (0.98)		0.05 (0.56)		−0.49 (0.95)
Type 4	1.35* (0.38)		0.99+ (0.58)		1.33** (0.34)		0.87 (0.56)

Statistically significant at: *** p < .001; **p < .01; * p < .05; + p < .10 level

Table 3. OLS and DID estimates for the effect of system characteristics on inequality in literacy/numeracy.

		Inequality in literacy (Gini coefficient)				Inequality in numeracy (Gini coefficient)
		OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2	OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2
M1	Duration	−0.53* (0.25)	.21	−0.75* (0.33)	.08	−0.42+ (0.23)	.24	−0.60+ (0.34)	.05
M2	Standardised curriculum	−0.30* (0.11)	.26	−0.20 (0.16)	.10	−0.24* (0.11)	.18	−0.21 (0.15)	.08
M3	Stage at first selection (ref: entry to lower sec.)		.16		.08		.33		.10
	Entry to upper sec.	0.70 (0.47)		1.22* (0.59)		0.80 (0.39)		1.12+ (0.58)
	During upper sec.	0.67 (0.63)		1.08 (0.82)		1.32* (0.52)		1.65* (0.79)
M4	Vocational prevalence	−0.02 (0.01)	.16	−0.03+ (0.01)	.06	−0.02* (0.01)	.32	−0.03* (0.01)	.08
M5	Vocational specificity	−0.01 (0.01)	.10	−0.02 (0.02)	.04	−0.01 (0.01)	.30	−0.00 (0.02)	.01
M6	Teaching hours/week	0.17* (0.05)	.39	0.13+ (0.07)	.10	0.14** (0.04)	.47	0.12+ (0.06)	.17

Statistically significant at: *** p < .001; p < .01; * p < .05; + p < .10 level

Table A1 OLS and DID estimates for the effect of 12 system characteristics on inequality in literacy/numeracy.

		Inequality in literacy (Gini coefficient)				Inequality in numeracy (Gini coefficient)
		OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2	OLS estimate β1Y (S.E.)	R^2	DID estimates γ1Y.age1820 (S.E.)	R^2
M1	Theoretical duration	−0.53* (0.25)	.21	−0.75* (0.33)	.08	−0.42+ (0.23)	.24	−0.60+ (0.34)	.05
M2	Standardised curriculum	−0.30* (0.11)	.26	−0.20 (0.16)	.10	−0.24* (0.11)	.18	−0.21 (0.15)	.08
M3	Stage at first selection (ref: entry to lower sec.)		.16		.08		.33		.10
	Entry to upper sec.	0.70 (0.47)		1.22* (0.59)		0.80 (0.39)		1.12+ (0.58)
	During upper sec.	0.67 (0.63)		1.08 (0.82)		1.32* (0.52)		1.65* (0.79)
M4	Vocational prevalence	−0.02 (0.01)	.16	−0.03+ (0.01)	.06	−0.02* (0.01)	.32	−0.03* (0.01)	.08
M5	Vocational specificity	−0.01 (0.01)	.10	−0.02 (0.02)	.04	−0.01 (0.01)	.30	−0.00 (0.02)	.01
M6	Teaching hours/week	0.17* (0.05)	.39	0.13+ (0.07)	.10	0.14** (0.04)	.47	0.12+ (0.06)	.17
M7	Age at completion of upper secondary (ref: 17 or 18 years)		.19		.04		.24		.06
	19 years	−0.69 (0.37)		−0.63 (0.52)		−0.63+ (0.35)		−0.46 (0.52)
M8	Teachers’ starting salary (US$ thousand)	−0.00 (0.02)	.07	0.00 (0.03)	.01	−0.00 (0.02)	.18	0.01 (0.03)	.02
M9	Expenditure per student at upper secondary (% per capita GDP)	−0.05+ (0.03)	.17	−0.04 (0.04)	.07	−0.04 (0.03)	.21	−0.01 (0.04)	.11
M10	Number of school types	−0.01 (0.15)	.08	−0.25 (0.16)	.16	−0.04 (0.13)	.15	−0.22 (0.16)	.14
M11	Upper secondary completion rate (%)	−0.00 (0.03)	.03	−0.02 (0.03)	.02	−0.01 (0.03)	.13	−0.01 (0.04)	.02
M12	Cohort years of upper secondary (Duration of upper secondary*completion rate)	−0.00 (0.00)	.16	−0.01* (0.00)	.11	−0.00 (0.00)	.21	−0.01+ (0.00)	.06

Statistically significant at: *** p < .001; p < .01; * p < .05; + p < .10 level

Data available at: https://www.llakes.ac.uk/the-effects-of-system-type-and-characteristics-on-skills-inequalities-during-upper-secondary-education-a-quasi-cohort-analysis-of-oecd-data/

Table A2 Skills inequalities in PISA and SAS.

			PISA (age 15)				SAS (age 18–20)
Country	Country code	System Type	Reading (P90-P10)	Reading (Gini)	Maths (P90-P10)	Maths (Gini)	Literacy (P90-P10)	Literacy (Gini)	Numeracy (P90-P10)	Numeracy (Gini)
Australia^9	AU	4	256.9	.103	226.6	.096		.087		.102
Austria	AT	3	226.6	.124	256.9	.110	110.1	.081	121.2	.088
Canada	CA	2a	219.6	.102	219.6	.092	-	-	-	-
Canada (English)	CA-en	2a	-	-	-	-	125.3	.097	141.1	.118
Canada (French)	CA-fr	2a	-	-	-	-	125.1	.088	128.1	.091
Chile	CL	4	205.3	.104	205.3	.107	136.6	.109	153.8	.131
Czech Republic	CZ	1	267.3	.130	267.3	.115	102.3	.073	110.9	.080
Denmark	DK	1	217.4	.101	217.4	.093	116.3	.075	126.1	.087
England (UK)	UK-en	4	263.0	.116	232.3	.102	123.6	.089	138.0	.104
Estonia	EE	1	206.7	.095	206.7	.088	111.8	.077	113.9	.083
Finland	FI	1	207.9	.084	207.9	.083	123.5	.075	127.6	.087
Flanders (BE)	BE-fl	1	265.0	.112	254.3	.102	119.1	.080	127.8	.090
France	FR	1	248.8	.120	248.8	.110	123.9	.088	141.8	.105
Germany	DE	3	257.2	.125	257.2	.111	121.6	.087	133.1	.097
Greece	GR	1	228.4	.111	228.4	.108	117.5	.089	122.1	.093
Ireland	IE	4	237.3	.101	211.6	.093	115.7	.080	129.3	.097
Israel	IL	4	270.3	.133	270.3	.132	138.7	.111	158.1	.132
Italy	IT	1	242.5	.130	242.5	.117	113.7	.083	126.3	.098
Japan	JP	1	233.7	.115	233.7	.098	99.8	.063	110.1	.082
Korea	KR	1	237.7	.089	237.7	.095	103.8	.059	114.6	.070
Lithuania	LT	1	227.1	.105	227.1	.104	104.7	.075	119.6	.085
Netherlands	NL	1	243.3	.107	232.5	.095	121.6	.075	125.1	.087
New Zealand	NZ	4	249.5	.112	249.5	.105	117.7	.081	137.4	.102
N. Ireland (UK)	UK-ni	4	275.2	.123	243.0	.110	118.1	.084	129.5	.094
Norway	NO	2b	235.4	.122	235.4	.105	115.3	.080	131.8	.093
Poland	PL	1	226.2	.112	226.2	.099	120.9	.081	127.9	.095
Russian Federation	RU	1	229.3	.120	229.3	.106	110.0	.084	104.7	.081
Singapore	SG	1	271.1	.105	271.1	.105	151.8	.077	175.4	.085
Slovak Republic	SK	1	241.0	.127	241.0	.108	99.4	.079	117.2	.092
Slovenia	SI	1	249.4	.107	249.4	.108	122.4	.078	135.3	.086
Spain	ES	4	226.6	.108	226.6	.104	123.5	.085	129.6	.094
Sweden	SE	2b	230.5	.109	230.5	.101	122.2	.077	132.8	.087
United States^10	US	2a	234.4	.110	234.4	.107	126.1	.083	144.8	.112

Figure A1 Skills inequalities in PISA and SAS