Selection and performance in post‐compulsory education

• Published date: 01 February 2022
• Author: Uzma Ahmad,Steven McIntosh,Gurleen Popli
• Date: 01 February 2022
• DOI: http://doi.org/10.1111/rode.12818

Rev Dev Econ. 2022;26:3–31. wileyonlinelibrary.com/journal/rode

|

3

© 2021 John Wiley & Sons Ltd

1

|

INTRODUCTION

In this paper, we investigate the determinants of (1) the decision to participate in post- compulsory edu-

cation (i.e., when continuing into education is a choice, usually made at the age of 16years) and (2) per-

formance in post- compulsory education, controlling for selection into post- compulsory education. In our

analysis, along with other key drivers such as prior attainment and family socioeconomic status (SES),

we specifically investigate the accessibility of post- compulsory education, measured as the distance to

the nearest institution that provides post- compulsory education, as a determinant of participation. We

Received: 11 September 2019

|

Revised: 12 February 2021

|

Accepted: 4 July 2021

DOI: 10.1111/rode.12818

REGULAR ARTICLE

Selection and performance in post- compulsory

education

UzmaAhmad1

|

StevenMcIntosh2

|

GurleenPopli2

1Sheffield International College, University

of Sheffield, Sheffield, UK

2Department of Economics, University of

Sheffield, Sheffield, UK

Correspondence

Gurleen Popli, Department of Economics,

University of Sheffield, Sheffield S1 4DT,

UK.

Email: g.popli@shef.ac.uk

Abstract

This paper investigates the determinants of participation

and performance in post- compulsory education, controlling

for selection into post- compulsory education and prior at-

tainment, using a unique primary data set on pupils study-

ing in the post- compulsory grade in 2011– 2012 from one

district of Punjab province in Pakistan. The main findings

of this paper show that participation and performance in

post- compulsory education are two different processes,

with participation being driven by the availability of post-

compulsory institutions within travel distance and perfor-

mance, once in post- compulsory education, determined by

ability. The results further highlight that distance reduces

participation most for those living in rural areas.

KEYWORDS

distance, Pakistan, participation, performance, post- compulsor y

education, selection

JEL CLASSIFICATION

I20; J24

4

|

AHMAD et Al.

use a unique primary data set from Pakistan for our analysis, with detailed information on individual

and family characteristics, which is then linked with administrative data to obtain educational outcomes.

There exist a literature investigating the determinants of participation in post- compulsory educa-

tion and a limited separate literature investigating the determinants of performance in post- compulsory

education. However, there are no studies to our knowledge that jointly investigate the determinants of

participation and performance in post- compulsory education. There are studies that investigate both

participation and performance in education together. A large proportion of these studies, however, are

in the context of conditional cash transfer (CCT) programs (see Fiszbein & Schady,2009, for a review

and Saavedra & García, 2012, for a meta- analysis), which focus on compulsory education (mainly

between ages 6 and 16). A few studies that are outside the CCT framework focus either on primary ed-

ucation (e.g., Duflo,2001) or on labeled cash transfer programs with a continued focus on compulsory

education (Benhassine etal.,2015). Our paper is significantly different from the existing studies in the

literature in two key aspects. (1) In the CCT framework participation is mandatory and monitored as a

part of the program; this is not the case in our study, where participation is a choice. (2) Unlike other

studies in the literature, we focus on participation and performance in post- compulsory education.

The often- discussed determinants of the decision to participate in post- compulsory education

include prior attainment of the individuals, their family SES, and perceived benefits of schooling

(Carneiro & Heckman,2002; Chevalier etal., 2013; Eckstein & Wolpin,1999). Some studies also

relate local labor market conditions with enrollment in post- compulsory education, with evidence sug-

gesting that individuals are more likely to participate in education if their local labor market is weak, as

education becomes more attractive if employment opportunities are relatively scarce (McIntosh,2001,

for Germany, Netherlands, Sweden, and England; Bradley & Lenton,2007, for Britain; Clark,2011,

and Meschi etal.,2019, for England).

Similarly, the factors determining performance in education, in general, include ability, fam-

ily SES, teacher quality, and school characteristics; for example, see Chevalier etal.(2013) for the

United Kingdom; Black etal.(2005a) for Norway; Engin- Demir (2009) for Turkey; Kasirye (2009)

for Uganda; and García (2014) for Russia. For post- compulsor y education, specifically, the impact of

working while in school has also been studied; a review by Neyt etal.(2019) finds that though work-

ing while in school has an impact on educational choices (engagement and continuation), it does not

have an impact on education performance.

In developing countries, there are additional determinants of educational participation and attain-

ment. On the demand side, poor and resource- constrained families see few immediate benefits of

education, and on the supply side, the availability of, or access to, schools is limited. Whereas CCTs,

information- based interventions, and scholarship programs have been a common policy response to

the demand- side constraints, for the supply- side factors, the common policy response has been to

increase access to schools and improve teacher quality and salaries (Glewwe & Muralidharan,2016).

Accessibility of schools is an important consideration as costs associated with education increase

with distance; these costs can be financial (travel and relocation costs), temporal (commuting time),

and psychological (inconvenience and unpleasantness of commuting every day).1 The existing literature

studying the impact of accessibility of schools on both participation and attainment has primarily fo-

cused on building new primay schools (e.g., see Duflo,2001, and Mazumder etal.,2019, for Indonesia;

Handa,2002, for Mozambique; Alderman etal., 2003, and Lloyd etal.,2005, for Pakistan; Burde &

Linden,2013, for Afghanistan; and Kazianga etal.,2013, for Burkina Faso) and decreasing the distance

cost for access to middle schools (Adukia etal.,2020; Muralidharan & Prakash,2017).2 There exists

very little evidence on the impact of accessibility to institutions providing post- compulsory education.

Two recent exceptions are Dickerson and McIntosh (2013), who study the impact of distance to an in-

stitution on the decision to participate in post- compulsory education at age 16 in the United Kingdom;