Creating a 2000 IES-LFS Database in Stata



PROVIDE Project Technical Paper 2005:1

February 2005


1. Introduction

The Income and Expenditure Survey of 2000 (IES 2000) conducted by Statistics South Africa
contains detailed information on income and expenditure of households. When merged with
the Labour Force Survey of September 2000 (LFS 2000:2), the data can be used for the
compilation of various sub-matrices of South African Social Accounting Matrices (SAMs).
Up until recently most of the existing post-1995 South African SAMs, including the
PROVIDE national and provincial SAMs, relied on the IES 1995 and the October Household
Survey (OHS) of 1995 for data on income, expenditure and employment patterns of
households. The use of the more recent 2000 dataset can be regarded as an improvement since
changes in employment and income and expenditure patterns are likely to have taken place
between 1995 and 2000. However, there are some concerns about the quality of the IES 2000
data in particular. Correcting these errors, which range from simple computing errors to more
serious inconsistencies in the data, has proven to be quite an elaborate process, especially
since there is no single correct way of treating such problems. However, as argued by Van der
Berg
et al. (2003a), the IES 2000 is the most recent available data and one should attempt to
work with it.

This paper is organised as follows. Section 2 gives a brief overview of the data files and
sample design of the IES 2000 and LFS 2000:2. A section is also devoted to the theory of
sampling and weighting. Next, the merging of the IES 2000 and LFS 2000:2 is discussed, and
finally some of the data problems, specifically in the IES 2000, are discussed in some detail.
Section 3 describes the Stata do-files that were used to read in the data from the original
ASCII-format data files and create person- and household-level IES 2000 and LFS 2000:2
Stata-format data files. Section 4 explains some of the final data adjustments made to prepare
the dataset for extraction of various household- and factor-related SAM sub-matrices. Some
concluding remarks are made in section 5.

2. An overview of IES 2000 and the LFS 2000:2

2.1. Data files

2.1.1. IES 2000

The IES is conducted by Statistics South Africa every five years. It measures the detailed
income and expenditure of households. These surveys were originally designed and are still
used to determine weights for the South African Consumer Price Index (CPI).
2 The IES is,

2 Because of this objective of the IES consumer goods and services are not grouped according to the Standard
Industrial Classification (SIC) codes. The SIC classification is used for activities and commodities in the
1
© PROVIDE Project



More intriguing information

1. NATIONAL PERSPECTIVE
2. Motivations, Values and Emotions: Three Sides of the same Coin
3. The name is absent
4. Effects of red light and loud noise on the rate at which monkeys sample the sensory environment
5. The name is absent
6. From music student to professional: the process of transition
7. The name is absent
8. The name is absent
9. The name is absent
10. Micro-strategies of Contextualization Cross-national Transfer of Socially Responsible Investment
11. Ability grouping in the secondary school: attitudes of teachers of practically based subjects
12. Transgression et Contestation Dans Ie conte diderotien. Pierre Hartmann Strasbourg
13. The name is absent
14. Cultural Diversity and Human Rights: a propos of a minority educational reform
15. The name is absent
16. The name is absent
17. The name is absent
18. The name is absent
19. The Role of Land Retirement Programs for Management of Water Resources
20. Cross border cooperation –promoter of tourism development