Sampling Procedure
The South Africa Enterprise Survey 2007 included enterprises with more than four employees as well as micro establishments, firms with less than five workers. There are 120 micro establishments in the sample.
The sample for enterprises with more than four employees was designed using stratified random sampling with strata defined by region, sector and firm size.
Establishments located in Johannesburg, Cape Town, Port Elizabeth and Durban were interviewed.
Following the ISIC (revision 3.1) classification, the following industries were targeted: all manufacturing sectors (group D), construction (group F), retail and wholesale services (subgroups 52 and 51 of group G), hotels and restaurants (group H), transport, storage, and communications (group I), and computer and related activities (sub-group 72 of group K). For establishments with five or more full-time permanent paid employees, this universe was stratified according to the following categories of industry:
1. Manufacturing: Food and Beverages (Group D, sub-group 15), Machinery and Equipment (Group D, sub-group 29), Electrical Machinery and Equipment (Group D, sub-group 31);
2. Manufacturing: Textiles (Group D, sub-group 17), Garment (Group D, sub-group 18), Leather and Footwear (Group D, sub-group 19), Paper and Paper Products (Group D, sub-group 21), Printing and Publishing (Group D, sub-group 22);
3. Manufacturing: Non-Metallic Mineral Products (Group D, sub-group 26), Basic Metals (Group D, sub-group 27), Fabricated Metal Products (Group D, sub-group 28);
4. Manufacturing: Wood and Wood Products (Group D, sub-group 20), Furniture (Group D, sub-group 36)
5. Manufacturing: Refined Petroleum Products (Group D, sub-group 23), Chemical Products (Group D, sub-group 24), Rubber and Plastics (Group D, sub-group 25)
6. Retail Trade: (Group G, sub-group 52);
7. Rest of the universe, including:
• Other Manufacturing (Group D excluding sub-groups in strata 1-5);
• Construction (Group F);
• Wholesale trade (Group G, sub-group 51);
• Hotels, bars and restaurants (Group H);
• Transportation, storage and communications (Group I);
• Computer related activities (Group K, sub-group 72).
Size stratification was defined following the standardized definition used for the Enterprise Surveys: small (5 to 19 employees), medium (20 to 99 employees), and large (more than 99 employees). For stratification purposes, the number of employees was defined on the basis of reported permanent full-time workers.
The implementing agency (EEC Canada) was unable to obtain a satisfactory sample frame from South African statistical agency (STASA) or its Department of Revenue. The best alternative solution was a list obtained from the Department of Trade and Industry Companies and Intellectual Property Registration Office (CIPRO), which contained about 800000 establishments when delineating in-scope cities and industries, but which had incomplete firm characteristics necessary for stratification purposes (e.g. contact information, size). In order to determine the sample frame, EEC Canada randomly drew 9550 units and contacted them.
In South Africa, the survey included panel data collected from establishments surveyed in the 2003 Investment Climate Survey (ICS) of South Africa. That survey included establishments in the manufacturing and the rest of universe strata, distributed across Gauteng (Johannesburg), KwaZulu Natal (Durban), Western Cape (Cape Town) and Eastern Cape (Port Elizabeth) provinces.
In order to collect the largest possible set of panel data, an attempt was made to contact and survey valid establishments (579) in the panel list provided which was part of the Enterprise Survey's scope. Of the 716 establishments provided to EEC Canada from those surveyed in 2003, there were 35 doubles, 8 out-of-scope, 89 excluded from this survey by The World Bank to avoid over representing Construction in a single Residual stratum, and 5 with undefined ISIC codes. This left a total potential of 579 panel establishments. EEC Canada surveyed 231 panel establishments or 40% of the total potential panels without eliminating those establishments which had closed. Once eliminated, this percentage coverage exceeded 55%. Given the non-random nature of panel establishment selection, these establishments are not allocated probability weights in the final dataset.
In this survey, the micro establishment stratum covers all establishments of the targeted categories of economic activity with less than 5 employees located in Johannesburg. The implementing agency selected an aerial sampling approach to estimate the population of establishments and select the sample in this stratum for all states of the survey.
First, to randomly select individual micro establishments for surveying, the following procedure was followed: i) select districts and specific zones of each district where there was a high concentration of micro establishments; ii) count all micro establishments in these specific zones; iii) based on this count, create a virtual list and select establishments at random from that virtual list; and iv) based on the ratio between the number selected in each specific zone and the total population in that zone, create and apply a skip rule for selecting establishments in that zone.
The districts and the specific zones were selected at first according to local sources. The EEC team then went in the field to verify the sources and to count micro establishments. Once the count for each zone was completed, the numbers were sent back to EEC head office in Montreal.
At the head office, the count by zone was converted into one list of sequential numbers for the whole survey region, and a computer program performed a random selection of the determined number of establishments from the list. Then, based on the number that the computer selected in each specific zone, a skip rule was defined to select micro establishments to survey in that zone. The skip rule for each zone was sent back to the EEC field team.
In Johannesburg, enumerators were sent to each zone with instructions how to apply the skip rule defined for that zone as well as how to select replacements in the event of a refusal or other cause of non-participation.
For complete information about sampling methodology, refusal rate and weighting please review "South Africa Enterprise Survey 2007 Implementation Report" in "Technical Documents" folder.