Biased, not blind: An experimental test of self‐serving biases in service users’ evaluations of performance information
| Date | 01 September 2018 |
| Published date | 01 September 2018 |
| Author | Julian Christensen |
| DOI | http://doi.org/10.1111/padm.12520 |
ORIGINAL ARTICLE
Biased, not blind: An experimental test
of self-serving biases in service users’evaluations
of performance information
Julian Christensen
Department of Political Science, Aarhus
University, Aarhus, Denmark
Correspondence
Julian Christensen, Department of Political
Science, Aarhus University, Bartholins Allé
7, Aarhus 8000, Denmark.
Email: julian@ps.au.dk
Based on literature about motivated reasoning, this article pro-
poses that choosing a public service provider from among compet-
ing options may bias service users in a positive direction when
evaluating the performance of their chosen provider. Users are
expected to defend their choice through processes of goal repriori-
tization, meaning that they will alter the weight they assign to given
pieces of information depending on the (in)convenience of that
information. This article uses nine experimental studies to test this
expectation on students who had recently chosen to study at one
university instead of competing universities. As expected, findings
show signs of biases in students’evaluations, but the biases are
small and not consistently significant. Thus, prior research may
have been too pessimistic regarding the general potential of perfor-
mance information in the public sector.
1|INTRODUCTION
How do users of public services make sense of performance information? That question is essential for an assess-
ment of the potential of performance information as a tool for performance improvements. Following the explosion
in recent years of performance measurement (Van Dooren 2011), which has led scholars to label the beginning of
the twenty-first century the ‘era of governance by performance management’(Moynihan 2008, p. 3) or even ‘the age
of quantified performance’(Van de Walle and Roberts 2011, p. 215), this question is more relevant than ever before.
Performance information can be collected for many purposes (Behn 2003), but most would agree that the major
purpose of measuring performance is to make performance improvements (Behn 2003). One prevalent strategy for
pursuing this goal is to publish performance information while letting citizens choose freely from among competing
providers of public services (Jilke 2015). Thus, marketization efforts have given citizens all over the world the ability
to choose between competing options when, for example, placing their child in a kindergarten, choosing a university
for their education, or receiving treatment at a hospital, and people are free to leave in favour of competing providers
if they are dissatisfied with the services they receive. In other words, citizens have been given an opportunity to hold
organizations accountable by reacting through channels such as voice and exit in cases of unsatisfactory performance
DOI: 10.1111/padm.12520
468 © 2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/padm Public Administration. 2018;96:468–480.
To continue reading
Request your trial