(Page 9 of 23) - If You Build a Political Website Will They Come? The Supply and Demand Model of New Technology, Social Capital, and Civic Engagement in Britain authored by Norris, Pippa. and Curtice, John.

‘I

F YOU BUILD A POLITICAL WEBSITE

…’

ORRIS AND

URTICE

8/20/2004

1:16

The summary Political Activism Index, providing an overview, is constructed simply by

adding together experience of each of these different types of acts (each coded 0/1). The study
has therefore formed additive indices of each item within each of the four groups and also created
an additive index across all four groups. It should be noted that the resulting summary index is
currently heavily dominated by the civic oriented scale, as this has the most items. Subsequent
analysis will see whether any weighting procedure makes any substantial difference to the
results. It should also be noted that in this conceptual framework, with the important exception of
partisan identification, this study focuses upon political activity; we are concerned with doing
politics rather than being attentive to public affairs or having psychological attitudes thought
conducive to civic engagement, such as trust in parliament or a sense of political efficacy,
explored elsewhere (Curtice and Norris 2004). The study therefore does not regard exposure or
attention to mass communications, exemplified by following political events in newspapers, as
indicators of political activism per se. These factors may indeed plausibly contribute towards
participation, and thereby help explain this phenomenon, as prior pre-conditions, but they are not,
in themselves, channels which citizens can use for expressing political concerns or mobilizing
group interests.

Part III: Analysis of the results

We can start by examining the simple bivariate relationships between Internet use and

these multidimensional indicators of political activism, with the important proviso that the
background of online users continues to be skewed towards the well-educated and more affluent
social sectors, which are both resources closely associated with political activism, so that
multivariate analysis is required controlling for these factors.

[Table 1 about here]

Table 1 presents the mean score on these indicators for the group of all Internet users

compared with all non-users, and the significance of the difference (estimated by ANOVA). In line
with our initial expectations, the results confirm that Internet users in Britain proved significantly
more politically engaged than non-users across the indicators of cause-oriented and civic-
oriented participation, as well as in the total activism scale. This pattern suggests that the forms
of political involvement which are most likely to benefit through the development of the Internet
are through single-issue politics, voluntary associations, and community groups, as expected
given the prior characteristics of the online population. By contrast, if we compare more traditional
forms of engagement, Internet users were slightly less likely to vote than non-users, and there
were no significant differences between users and non-users in levels of campaign activism.

The comparison of activism by length of Internet use displays a slightly different pattern,

illustrated in Figure 4, where early adopters with the longest experience of going online proved
significantly more politically active across all dimensions except voting turnout (where there was
no difference) when compared against those who had first ventured online more recently. Yet
without any controls it is not possible to determine whether this pattern was due to the effects of
length of exposure to the Internet per se, or whether this reflected the gradual changes in the
social composition and political attitudes of the Internet population.

[Table 2 and Figure 4 about here]

The multivariate regression analysis presented in Table 2 displays the results of the full

model predicting overall levels of political activism, including the range of social and demographic
controls as well as the impact of regular newspaper reading and use of the Internet. The results
show that, as expected, age had a significant and strong effect upon political participation, with
people becoming more active as they enter middle-age, with a slight fall in a curvilinear pattern in
the over 70s. Interestingly, gender and race appear to be insignificant predictors of activism in
this model, although women and men in Britain have been found to have different patterns of
participation in other studies (Norris, Lovenduski and Campbell 2004). As many others have long
reported, by providing civic skills and boosting a sense of internal efficacy, education is one of the