29
Table 2: European integration in Britain, 1994 to 1997
Place Labour more pro-EU than the
Conservatives
European integration coef.
1994
1997
N
%
1994 1997 Diff.
W.Contr.
All
-
-
671
100%
0.89
1.41 0.52
(0.24) (0.24) (0.34)
Yes
Yes
298
44%
2.53 2.61 0.08
0.04
Priming or other salience effects
(0.42) (0.49) (0.65) (0.31)
Learning effects
No
Yes
148
22%
0.23
2.38 2.15
0.47
(0.39) (0.46) (0.60) (0.14)
Partial learning effects
No
Better
82
12%
-0.46 0.29
0.75
0.09
(0.55) (0.50)
(0.74)
(0.17)
No learning
No
No
77
11%
-0.52 -1.18
-0.66
-0.08
(0.79) (0.81)
(1.13)
(0.14)
Forgetting
Yes
No
66
10%
0.65 0.19 -0.46 -0.05
(0.72) (0.68) (.99)
(0.09)
Note: Probit estimates (standard errors in parentheses.) The first row shows that attitudes about European
integration became more related to vote choice in 1997 compared to 1994. It presents coefficients for
integration attitudes from a probit model of major-party vote choice (Labour versus Conservatives). The
1994 and 1997 estimates are from separate models with variables only from the respective year. Since the
UK held no election in 1994, the question asks for vote choice “had there been an election.” The next set of
rows shows that the increased relationship arose among those who learned the parties’ positions on this issue
by reestimating these models with interactions for each of the knowledge categories. The weighted
contribution (W.Contr.) column estimates the extent to which each row contributes to the overall increase by
multiplying the Diff. column by the proportion of the sample in that category. The standard errors in this
column are bootstrapped to account for uncertainty in the coefficients and in the proportion of individuals in
each category. See the appendix for control variables.
Convention