32
M. Wilke et al.: Risk Specificity for Risk-Defusing Operators
means of preventive actions. Thus, we expected ques-
tions and statements to most frequently concern possi-
bilities for control.
Exploratory Research Questions
We know from previous research (e.g., Huber & Huber,
2003) that information besides searching for RDO might
be important for the decision-making process. Therefore,
we wanted to explore different aspects to gain ideas for fur-
ther improvements in the theory.
Background knowledge and prior experience
Because the risk situations in our study (especially our cat-
astrophic and global types of risk) extend beyond the every-
day risks that Huber studied, we wanted to check the role
of prior experience and background knowledge in the de-
cision-making process. There are clear data on the role of
prior experience and expertise in decision making (e.g., El-
terson, Shanteau, & Krogstad, 1987; Weiss & Shanteau,
2003). We assumed that it would be activated in connection
with RDOs throughout the different types of risk. For ex-
ample, prior experience with negative events such as flood-
ing is related to the intention to buy natural catastrophe in-
surance (Kunreuther et al., 1978; Zaleskiewicz, Piskorz, &
Borkowska, 2002). Ranyard, Hinkley, and Williamson
(2001) found that risk management strategies in buying con-
sumables using credit depended on prior experience and
emotional reactions. We expected statements concerning
specific background knowledge and prior experience to be
important elements for the decision maker. This is in line
with Huber and colleagues (Huber, 2004; Huber et al., 1997;
Huber et al., 2001), who mentioned the potential role of
background knowledge and experience with respect to the
subjective representation of the situation and to the search
for RDOs.
Attitudes and values
Based on experiences from our pilot study, it seemed use-
ful to take into account our participants’ general attitudes
and values (e.g., plans, rules, and basic principles), which
are independent of the particular scenario, but help partic-
ipants make decisions. An example could be the opinion
that “One should never interfere with nature’s plans.” This
basic principle would support decisions concerning health
problems and ecology. Decision making might be based on
general attitudes and values for the global type of risk more
frequently than for other types of risk.
Situation, probabilities, negative and positive
consequences
Following Huber et al.’s (1997, 2001) category system, we
analyzed type-specific differences in decisions about risk
with respect to the frequency of questions and statements
in the categories of probabilities, cost-benefit, and the gen-
eral situation (see Table 2).
Choice of alternatives
We analyzed whether participants chose the risky or the less
risky alternative.
Method
Participants
A total of 120 individuals (64 women, 56 men) participat-
ed in our study. Their mean age was 30.1 years. Nearly half
the participants were employed (with and without univer-
sity-level education); the others were schoolchildren and
college students. None of them had participated in previous
studies on decision making. All materials were presented
in German (the participants’ first language).
Design
Two independent variables were used for our main study.
The first one was the type of risk (normal, medium, cata-
strophic, and global), the second represented the risk do-
mains (ecology, health, and politics/economics), resulting
in a 4 × 3 within-subjects design. Factor levels were oper-
ationalized by the content of risky scenarios. For each of
the four types of risk, we chose examples from the three
risk domains (ecology, health, and politics/economics).
A complete design would have required participants to
evaluate all 12 scenarios. Following Ranyard, Williamson,
and Cuthbert’s (1999) recommendation, we decided to use
an incomplete design by reducing participants’ load to the
evaluation of three scenarios. We selected one scenario from
each domain for each participant, whereby each of the three
scenarios corresponded to a different type of risk. So, data
from a total of 360 evaluations were collected (120 partic-
ipants with three scenarios each). The four types of risk were
each covered by 90 evaluations; the three risk domains were
each covered by 120 evaluations. For all participants, the
sequence in which the risk domains were presented was held
constant (health, economics/politics, ecology), but the se-
quence in which the types of risk were presented was var-
ied. Huber and Huber (2003) as well as Huber and Macho
(2001) found no sequence effects in their scenarios. There-
fore, we decided to keep the sequence of risk domains con-
stant.
Swiss J Psychol 67 (1), © 2008 by Verlag Hans Huber, Hogrefe AG, Bern