Stigma In Global Context: Mental Health Study
Indiana University . Bloomington . Indiana
 

Home

 

Research Plan 

 

Fielding Schedule and Sample Design

 

Participating
Countries

 

Instrument

Other Links

Please address all general inquiries to:

sgcmhs@indiana.edu

Search:

 

 

 

South Africa Sample Design

Sampling Plan and Timeline:

This study will be included in the Markinor M-Bus (Omnibus) study. These studies are weighted to represent South African adult population 16+ years.

I have however in the meantime also attached the 2004 M-Bus schedule which not only provides field dates, it also provides some technical specifications regarding our sample.

Sample Size:

We offer clients two options in terms of sample sizes pertaining to our M-Bus: a metropolitan sample (sample size is 2000) and national sample (sample size is 3500:  2000 in metropolitan areas and 1500 in non-metropolitan areas).

50% of sample is females and 50% is males.

In order to put the community sizes in South Africa  in perspective,  I will outline a description of what population sizes are included in each area:

Metropolitan  areas (population of 250000+)                         33.7% of adult population

Metropolitan areas in South Africa are as follows:

  • Johannesburg (North incl. Sandton and Randburg, South, Central, East Rand , West Rand , Greater Soweto )
  • Pretoria
  • Vaal Triangle
  • Bloemfontein
  • Cape Town
  • Port Elizabeth
  • East London
  • Durban
  • Peitermartizburg

Non-metropolitan  areas include the following:

Urban:

Cities (population 100000 - 249999)                      8.4% of adult population

Large towns (population 40000 - 99999 )               5.2% of adult population

Small towns (population 8000 - 39999)                 8.9% of adult population

Rural:

Large/small villages (population (500- 7999)          3.4% of adult population

Rural (population less than 500 people)              40.3% of adult population.    

Our sampling methodology for our M-Bus is as follows:

The sample is stratified by province and within province, by community size, city, township and suburb.  Within each stratum, sampling points are selected at random.

In the case of blacks, five plots are selected from each sampling point in a random walk procedure.  Within a plot all households living on this plot are listed, and one household is selected according to a random selection procedure.  Within the selected household all male or female adults are listed and one is selected, again according to a random selection procedure.

If this person is not available for an interview (away, refused, sick), a substitute living in the same street is chosen who is of the same sex, age and working status, 

In the case of whites, coloureds and Indians, five interviews are conducted around the selected sampling point, according to a quota on gender, and within gender, age and working status.

 

1022 E. Third Street, Bloomington, IN 47405 (812) 855-3841

Indiana University

Last updated: 6 September 2005
Comments: sgcmhs@indiana.edu
Copyright 2004, The Trustees of Indiana University
Copyright Complaints