Activism – active contact with duty bearers such as MPs, local government
officials, government officials, etc.
Political Trust – trust in public officials and bodies, such as the president,
parliament, ZEC, the police, etc.
Support for democracy – preferring democracy and being satisfied with
Zimbabwe’s democracy;
Political party affiliation – expressed support for a political party.2
All AB measures were converted to binary variables in order to construct a score for each
index. The data was entered on a spread sheet and analysed in SPSS (20).
Hypotheses
Given our overall research question - are the views of the elite grounded in their experience
of engagement in the general political life of Zimbabwe – it was appropriate to generate
some testable hypotheses. These are described as follows:
Voice – Elites should have higher scores on voice;
Community participation – Elites should have higher scores on community
participation;
Political participation – Elites should have higher scores on political participation;
Activism – Elites should have higher score on activism;
Support for democracy – Elites should have higher scores on support for democracy;
Political trust – Elites should have lower scores on political trust;
Political affiliation – Should be no difference between Elites and Non-elites. We
might predict that Elites would support MDC-T and that Non-elites would support
ZANU-PF, but it seemed parsimonious to merely predict a split as suggested.
Results
Table 1: Elite versus Non-elite, tests of means
Elite
Voice
Community Participation
Political Participation
Activism
Support for Democracy
Political Trust
2.35
1.01
1.34
0.44
0.36
2.88
Nonelite
2.67
1.36
1.72
0.75
0.46
4.24
df
1198
1198
1198
1198
1198
1198
t
-1.701
-3.04
-3.947
-3.31
-2.456
-6.681
Sig
ns
0.002
.000
0.001
0.014
.000
As can be seen in Table 1, virtually all hypotheses are rejected. Although there is no
difference between the groups in respect of voice, Elites have lower community and political
2
In practice, there are only two parties (ZANU-PF and MDC-T), with negligible numbers supporting other
parties, and nearly 50% either refusing to answer or stating the question was inapplicable. Here we
concentrated only on expressed choice: it is not possible from the data to infer a choice from those not
expressing one, even though some critics would argue that this is because of the “fear factor”, with the
unwillingness to disclose support of an opposition party for fear of reprisal.
8