The Impact of Education Quality on Rates of Return to Education

1
The Impact of Education Quality on Rates of
Return to Education

• By
• Tekaligne Godana John Ashipala

2
Format of the presentation

• Educational History In Namibia
• Pre-independence
• Post-independence
• Research objective
• Research Methodology
• Results
• Policy implications

3
Education History, Pre-independence

• Introduced by Missionaries as by-products of
  Evangelical work
• The Prime objective of the Mission was the
  spreading of Christianity, trade and colonialism,
  thus education was neglected
• Their provision of education concentrated on the
  ability to read the Bible and on industrial
  education
• Settlers feared education of blacks would
  threaten their livelihood as traders, farmers and
  artisans

4
Pre-independence
5
Pre-independence
6
Pre-independence
7
Pre-independence
8
Post-Independence
9
Article 20 States that

• All persons have the right to education
• Primary education shall be compulsory and the
  state shall provide reasonable facilities to
  render effective this right for every resident
  within Namibia, by establishing and maintaining
  State schools at which primary education will be
  provided free of charge.
• Children shall not be allowed to leave school
  until they have completed their primary school or
  have obtained the age of sixteen
• All persons shall have the right to establish and
  maintain private schools, colleges or other
  institution of tertiary education

10
Major Goals for education

• Access
• Equity
• Quality
• Democracy
• Efficiency

11
Learner-Teacher Ratio
12
Research objectives

• To assess the relationship between the level of
  different types of school resources and education
  quality.
• To assess the impact of education quality on the
  rates of returns to education.

13
Research Methodology -Questionnaire

• Respondents Biography
• Respondent's family background
• School of respondent
• Respondent employment details
• Income details for respondent in wage employment
• Income for respondent in self employment

14
Math Test

• 18 Multiple choice questions - 15 minutes

15
The Sample

• Sample was designed in consultation with National
  Planning Commission (NPC) - Central Bureau of
  Statistics (CBS)
• CBS provided a representative sample of
  households from their mainframe. This signifies
  the importance attached to the project by
  government, which is crucial in terms of policy
  recommendation. CBS provided
• A random sample of Primary Survey Units (PSU)
• Random sample of HH in each of the selected PSU
• The survey covered the whole country (all 13
  regions) with a total of 42 PSU selected.
• A total of 24 HH were selected from each PSU
  giving a total of 1008 HH

16
Sample Cont,

• Eligible interviewees people older than 15 and
  not in school. Our target 1200. 1168 interviewed.
• 29 out 42 PSUs urban, 13 rural
• The survey covered towns, villages, settlements
  with some PSUs from the communal area.

17
Main Survey
18
Data Capturing and Cleaning

• School level data, 1989 2002
• PTR, Teachers qualification for about 1500
  schools collected for 1992, 1996, 1998 and 2002.
• District level data for 1984, 1986 and 1989
• Challenge during data gathering and processing.
• Approaches to data analysis
• District level analysis
• School level analysis

19
Data merging processes

• Re-allocation of schools to pre-independence
  districts.
• 7 data points covering almost 20 years.
• Quality variables Pupil-Teacher Ratio (PTR) and
  percentage of teacher qualified (quality), PTR1,
  PTR2PTR6, Quality1, Quality2Quality6
• Advantages and Disadvantage of the data merging
  process

20
Descriptive data

• Out of a total of 1168 individuals over the age
  of 15 interviewed wage information was obtained
  for 510 individuals.
• average education of the population has increased
  dramatically between the generations, parents
  having only 2 years of education compared to 10
  years of education for their children
• There is very little difference between mothers
  and fathers education
• The performance on the numeracy test is quite
  low, half of them scoring 5 or lower out of 18
  questions
• Half of the respondent earned less than N8 per
  hour

21
Rates of return estimation
22
Result comments

• All the four models have high explanatory power
  (high adjusted R-square and F-statistics) which
  is not common in cross section data
• All the standard variables (education, experience
  and its square) have the expected signs and
  statistically highly significant
• The rate of return to education is high and quite
  stable across the four estimations

23
Comments continue

• Pupil-teacher ratio has a negative impact on
  returns to education
• Teacher qualification has however the unexpected
  negative sign
• Estimation using pupil teacher ratio in the last
  school attended by respondents did yield
  statistically significant results but the
  magnitudes remained very small

24
Comments continue

• This conclusion though sounds counter-intuitive
  has support from other empirical work
• little impact of school resources on students
  later earnings. (Betts 1999)
• Godana and Ogawa (2003) also find that the
  pupil-teacher ratio seems to have very little
  impact on student performance
• The same study finds however that teacher
  qualification has a positive and significant
  effect on student performance

25
Labour market imperfections
26
Labour market imperfections

• Blacks earn less than coloureds and white earning
  more than coloureds with the same level of
  education.
• The gender bias is also strong with men earning
  higher than women.
• The urban and rural divide is also significant,
  urban labour having a higher return on education
  than rural labour.
• Last but not least unionised labour receives
  better return than non-unionised labour.
• With all these market imperfections it is not
  perhaps surprising that the effect of the quality
  of education on returns to education did not come
  out clear

27
Summary

• School resources as measured by pupil-teacher
  ratio and teacher qualification were found to
  have very little impact on rate of returns to
  education
• Significant distortions and imperfections in the
  labour market which perhaps overshadow the impact
  of education quality on returns to education
• This study gives strong evidence that factors
  like mothers education, household assets,
  attendance of pre-primary education as well as
  cognitive ability of the individual have strong
  and positive impact on earnings

28
Policy implication

• Improving and equitably distributing school
  resources may not be adequate to achieve
  equitable outcome of education in terms of
  earning capacity
• measures which improve household equity or/and
  additional measures that compensate for household
  inequity (school-feeding programmes, provision of
  education materials, improved dwelling, medical
  care, water and sanitation, etc) are necessary to
  attain equity in education outcome
• the need for policy designed to reduce labour
  market distortions which militate against the
  acquisition of better education and skills

29
THANK YOU