Disparities in education along socio-economic lines in Pakistan

Sehar Saeed and Huma Zia

Idara-e-Taleem-o-Aagahi / ASER Pakistan
“I want to send my children to school. I want them to
have a better life than mine. But I cannot afford to pay Such analysis on patterns of inequality in learning outcomes
their fees, buy them books or get them pencils every will bring the attention of the policy makers to formulate
month. Mostly people of this village don't send their policies that empower children from poor backgrounds to
beat the odds. The imbalances if not checked will push the
children to school and make them work instead as all of
inequities in the education sector further down the abyss.
us are too poor. My children have been dropped out of Providing equal opportunities in schooling along with
primary school as I had no money” strengthening quality of education can serve as a benchmark
for bringing a change in social and economic outcomes. An
The voice is that of Sakina Bibi; a mother and a resident of a equitable distribution of educational opportunities will allow
remote village in Balochistan. One of the harsh realities the poor to gain from the benefits of economic growth and
behind the education crisis in Pakistan cannot be better contribute towards higher growth rates. Whereas, depriving
summarized than the story presented above. the poor from educational opportunities will result in lower
economic growth and amplification of income inequality4.
Today where due to rapid globalization, economic activity is
becoming increasingly knowledge based and education has Hence, equitable access and learning is a key to “sustained
gained importance more than ever, the education system of development”. This research appraises education inequalities
Pakistan continues to stay ineffective and unproductive. The in Pakistan with the help of ASER data (2012) covering 136
vital role and significance of education is largely mistreated rural districts of Pakistan and investigates if the children from
and ignored in Pakistan1. Moreover, the provision of the lower income groups are worse off.
educational opportunities is unfortunately determined and
made available on the basis of regional disparities, rural- In order to highlight the above mentioned aspect of our
urban location, gender, types of schools, income and wealth education system, an ASER composite wealth index has been
of parents etc. Pakistani society has become largely constructed by integrating all the households indicators
fragmented and segregated on various socio-economic lines mentioned in the survey form. These indicators measure the
since the last couple of years. The inequality in income and economic potential and achieved levels of income and wealth
wealth not only continues to grow with every passing year but of a household. The table representing the variables used to
also has triggered disparities in education2. The propagation create the wealth index is described below.
of private schooling system has further intensified the
disparities resulting in polarization of education along socio- Variables Description
economic lines3. People falling in lower-middle income group Type of house is a cate gorical
remain deprived of quality education provided by private variable with kutcha given the value
schools due to exorbitant fees charged by them while the Type of house
1, semi-pucca equals 2, and pucca
government schools fail to come up at par in terms of quality
equals 3.
of education. ASER (The Annual Status of Education Report)
data reflects such inequalities very clearly. ASER 2012 pointed Dummy equaling 1 if the house is
House owned
out the dismal performance of government schools as owned, 0 otherwise.
compared to private ones in language and arithmetic Dummy equaling 1 if the house had
Electricity
assessments. electricity (visible wires and fittings),
connection
0 otherwise.
ASER data can further be used to identify the relationship Dummy equaling 1 if the household
between students' performance and the disadvantages they Toilet
had a toilet, 0 otherwise.
face because of their home background. The household Dummy equaling 1 if anyone in the
indicators tapped during the survey can be used as a baseline Mobile
house has a mobile, 0 otherwise.
to determine the wealth status of households. A comparison Dummy equaling 1 if the household
of wealth status of households with the learning levels of Television
has a television, 0 otherwise.
children can provide a snapshot of the extent of inequality in
learning levels across wealth distribution.

1
Jamal. H and Khan. A. 2005. “The Knowledge Divide: Education Inequality in Pakistan”. The Lahore Journal of Economics.
2
UNESCO, EFA Global Monitoring Report. 2012.
3
A Study of Education, Inequality and Polarization in Pakistan. Tariq Rehman. Oxford University Press.
4
Javid. K. 2011. “Rural-Urban Divide in Education- Inequities Reinforcing Inequities”. ASER 2011

20 ASER 2012 - National

 Construction of ASER wealth index: The status of wealth was also found to be influencing gender
ASER wealth index has been developed by using principle inequity. The males and females of the lowest quintiles are
component factor analysis procedure in the STATA software.5 particularly disadvantaged as only a limited set of educational
Using the above mentioned method of creating quintiles, opportunities is available to them. The percentage of males
ASER 2012 data has been divided into four and females enrolled in schools goes up as we move along the
categories/quintiles (i.e. poorest, poorer, richer, and richest) wealth index towards the richest. Inadequate public
which represent the entire population of Pakistan in a socio- expenditure in rural areas, limited number of schools,
economic context. obsolete teaching methodology etc. might be the reasons
leading towards restricted access to basic education which
Results of the ASER 2012 data reveal that the poorest quintile further transforms into learning gaps across different income
has the highest level of children enrolled in government groups.
schools (91%) whereas the remaining 9% of the children are
enrolled in the private sector schools. The second quintile,
which is poorer, has 82% children enrolled in government Wealth index: out of school children
schools and 18% children enrolled in private schools. The 100
third quintile, richer, has 75% children enrolled in government 80

% children
schools and 25% in private schools. The richest quintile has 60
46 33
the highest number of children enrolled in private schools 40
27 19
20
(40%) and the lowest percentage of children in government
0
schools (60%). It is evident from the figures that enrollment in
Poorest Poorer Richer Richest
government schools falls and for that of private school
increases as we move along the wealth index towards the
richest. A strong correlation between wealth and enrollment
Wealth index: learning levels (highest competency levels)
in private schools is established. Though a number of low fee
private schools exist in the country, they are still more Poor Poorer Richer Richest
expensive than their public counterparts and thus are not 100
affordable for all income quintiles6. 80
% Children

60
32 42 31 42 29 38
A large proportion of households are not able to send their 40 26
16 15 25 14 23
children to schools at all because of poverty. Result of ASER 20
2012 displays the percentage of out-of-school children to be 0
highest in the poorest quintile (46%) as compared to other Urdu/Sindhi/Pashto: English: Reading Mathematics:
Reading Story Sentence Division
quintiles.

Given the bleak picture portrayed by the disparities in

enrollment according to types of schools, a similar image Wealth index: enrollment by gender
comes to light when the “learning levels” according to wealth Males Females
status are taken into account. The graph clearly indicates that
100
the learning levels of children are directly related to their 73 78 84
80
wealth status. The learning level of children in all three
% Children

62
60
subjects increases as we move along the wealth index 64
76
40
towards the richest quintile. Poorest have the lowest learning 39 55
20
levels (16% Urdu/Sindhi/Pashto, 15% English, and 14% Math)
0
and richest have the highest learning levels (42%
Poorest Poorer Richer Richest
Urdu/Sindhi/Pashto, 42% English, and 38% Math). The
households with better wealth status are able to spend
significantly more on their children's education improving
their opportunities for better quality schooling as reflected by
the enrollment figures mentioned above.

5
It factorizes variables in a way such that it creates weighted combination of the input variables in the following manner e.g.
F1 = a11X1 + a12X2 + ….
In order to select factors, eigen values from a principal component analysis are used and the factor coefficient scores are created.
Further, the indicator values are multiplied by the coefficient scores and added to come up with the wealth index.
The index is then divided into groups/quintiles to categorize the population according to their wealth status.
6
Bari. F and Sultana.N. 2011. “Inequality in Education”. ASER 2011.

ASER 2012 - National 21

 The differences in learning levels for both males and females
across different quintiles present an alarming picture. Wealth index: learning levels - males
Learning levels of males and females improve as we move
Poor Poorer Richer Richest
from the poorest quintile to the richest quintile. Highest
learning levels of females are seen in the richest quintile 50
43 43
across the three competency levels (41% 39
40
Urdu/Sindhi/Pashto, 40% English, and 36% Mathematics). 34 33
Similarly males falling in the richest income group are better

% Children
31
30
able to perform the language and numeracy tasks than 28
27 26
children falling in low income groups. This also confirms with 20
19
the findings of PISA survey 20097 that established: “the higher 18 17
10
the quartile of the socio economic index to which a student
belonged, the better the performance, with a similar pattern 0
for boys and girls.”(EFA Global Monitoring Report 2012) Urdu/Sindhi/Pashto: English: Reading Mathematics:
Reading Story Sentence Division

The current state as demonstrated by ASER 2012 clearly

illustrates the appalling access and gender disparity created
by differences in wealth status. This also corroborates with Wealth index: learning levels - females
the results of World Inequality Database on Education (WIDE) Poor Poorer Richer Richest
produced by EFA Global Monitoring Report Team at UNESCO.
50
The WIDE Database has provided figures for over 50 countries
41
to allow for comparison in disparities across countries and to 40
40
identify which groups are most disadvantaged within these 36
% Children

29 27
countries on the basis of gender, wealth and location. 30
25
22
20
Article 25A embedded in the 1973 constitution of Pakistan 21 18
that promises free and compulsory education for all children 10
12
aged five to sixteen appears to be meaningless in a country 11 10

where the education system is fragmented and inequality 0

persists to such an extent. If our objective is to educate all Urdu/Sindhi/Pashto: English: Reading Mathematics:
Reading Story Sentence Division
children, we need to challenge the existing differences and
divisions in order to provide equal set of opportunities to all
children of the society. Moreover, at a time when the
international community begins to plan post-2015 education
goals and framework, it is vital to ensure that equity based
targets are included and measuring marginalization in
education is given a high priority.

7
Amongst learning assessments, PISA has done the most comprehensive coverage and surveyed 74 countries:
all the OECD countries and forty other countries. The survey assessed the performance of 15 year olds and in addition
collected data on parental occupation and education, selected home characteristics such as availability of books.

22 ASER 2012 - National