Promoting the wellbeing of Africans through policy relevant research on population and health

Promoting the wellbeing of Africans through policy relevant research on population and health

African Population and Health Research Center

Authors: Moses NgwareMoses Oketch

Alex Ezeh

Assessing the impact of free primary education policy on access and schooling outcomes in Kenya

Outline

• Background

• Purpose

• Design

• Results

• Conclusion

Background 1• Kenya Introduced FPE in 2003

FPE has the following elements1. Universal coverage2. Universal eligibility

• Aims of FPE:1. Improve enrolment by increasing

public expenditure to education 2. Increase education attainment and

reduce overall poverty by mitigating against intergenerational transfer of low human capital.

Background 2

3. Improve the quality of public education

• NB: The introduction of FPE coincided with the political transition in Kenya.

• FPE is a vehicle to realising UPE, which is an EFA-MDG

Purpose

• How did different population groups respond to FPE?

Did FPE change enrolment patterns? Did the patterns differ by population

groups? If the patterns were different, what

explains it? How does enrolment patterns of

different population groups relate with children’s indulgence in risky behaviour?

Study Design (Case study)•A longitudinal household surveyNested in NUHDSS that tracks

approx. 60,000 individuals – 2 sitesEducation research program

household longitudinal survey- 4 sitesNB: 1. DSS existed before the FPE.

2. This motivated the study of assessing how different groups responded to FPE since DSS was in 2 slums of Nairobi, ERP was designed to cover 2 additional nonslum sites to provide a different population group.

Study Design (Case study) Cont….

What did the design enabled us to do?

To compare how slum population groups and nonslum population groups responded to the introduction of FPE. By doing so we are able to assess how the poor and the less poor responded to the policy.

Study sample• Sampling methods:• Purposive:

ERP household survey : In each site households were identified based on CBS cluster enumeration

–7405 households–13,257 individuals aged 5-19–Twice every year

Cont ….

Study sample cont’d…..DSS household survey: nested

–We have about 60,000 individuals living in about 21,000 households

–We collect data every four months (120 days), so thrice a year

–We are currently in round 17 of data collection

Population in slums

Korogocho Viwandani

12 10 8 6 4 2 0 2 4 6 8 10 12

0-4

10-14

20-24

30-34

40-44

50-54

60-64

70-74

80-8485+

12 10 8 6 4 2 0 2 4 6 8 10 12

Male

Percent of Total Population

Female Male Female

Percent of Total Population

Study instruments

Parent/guardian questionnaireHousehold Characteristic questionnaire

Child schooling History questionnaire (or Update)

Chid Behaviour questionnaire (12 yrs and above)

Movement forms (in-migration and out-migration)

Data entry

Datasets generated

Cross-sectional datasetsLongitudinal datasets/panel datasets

Qualitative data

Methods of data analysis

Descriptive analysisRegression analysis (OLS, logit, probit)

Qualitative analysis

Results•A revisit to the questions:

Did FPE change enrolment patterns?

Did the patterns differ by population groups?

If the patterns were different, what explains it?

How does enrolment patterns of different population groups relate with children’s indulgence in risky behaviour?

Trend in school enrolment 2000-2007: slums

Trend in school enrolment 2000-2007: slums

52.4750.81

49.31

64.96 65.64

60.7358.17

56.73

25

30

35

40

45

50

55

60

65

70

2000 2001 2002 2003 2004 2005 2006 2007

Year

Perc

enta

ge

Slum % Public

Slum % Private

Trend in school enrolment 2000-2007: non-slums9.

73

10.4

6

11.7

2

13.4

2

16.0

5

18.8

9

19.1

3

20.4

3

0

20

40

60

80

100

2000 2001 2002 2003 2004 2005 2006 2007

Year

Per

cent

age

Non-slum % Public Non-slum % Private

Trend in school enrolment 2000-2007: non-slums

Results Cont.. (slum model)OR SE

HHS (coef) 0.05

0.01

ATHEA (Coef)

0.06

0.02

HHW: Poorest

2 1.19

0.1

3 1.03

0.09

4 1.13

0.1

Least poor 1.37

0.12

SITE : KOCH

VIWA

1.94

0.13

HHG: Male 0.80

0.05

The odds of enrolling in a public school are high in Viwandani than Korogocho

More of slum least poor households are attending public schools (OR=1.37) compared to the poorest

Pupils from male headed households have low odds of enrolling in a public school

Result cont.. (non-slum model)OR SE

HHS (coef) 0.23

0.05

ATHEA (Coef)

0.30

0.07

HHW: Poorest

2 0.48

0.18

3 0.31

0.11

4 0.25

0.09

Least poor 0.16

0.06

SITE : Jericho

Harambee

1.55

0.34

HHG: Male 0.68

0.014

The odds of enrolling in a public school is high in Harambee than Jericho

Less of non-slum least poor households are attending public schools (OR=0.16) compared to the poorest

Pupils from male headed households have low odds of enrolling in a public school

Results Con’t…Despite FPE, children from poorer households are still

less likely to be enrolled compared to the less poorChildren living in non-slum locations are more likely to

enrollChildren from female-headed HH are more likely to

enrollChildren from smaller in sized households had a better

chance of enrolling. Even among the poor slum residence, those children

from households where the head had more education were more likely to enroll

Results Con’t…

Orphan type matter more than orphanhood in school enrolment

Maternal orphans were more associated with negative attitude towards schooling and had lowest attendance

ConclusionSlum residents schooling patterns show that

they have not responded to FPE as would have been expected

In spite increased public expenditure in public schooling, the less poor remain more represented in the public school system than the poor, i.e. the odds of the poor enrolling in public schools is higher relative to the less poor.

Policy engagement with the government has led the MOE in Kenya to acknowledge that FPE has not included the slum residents as was intended.

With 60% of Nairobi residents living in slums, no wonder Nairobi province registers the lowest enrolment in public schools in spite of Nairobi being overall a wealthy urban province.