school uniforms

Activists always campaign for rendering education more accessible in the developing world by eliminating school fees and Cambridge International removing other expenditures. The World Bank 2015 has posited that user fees are the main hindrance to universal education in developing countries. Some countries in sub Saharan Africa have abolished school fees but other substantial expenditures persist such as the expense of issuing a school uniform to a child.

Cambridge and non government agencies can intervene in all manner of ways to get children into school including providing free Cambridge uniforms as detailed in this paper free meals Vermeersch and Kremer 2004 free medication Miguel and Kremer 2014 offering a bundle package of services like uniforms textbooks and classroom construction Kremer et al 2012 or increasing the offer of a potential opportunity to win a scholarship Kremer et al 2014. This paper adds to the debate of the best means of getting African children into school. Although some measures like supplying meals or medicine enhance the advantages of Cambridge school uniform provision, removal of school fees minimizes the amount of money needed.

We collaborated with ICS Africa a nongovernmental organization that sponsored some school children to obtain free school uniforms. ICS utilized a lottery system to decide whose children would
receive uniforms this offers a special opportunity to estimate the effect of uniform provision without the selection biases Cambridge that Cambridge curriculum affect observational data.

Although not all lottery winners were given uniforms, winning the lottery is strongly associated with receiving a uniform and therefore makes a good instrumental variable. We estimate strong positive effects of getting a school uniform on student school attendance. Providing a uniform decreases school absence by percentage points 43 from a baseline of 15% school absence. The impact is percentage points greater for pupils with a uniform at the baseline.

Context

The education system of Kenyan is made up of primary secondary and tertiary levels. Primary school consists of standards grades 1 to. Most schools also provide an Early Childhood Development ECD for students too young to attend normal standard 1 but these are usually informal with teachers supplemented by contributions from Cambridge Advanced, Cambridge Checkpoint, Cambridge Lower Secondary, Cambridge Upper Secondary, Cambridge ICE, families not paid a regular salary by the government. The academic year aligns with the calendar year and consists of three terms Term 1 takes place between January and March Term 2 between May and July and Term 3 between September and November. Primary school students usually vary in age between six and seventeen years.

One of the reasons why the age group is so broad is because the number of times pupils repeat standards is so frequent. Students who fail to finish the standard national curriculum are retained to take the year over again. Country students from poor families and with minimal educational experience in the home tend to find the curriculum difficult and thus are retained frequently. It is not unusual for a student to be retained in a standard two or more.

uniforms school

Parents in most countries have numerous education costs like school fees and procurement of uniforms. In Kenya students paid school fees to access primary education up to 2002. In January 2003, a new policy by the government Summative assessment offered not just fees but also minimum textbooks and notebooks. This resulted in phenomenal rises in school enrollment. Nevertheless schools still needed. In the past, those who did not pay their school fees or did not wear uniforms could be sent away from school.

There is anecdotal evidence that students are less likely to be expelled from school for not wearing a uniform after 2002 than before but that students do feel stigmatized through the non-wearing of a uniform and are likely to be disciplined Cambridge examiners by teachers. Current evidence indicates that lowering the cost of schooling through the provision of uniforms among other inputs raises participation in school.

Kremer et al 2002 assess a program where ICS supplied uniforms textbooks and classroom construction to seven randomly sampled schools from a group of 14 underperforming schools in Western Kenya. Dropout rates in treatment schools plummeted and after five years students had completed 15% more schooling. They contend that availability of textbooks only can account for the impact and that the dropout rates decreased.

School Project

ICS Africa has been working in Western Kenya since 1996. One main program has been the Child Sponsorship Program CSP where children are sponsored by donors in the Netherlands and abroad and consequently get school fees and school uniforms. ICS Africa phased its sponsorship program into a number of new schools during 2002 and assessed those components of the program that target individual children compared to those that target the entire school. During fall 2001 ICS Africa chose twelve primary schools in Western Kenya for the CSP.

The twelve primary schools are all located in Busia district, the westernmost district of Kenya, Western province. Busia district shares its border with Uganda and lies to the north of Lake Victoria. Children’s households in CSP schools do not seem to be systematically distinct from other rural households in Kenya. For instance 86% of kids in CSP Cambridge schools owned a toilet at home compared to 79% of rural homes based on the DHS 2003. In January 2002 ICS conducted a children’s census in standards one to four of the twelve schools selected.

From the census ICS chose all the children who had lost one Past papers or both parents and became orphans to be automatically sponsored. It then employed a lottery to randomly pick the remaining beneficiaries. Then an ICS field representative visited the twelve schools to register those children picked for sponsorship into the program. In order to enroll into the program a child had to be present for a photo to be taken and a small information card to be completed which would then be mailed to the sponsor along with some basic information about the child like her preferred.

Data and Empirical strategy

The data we work with is a merge of five separate datasets containing information on pupil attendance students who had won the sponsorship lottery students who had been in school when they enrolled students who had been given uniforms in 2002 and a pupil questionnaire that was conducted in 2002. Illustrates how the Attendance dataset relates to the other datasets. We have attendance data from each of the twelve schools for the period 2002 through the end of 2005.

Attendance was collected as field officers made unannounced visits to each school several times per year and marked down if each child was present. From these multiple visits an overall annual average per child attendance is gathered. 2002 Pupil Cambridge teacher training Questionnaire was conducted in mid January 2002 prior to The Attendance dataset has distinct pupil identification numbers and the 2002 Pupil Questionnaire employs those same pupil numbers.

But datasets with the data of the students who won the lottery the ones present on the lottery day and those who got uniforms do not have those pupil identification numbers. Hence the names were matched manually as well as with the help of a computer program. There were a number of pupils in the different other datasets who did not match against any pupil in the Attendance dataset.  This could be because children at times do have numerous names and might only provide a subset of those in a single data collection exercise and a non overlapping subset in another exercise.

Results

First we look at the first stage regressions What is the likelihood that a child having been randomized into the project actually was registered or went on to receive a uniform Table 2 shows that 74% of children who won the lottery were Cambridge professional development indeed randomized into the project. 77% of children who won the lottery received uniforms in June of that year.

In each of the regressions approximately 20% of children who lost the lottery ended up enrolling in the program. This is because if there were a lottery winner not present on the enrollment day a second child was assigned to replace her. Getting School accreditation, Teaching resources, Professional development qualifications, School support hub, International standards, randomized into the program is therefore a good instrumental variable for true enrollment and subsequent uniform receipt. Provides some of these instrumental variables estimates.

In Regression we look to see if early registration into the program had an influence on children’s school attendance in the six months prior to uniform distribution. We observe an insignificant positive effect on attendance. Some of the low precision of this estimate is due to the fact that it is calculated based on few observations on school attendance. In Regression we offer the intention to treat regression estimating simple effect of being randomized to the project on attendance after uniform distribution and find an effect of percentage points. Then we offer IV estimates of the effect of registration in the program Regression and of receiving a uniform in 2002.