MPOI applied a random selection method at every stage of sampling and also a probability
proportionate to population size (PPPS) wherever possible. A sample size of 1 200 allows for a margin of
error of plus or minus 2.8% at the 95% level of confidence. This is regarded as sufficient to make
population inferences. The sample was stratified first by province and secondly by rural/urban
distinctions within provinces. In total 100 enumeration areas (EAs) were sampled with the latter
subdivided into 71 rural EAs and 29 urban EAs (Table 2).
Province
Manicaland
Masvingo
Harare
Mashonaland Central
Midlands
Mashonaland West
Mashonaland East
Matebeleland North
Matebeleland South
Bulawayo
Total
Table 1:
Projected Population for Zimbabwe 2010
Urban
Rural
Total
214 647
1 226 203
1 440 850
32 197
1 383 858
1 416 055
1 395 207
0
1 395 207
135 943
1 217 444
1 353 388
354 168
980 962
1 335 130
343 436
919 652
1 263 088
146 676
1 033 514
1 180 190
64 394
1 059 790
1 124 184
46 507
937 169
983 676
844 279
0
844 279
3 577 453
8 758 593
12 336 046
% of national population
11.7
11.5
11
11
11
10.2
9.6
9
8
7
100
In order to use up-to-date population figures, the Institute requested the Central Statistical Office (CSO)
to project the Zimbabwe population figures for 2010. Zimbabwe population projections for the period
2002 to 2010 were computed using the cohort component method. The method is defined simply by the
use of estimates and projections of births, deaths and net migration to update the population. The
cohort component method is inherent in Spectrum computer package. The Spectrum software was
designed by Futures Group International for making population projections taking HIV-AIDS into
account. The issue of internal migration remains a major challenge for both the CSO and research
organisations in Zimbabwe, because of difficulties associated with factoring that component into the
projections. Tables 1 and 2 summarise Zimbabwe’s projected population figures for 2010, as well as EA
sample allocation of EAs by province.
Preparation for and implementation of fieldwork
MPOI undertook fieldworker training in late November 2010. Training included trial interviews and
feedback on technical aspects of the questionnaire. At the end of November, MPOI dispatched three
teams of six interviewers each, each under the leadership of a supervisor. The ‘Harare team’ covered
Harare province, Mashonaland Central, parts of Mashonaland East and Mashonaland West; the
‘Bulawayo team’ covered Bulawayo province, Midlands, Matebeleland North and Matebeleland South;
and the ‘Manicaland team’ covered Manicaland, Masvingo and part of Mashonaland East.
12