19
The last part of our social capital index to consider is the amount of
interpersonal trust our interviews revealed. It is important to point out the
distinction between interpersonal or social trust, on the one hand, and
trust in institutions or political trust, on the other. Sometimes the two are
confused conceptually, but having trust in your fellow citizens does not
necessarily mean you have trust in the political institutions; or vice versa.
Several studies in OECD countries have shown that there often are strong
correlations between trust in people and trust in government or political
institutions.
20
There are heated debates whether the first causes the second
(e.g. Putnam 2000) or the second causes the first (e.g. Rothstein &
Kumlin 2001). But our data indicate that the connection between
interpersonal trust and government performance (in India) may not be
particularly strong at all. I have shown elsewhere (Blomkvist 2001) that
OLS regression show there is a negative relation between people’s trust
and (their evaluation of) government performance.
What role does education play in this? Looking at the whole sample of
3,200 respondents there is a statistically significant effect of level of
education on political participation.
21
But if we look at this relationship
for each state another and very interesting picture emerges. The effect of
the individual’s education on his/her political participation disappears in
Kerala and West Bengal and even becomes a negative factor in Gujarat.
Given the conventional wisdom that education always and everywhere
affects political engagement and attitudes, and the central role (primary)
education often is given in aid and discussions about development this
result is astonishing. Are we simply wrong? Or can these data be
interpreted in a light that make them more comprehensible?
If we sharpen our lens and make the picture more nuanced by looking at
the separate caste groups for the whole sample and the respective states
we find what is reported in tables 17 and 18.
20
In our data the picture is much more varied, depending on what kind of interpersonal trust we talk
about.
21
OLS regression, controlling for the individual’s economic standard, trust in Indians, participation in
associations, support network, bridging network, and discussion network.
Convention