activity between the two German regions since East and West Germany while different,
are not historically distinct along this particular regional division.
Both German cases have markedly higher aggregate figures for active membership
than Romania. According to the data collected in the second wave of the WVS, there is a
39 percent difference between West Germany and Romania for active membership. East
German and Romanian figures also show a substantial difference of just over 24 percent.
The aggregate figures for active membership further substantiate how post-communist
experiences vary among cases.
In all cases, the rate of active membership increased between the two waves of the
WVS with the greatest increase occurring in West Germany. East Germans became more
actively involved but at a lower rate than their Western counterparts. The difference
between the two German cases and Romania remained proportionally similar across the
application of the two surveys.
The regional distinction for Romania shows that Transylvanians were more frequently
active members in associational life than those outside the region showing that civil
society varies across region. However, the number of active members in Transylvania
still is lower than the figures for both West and East Germany.
members in West Germany to East Germany is 1.35, while the ratio of Transylvania to
the non-Transylvanian regions of Romania is 2.22. So, we observe variance in both
countries, but the regional discrepancy is much higher in Romania. The ratio figures
accentuate how the communist legacy cannot explain the difference for Romania in
simple terms if we assume that all regions of Romania had a similar experience under the
communist regime.
3
The 1993 WVS for Romania presented problems with the coding of regions. The population included
in Transylvania significantly deviates from the proportion represented in multiple censuses. With no
reliable variables to check its accuracy within the data set, such as religion or ethnicity, we elected to
exclude it from the analysis. For the 1998 data, we aggregated the coded “region” variable for
Transylvania, Banat and Crisana-Maramures to represent “Transylvania” in our analysis.
8
Convention