19
voting pattern of this region is closer to Seoul/Megapolis than to either
Youngnam/Ganwon or Honam. Although Daejeon/Chungcheon did show some
significant regional voting (as compared with Seoul/Megapolis) in 1997, the regional
effect diminished in 2002. This is consistent with the fact that a third-part candidate from
Daejeon/Chungcheon was on the ballot in 1997 (but not in 2002).
(Table
2
about
here)
Table 3 presents the standard deviations of the vote share that the winning
candidates obtained from each region. As expected, Honam shows the smallest
dispersions in the vote in both years. Surprisingly, however, the standard deviations of
the of Daejeon/Chungcheong reflect a smaller degree of homogeneity in 1997 than in
2002, contrary to the belief that regional voting was more prevalent in 1997. For the
other regions, the standard deviations are not consistent with the well-known pattern of
regional voting, either. Specifically, the standard deviations of Youngnam/Gangwon are
not very different from those of the Daejeon/Chungcheong, and they are greater than
those of Seoul/Megalopolis. Since Youngnam/Gangwon is supposed to exhibit a greater
degree of regional voting than both Daejeon/Chungcheong and Seoul/Megapolis, the
standard deviation as a measure of homogeneity fails to reflect that belief. What this
means is that even though Youngnam/Gangwon may have a level of average vote share
that is significantly different from those of other regions, it is not necessarily as
homogeneous a voting bloc as the other regions.
(Table
3
about
here)
Table 4 through Table 7 demonstrate the estimates of the mixed regressive-spatial
autoregressive (MRSAR) model of the vote share of the winning candidates for the 1997
Convention