Creating a 2000 IES-LFS Database in Stata



PROVIDE Project Technical Paper 2005:1

February 2005


The personwgt.dta file, which contains alternative but unofficial person-level weights for IES
2000, is also merged with
person.dta at this point.32

Next two sub-do-files are run within person.do. The first is do-file factors.do, which
converts employment information contained in variable
jobcode to create factor codes as used
in the SAM (variable
factors). Variable jobcode is based on answers given to the following
question in the IES 2000 questionnaire: “
What kind of work did [the respondent] do in
his/her main job during the past seven days?
” Statistics South Africa uses this information to
give each respondent a four-digit occupation code based on the International Standard
Classification of Occupations (ISCO 88). The meta-data file included with the LFS 2000:2
data shows how these codes can be used to derive an aggregated occupation code variable
(
factors) containing ten types of work and one category for ‘unspecified or not adequately
defined’ (see Table 7). This aggregation differs only in one respect to the occupation codes
used before in PROVIDE (2003b) in that ‘domestic workers’ have now replaced ‘armed
forces’.
33

Table 7: Occupation codes (variable factors)

Factor code

Description

0

Not applicable/not working

1

Legislators, senior officials and managers

2

Professionals

3

Technical and associate professionals

4

Clerks

5

Service workers and shop and market sales workers

6

Skilled agricultural and fishery workers

7

Craft and related trades workers

8

Plant and machine operators and assemblers

9

Elementary Occupation

10

Domestic workers

11

Not adequately or elsewhere defined, unspecified

Source: (SSA, 2002a)

The second do-file in person.do is do-file activities.do. Individuals were asked to indicate
in which industry they work. The answers were used to derive an industry code variable
called
stccode, based on the International Standard Industrial Classification (ISIC 1993) of all
economic activities. Variable
stccode was then used to group workers into 96 different

32 Statistics South Africa has been unable to confirm whether these new weights can be used, hence the
continued use of the old weights.

33 Previously domestic workers were included under unskilled factors, but given that a separate industry for
domestic services is also included in the SAM this distinction is useful. Armed forces used to be a
separate group, but there were concerns about the representativity of this group as a separate factor
account. Only 0.3% of African, 0.2% of Coloured and 0.2% of White workers were members of the
armed forces. On aggregate only 0.2% of all workers were employed as members of the armed forces
(IES 1995). On the other hand, 5.9% of African, 3.6% of Coloured, 0.4% of Asian and 0.1% of White
workers are domestic workers, giving an aggregate of 5.0% of all workers (IES 2000).

35

© PROVIDE Project



More intriguing information

1. DURABLE CONSUMPTION AS A STATUS GOOD: A STUDY OF NEOCLASSICAL CASES
2. TOMOGRAPHIC IMAGE RECONSTRUCTION OF FAN-BEAM PROJECTIONS WITH EQUIDISTANT DETECTORS USING PARTIALLY CONNECTED NEURAL NETWORKS
3. THE RISE OF RURAL-TO-RURAL LABOR MARKETS IN CHINA
4. The duration of fixed exchange rate regimes
5. A Critical Examination of the Beliefs about Learning a Foreign Language at Primary School
6. Strategic Effects and Incentives in Multi-issue Bargaining Games
7. The name is absent
8. The name is absent
9. Sex differences in the structure and stability of children’s playground social networks and their overlap with friendship relations
10. The name is absent