Comparing teaching practices, teacher content knowledge and pay in Punjab
Introduction
Despite recent improvements in learning outcomes, Pakistan struggles to impart quality education to school children, with only 59 per cent of children in class five being able to read a story in Urdu, Sindhi or Pashto (ASER, 2019). With teachers being a critical input in the learning process, provincial governments have turned their attention to strengthening recruitment practices, as well as leveraging the private sector to improve the quality of education. Punjab-Pakistan’s most populous province-is home to approximately 115,000 public primary school teachers and 74,000 private primary school teachers (National Education Management Information System, 2018). In addition to these teachers, there are more than 90,000 teachers teaching in Public Private Partnership (PPP) schools in the province. Given these large numbers and the attention of the provincial government on teacher recruitment and PPP reform initiatives, it is essential to learn more about the teachers who teach in public, private and PPP schools to begin to understand the extent to which the current stock of teachers are equipped to facilitate improvements in student learning.
The objective of this paper is to present insight into the backgrounds of teachers who teach in public, private and PPP schools. Specifically, the paper contributes to the literature on teaching practices, teacher content knowledge mastery and teacher pay in Punjab, Pakistan. The study also attempts to provide some correlational evidence on the differences in content knowledge mastery between public school teachers that were recruited after the introduction of test-based teacher recruitment and teachers that were recruited through the previous recruitment method. Lastly, the paper explores the correlates of teacher pay in the three types of schools which, to the best of my knowledge, is the first study of its kind to do so within this context. The four research questions addressed in this paper are presented below:
- 1
To what extent does school type, teacher content knowledge, years of teaching experience and teacher training explain the variation in teaching practices?
- 2
What are the correlates of teacher content knowledge mastery for public, private and PPP school teachers?
- 3
Are public school teachers recruited through the test-based recruitment method more likely to have higher content knowledge than teachers recruited through the old method (prior to 2013)?
- 4
Is there a relationship between teacher content knowledge and teacher pay, and how does this vary for public, private and PPP school teachers?
This paper proceeds as follows. Section 1.1 presents a brief overview of teacher recruitment and PPP reform programmes in Punjab. Section 1.2 presents a brief literature review, while section two presents the conceptual framework underpinning the study, methods, and data used to address the research questions. Section three discusses the main findings of the study while section four presents the conclusion.
Two of the major reforms implemented by the School Education Department (SED) in Punjab have been introducing test-based teacher recruitment policies for public school teachers, and expanding the provision of education services through the establishment of PPPs in Education. This sub-section provides a brief overview of both reform initiatives in the province.
Public sector teacher recruitment processes have drastically evolved over the years. Prior to 2011, teacher recruitment in Punjab was susceptible to nepotism with politically connected candidates being able to manipulate the system and secure teaching positions. However, the 2011 public sector teacher recruitment policy, raised the professional qualification/certification requirements for candidates, and introduced mandatory induction training which culminated with an assessment (which counted towards the final merit list score)1 (Bari et al., 2013). In 2013, the School Education Department (SED) introduced a mandatory assessment for all teaching candidates with the objective of assessing candidates’ content knowledge in core school subjects, IT skills, knowledge of pedagogy and general knowledge. In order to progress to the interview stage, candidates had to have scored 50 per cent or higher on this assessment.2 All subsequent teacher recruitment policies have included a mandatory test for teachers, however, the pass marks for the test and the weightage has varied. In 2014 for example, the pass/fail benchmark on the assessment was lowered from 50 per cent to 45 per cent. In 2016, the pass/fail benchmark on entrance test was changed once again back to 50 per cent, and the entrance test included additional marks for math and science for Elementary School Educators (ESE).3 Since 2014, the SED has hired public school teachers on a contract basis for a duration of five years, however unlike previous contracts, teachers are paid basic pay scale salaries plus 30 per cent (in lieu of pension benefits that are offered to permanent teachers).4 Although there is no direct evidence that nepotism in teacher recruitment has decreased during this period, there are positive indications: the education profiles of public school teachers have improved- in 2010 only 41 per cent of primary public school teachers had a bachelors’ degree or higher, now this figure is approximately 69 per cent, and for the last few years the SED has conducted third party validations5 of the recruitment process to ensure greater transparency in teacher recruitment (National Education Management Information System, 2018).
In parallel, the SED has launched several PPP initiatives in education through the Punjab Education Foundation (PEF) with the dual objectives of increasing school participation and improving the quality of education. The central argument for partnering with the private sector for this purpose is based on evidence that despite having less qualified, less experienced teachers, low free private schools tend to outperform public schools in terms of student achievement at a fraction of the cost (Andrabi et al., 2006). Since 2005, PEF has scaled up its PPP programmes-currently there are four main programmes operating in the province, catering to 2.6 million children. Although the programmes vary in their focus and design, the basic modus operandi is the same. PEF screens and identifies potential partners (entrepreneurs, NGOs etc.) who enrol children free of cost, and are offered payments by PEF on a per-student basis. Partner schools must pass a quality assurance test administered to students and meet other infrastructure and facilities related requirements in order to remain eligible for partnership with PEF (Ansari, 2020a). An important feature of these is that partner schools are paid a fixed amount per student (which only varies by level of schooling) that parents cannot top-up since partner schools are not allowed to charge any tuition fees to parents. This design feature of the PPP programmes means that PEF controls the fee/subsidy rates offered to partner schools, and these are not determined by the schooling market. This raises questions about whether these imposed fees/subsidy rates are having any potential impact on teacher pay packages within these schools. For example, if the per student subsidy amount paid to partner schools is less than the prevailing market tuition fee in low fee private schools, this could have a detrimental impact on teacher salaries in PPP schools, providing an incentive for partner schools to cut salaries to maximize profits.
The importance of teachers and teacher quality in improving learning outcomes for students is well documented (Metzler and Woessmann, 2010; Hanushek and Woessmann, 2011). Researchers have investigated the impact of specific teacher characteristics on student learning in various contexts. For example, there is substantial evidence that standard teacher resume characteristics such as education level and professional qualifications, teacher training and experience (beyond the first two years of teaching) do not matter much when it comes to improving student achievement (Staiger and Rockoff, 2010; Aslam and Kingdon, 2011; de Talancé, 2017). Studies have also reported that having a teacher with greater content knowledge is associated with greater student achievement (Rawal et al., 2013; Bau and Das, 2017; Bold et al., 2017). Given traditionally higher rates of teacher absenteeism in the public sector and the difficulty in firing low performing public school teachers due to their permanent contracts, studies have explored whether teachers on temporary contracts are associated with greater student achievement (Chaudhury et al., 2006). The evidence from developing countries suggests that students of contract teachers experience larger test gains than students assigned to permanent teachers (Muralidharan and Sundararaman, 2013; Duflo et al., 2014; de Talancé, 2017). Given the relationship between teacher content knowledge and student learning, studies have dug deeper to identify some of the correlates of teacher knowledge (for example see Hill and Lubienski, 2007). A study of math teachers from Germany found that sex, grade point average, and cognitive abilities were associated with content knowledge scores (Kleickmann et al., 2013). Studies have also found that teacher education is another factor that is correlated with teacher content knowledge (Schmidt et al., 2007; Bold et al., 2017). Bau and Das (2017) report that in public schools, female teachers are associated with lower content knowledge scores in math and Urdu while teachers with higher education qualifications are associated with higher content knowledge scores across math, English and Urdu. The authors also found that teachers with temporary contracts were associated with higher English content knowledge scores than teachers with permanent contracts.
While there is evidence on specific teacher characteristics that may or may not improve student learning, information on effective teacher behaviours or practices is more limited. Researchers have recently attempted to utilise classroom observation tools such as the Classroom Assessment Scoring System (CLASS) to estimate teacher effectiveness (Grossman et al., 2010). Kane and Staiger (2012) compared five different classroom observation instruments in the US and found that the various measures were all positively associated with student achievement and similar relationships between observed teaching practices and student achievement have also been observed in developing countries (Araujo et al., 2016; Wolf et al., 2018). More recently, the World Bank designed and administered a classroom observation tool-Teach- with a particular focus on measuring teaching practices in primary schools in low and middle income countries (Molina et al., 2018). Using data collected from the Teach tool in Punjab, Pakistan Molina et al. (2018) found a positive association between teach scores and student achievement in math, Urdu and English.
Given the central role of teachers in improving student learning, an important question from a policy perspective is whether teacher quality and ultimately student achievement, is associated with higher teacher salaries. And while the evidence on the effect of teacher pay on student achievement is mixed, understanding how teachers are remunerated and incentivised is critical to ensure that teaching as a profession is an attractive option for skilled candidates (Hanushek, 2003). Studies have examined differences in teacher pay in public and private schools. Studies from India and Pakistan report that low-fee private school teacher wages are one fifth of public school wages (Muralidharan and Kremer, 2006; Andrabi et al., 2008). Researchers have attempted to identify some of the correlates of teacher pay and evidence from the United States suggests that teacher certification and teacher education have a positive association with teacher wages (Chambers, 1985). Aslam and Kingdon (2011) found that for private school teachers in Pakistan, teacher tenure at the school is positively associated with wages, while in public schools teacher certification and teacher experience are associated with higher wages. In a study from Punjab, Bau and Das (2017) found that educational qualifications and teacher training were positively correlated with teacher wages in both public and private schools while contract teachers were paid less in public schools. The authors also reported that female private school teachers are paid less than their male counterparts, even after controlling for background characteristics. Lastly, the authors found no correlation between mean teacher productivity (measured as teacher value added), teacher content knowledge and teacher pay in the public sector, while they did find some evidence of a relationship between teacher value added and teacher pay in the private sector. However as Andrabi et al. (2008) caution, observed characteristics do not necessarily explain the public-private teacher wage gap in Punjab, Pakistan.
While there has been significant attention paid to teacher quality and teacher pay, this study addresses important gaps in the literature. This paper builds on previous research on public, private and PPP schooling in Punjab by exploring differences in teacher characteristics and behaviours across the three school types. To the best of my knowledge, it is the first paper of its kind to explore the correlates of teacher behaviour in the classroom for public, private and PPP school teachers in Punjab. Given the important relationship between teacher content knowledge and student achievement, the paper attempts to identify correlates of teacher content knowledge mastery separately by school type, and provides some correlational evidence on the difference in content knowledge for public school teachers recruited through the test-based system and the previous recruitment system. Lastly, the paper goes beyond previous studies to examine the correlates of teacher pay in Punjab across public and private schools, with the inclusion of PPP school teachers as an additional category. The purpose of this analysis is to determine the extent to which different types of schools reward teachers with greater content knowledge. As mentioned earlier, the imposition of a ceiling on tuition fees for PPP schools raises concerns about whether partnering with the public sector may alter teacher remuneration packages in participating private schools.
Section snippets
Conceptual framework
In order to understand the linkages between the respective research questions, it is useful to identify the underlying conceptual framework for assessing teacher effectiveness. This study is predicated on Blömeke and Delaney’s (2012) framework for teacher competency which integrates earlier work of Shulman (1985); Richardson (1996) and Thompson (1992) into one comprehensive framework. Blömeke and Delaney’s (2012) framework represents teacher competency as a function of both cognitive abilities
Correlates of teaching practice
This subsection presents the main findings for the correlates of teaching practice, specifically exploring whether school type, teacher content knowledge, and teacher training, are associated with teacher practices. Descriptive statistics for the 751 teachers for which classroom observations were available along with complete information for other covariates are presented in Table A2 in the appendix. 67 per cent of the teachers in the sample work in rural areas, and 47 per cent of the sample is
Conclusion
This paper contributes to the existing literature on teacher behaviour, content knowledge and teacher pay in the context of Punjab, Pakistan. Specifically, the paper explores the correlates of teaching practice, the correlates of teacher content knowledge mastery across public, private and PPP schools and the correlates of teacher pay in the three types of schools.
The findings reflect that private school teachers are associated with lower classroom observation scores (teach scores) than their
Acknowledgements
I am indebted to Anna Vignoles and Cristina Santos for their comments on this paper. Any errors and omissions are my own.
References (37)
Cream skimming? Evaluating the access to Punjab’s public-private partnership programs in education
Int. J. Educ. Dev.
(2020)- et al.
What can teachers do to raise pupil achievement?
Econ. Educ. Rev.
(2011) Patterns of compensation of public and private school teachers
Econ. Educ. Rev.
(1985)- et al.
The economics of international differences in educational achievement
(2011) - et al.
A Dime a Day: the Possibilities and Limits of Private Schooling in
(2006) - et al.
Learning and Educational Achievements in Punjab Schools: Insights to Inform the Policy Debate. Report
(2008) Evaluating the Effectiveness of Punjab’s Public-private Partnership Programs in Education
(2020)- et al.
Teacher quality and learning outcomes in kindergarten
Q. J. Econ.
(2016) - et al.
An Investigation into Teacher Recruitment and Retention in Punjab
(2013) - et al.
The Misallocation of Pay and Productivity in the Public Sector: Evidence from the Labor Market for Teachers
(2017)