Principal investigator(s):
Joachim Bruess, Klara Debeljak (InterMedia)
Data Collection Organization:
Synovate Pan Africa
Survey Period: Jul-12-2008 to Jul-30-2010
Sample:
Sampling frame: 2002 census, carried out in 22 regions of Tanzania. The target population at 2009 was projected to be 23.3 million, with a 48% : 52% Male : Female ratio and a 29% : 71%
Urban : Rural ratio
Sample design: Nationally representative, stratified multistage (5-stages) cluster method
PSU selection: The basic unit available from the report was at Ward level. The wards were defined as the primary sampling units (PSUs), since they are the smallest well-defined geographical units for which population and household data are available from the published report available to the public
Household selection: Households were selected using the ‘left hand rule’ selection methodology, starting from a fixed landmark in each sampling point, using the date to determine the starting house.
Respondent Selection: Where there was more than one eligible respondent in the household, the random selection procedure known as the Kish Grid was used to select the respondent to be interviewed.
Universe:
Both sexes, 15 and more years
Survey procedure: Individual face to face Interview
Sample size: 2000, of which 1,425 are rural and 575 are urban, distributed using Probability Proportionate to Population Size (PPPS) across the regions
Estimated Error: +/-2.2% at 95% confidence level at the national level. It should be noted that the degree of precision will vary (i.e. be higher than +/-2.2%) at the urban and rural levels
Response rate: 53.79%
Weighting – Age Group and Gender
Universe:
Both sexes, 15 and more years
Survey procedure:
Personal Face to Face Interview
Sample size:1566
