Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.573620
Title: Improving school attendance by raising school quality
Author: Khan, Muhammad Jehangir
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2012
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
Using a two generational model of investment in education, this thesis shows that school quality enhancement is very highly valued by the average rural Pakistani family, and by those below the official poverty line. Corollaries are that quality enhancement will be an effective policy for boosting school attendance and that subsistence poverty is not a major reason for keeping children out of school. The empirical strategy is to demonstrate (chapter 4) that school quality variables influence perceived child cognitive achievement. Parental perceptions are critical rather than objective performance (if they diverge) because it is parents that decide whether their children should attend school. These results provide support for the finding in the following chapter (5) that some school quality variables also significantly influence school attendance. Note that the empirical focus of this study is on the more important attendance rather than the commonly discussed school enrolment. The data source, the 2004-5 Pakistan Rural Household Survey, PRHS-II, is the only one currently available for Pakistan that makes available a multitude of school quality measures. Identification in the empirical models of school attendance and cognitive achievement is achieved first by establishing and utilising a distinctive feature of the Pakistani environment; the arbitrary and random allocation of resources to state schools economy and society. The quality variable library is confirmed exogenous with the only instrument in the data set suitable on grounds of relevance (correlation with library), community population. If school resources were adjusted to school need, school attendance would not influence the staff-student ratio. But instrumenting this ratio changes its sign in the predicted fashion. For cognitive achievement, cmp (conditional (recursive) mixed process estimator) (Roodman 2009) is implemented to endogenise staff-student ratio in ordered probit models. Since we have either controlled for endogeneity or established the exogeneity of the quality measures in the attendance and cognitive achievement equations, we can be confident that the parameter estimates correctly capture the impact of school quality variables. The possibility that other variables, not instrumented in the school attendance or cognitive achievement equations, are endogenous does not bias these estimates. Nor does including a range of extra community characteristics in the school attendance and cognitive achievement equations affect the school quality parameter values estimated. School quality measures must be measured at the community/village level to capture the options for school non-attenders. This reduces the precision with which the standard errors can be estimated. Normally it is appropriate to cluster standard errors at the village level, and these results are presented. Judging by parental assessment of child performance, in rural Pakistan private schools achieve better outcomes than state schools. Despite comparatively low private school fees, the average rural household with three children is unable to afford to send their children to private schools. Simply expanding private schooling provision then is not a solution to the currently poor education available to these households. In any event such expansion is not without its problems because there are typically arbitrary political or regulatory barriers to establishing private schools. Therefore this thesis suggest that improving (perceived) state school quality may be the most effective strategy for improving human capital in rural Pakistan.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.573620  DOI: Not available
Keywords: L Education (General)
Share: