Page 364 - Invited Paper Session (IPS) - Volume 1

IPS169 Markku L.
conditions (legal, political, institutional) need to be fulfilled in order to enhance
such confidence?” However, rather than focusing on means of ‘enhancing
confidence’, I will seek to provide a more nuanced picture of the multiple roles
of trust in indicator work, and highlight the potential downsides of “overtrust”
and virtues of mistrust and distrust. The operational question then becomes:
what role, if any, might constructive mistrust and distrust play in strengthening
rather than undermining the production and use of composite indicators, in
the context of alleged loss of trust and post-truth politics? I suggest that the
concepts of trust, mistrust, and distrust briefly introduced here could be
helpful in attempts to better understand the role of multiple roles of statistical
offices and composite indicators. Further empirical research will be needed to
corroborate (or invalidate) the applicability and usefulness of the framework.

2. The multiple dimensions of trust, mistrust and distrust
2
Trust can be defined generally as a stance whereby an individual accepts
‘believing without knowing’, thus placing herself voluntarily in a position of
vulnerability towards ‘the other’, be it another individual or an institution
3
(Earle & Siegrist 2006). Trust represents a ‘leap of faith’ (e.g. Davies 2018),
because there is always a risk that the ‘trustee’ proves untrustworthy, yet as a
voluntary choice, trust does not have to imply the feeling of loss power and
control (Espluga et al. 2009).
Three mutually interacting dimensions of trust can be distinguished. First,
social trust is interpersonal. It can entail generalised trust in other, unknown,
members of society (Rothstein & Stolle 2008) or particularised (specific) trust
in people we already know, with whom we interact regularly, for example in
our own social or demographic group (Bäck & Christensen, 2016, 180). For
indicators, a key issue then concerns particularised trust in statisticians as
individuals. Arguably, this interpersonal type of trust has little to do with
current problems of loss of trust in indicators and statistics – few citizens have
regular encounters with statisticians, or have particular trust/mistrust relations
with statisticians.
Institutional trust denotes the public trust in institutions such as
statistical authorities, the government, government regulation, or NGOs.
Institutional trust can entail specific support, in other words, individuals’
judgement of what the institution does (its performance), or diffuse support,

2 For the sake of simplicity, I use the term trust to encompass both its traditional meaning as a
normative judgement concerning an individual or entity, and confidence, that is, a belief based
on earlier experience that certain events will occur as predicted (Earle & Siegrist 2006; Luhmann
2006; Kinsella 2016).
3 Following Hodgson (2006, 18), institutions are here defined broadly, as “systems of established
and embedded social rules that structure social interactions”. Organisations, in turn, are a
specific type of institution.
353 | I S I W S C 2 0 1 9