Creating a 2000 IES-LFS Database in Stata

PROVIDE Project Technical Paper 2005:1

February 2005

Table 3: Cross-tabulating gender, location and race

Gender________	_____________Person-level_____________		__________Household-level__________
LFS^ IES→	Male__________	Female___________	Male___________	Female__________
Male___________	________48799	_____________424	__________15785	_____________79
Female__________	__________414	___________53960	____________46	_________10196
Accuracy________	_______99.2%	__________99.2%	_________99.7%	_________99.2%

Location________	__________Household-level__________
LFS^ IES→	Urban_________	Rural______________
Urban__________	________14466	_____________383
Rural____________	__________1415	____________9854
Accuracy________	_______91.1%	__________96.3%

Race___________	________________________________Person-level________________________________
LFS^ IES→	African_________	Coloured__________	Asian___________	White___________
African__________	________83896	______________60	______________2_	______________12
Coloured________	_____________93	___________11462	______________12	_______________7
Asian___________	____________3_	_______________0_	__________2040	_______________4
White___________	____________36	_______________6_	_____________0_	__________5914
Accuracy________	_______99.8%	__________99.4%	_________99.3%	_________99.6%

	_____________________________Household-level_____________________________
LFS^ IES→	African_________	Coloured__________	Asian___________	White___________
African__________	________20719	________________15	______________1_	_________________5
Coloured________	_____________33	____________2689	_____________2_	______________1_
Asian___________	_____________1_	_______________0_	____________523	_______________2
White___________	____________6_	_______________2_	_____________0_	__________2098
Accuracy________	_______99.8%	__________99.4%	_________99.4%	_________99.6%

2.4. IES 2000 data problems

2.4.1. Literature review

Most of the data problems in the IES 2000 dataset can be ascribed to accounting and coding
errors (Poswell, 2003). There are also a few inconsistencies when compared to the IES 1995
dataset. This restricts the confidence with which conclusions can be drawn about changes in
income or expenditure over time. Van der Berg et al. (2003a) point out some of the specific
problems that they have encountered in the IES 2000 dataset:

• When compared to IES 1995 the 2000 results indicate that income in South Africa has
been declining strongly in real terms between 1995 and 2000. This contradicts national
accounts and demographic statistics also compiled by Statistics South Africa. Statistics
South Africa has since admitted that the surveys are incomparable.^{15, 16} When building a

¹⁵ This inconsistency is also pointed out by Simkins (2003). Simkins looks at the components of income and
finds that there are large and inexplicable drops in some of these components, particularly net profits
from business (half the 1995-level in nominal terms), occupational perquisites (down 42%) and
16

More intriguing information

1. Spectral calibration of exponential Lévy Models [1]
2. Structure and objectives of Austria's foreign direct investment in the four adjacent Central and Eastern European countries Hungary, the Czech Republic, Slovenia and Slovakia
3. Gerontocracy in Motion? – European Cross-Country Evidence on the Labor Market Consequences of Population Ageing
4. Comparison of Optimal Control Solutions in a Labor Market Model
5. Opciones de política económica en el Perú 2011-2015
6. The name is absent
7. The name is absent
8. Are Japanese bureaucrats politically stronger than farmers?: The political economy of Japan's rice set-aside program
9. Skill and work experience in the European knowledge economy
10. Placenta ingestion by rats enhances y- and n-opioid antinociception, but suppresses A-opioid antinociception