Risk Assessment: The Consumer as an Intuitive Statistician - Innovation and Marketing in the Pharmaceutical Industry - page 303

inventory was a reason for respondents to update their perceptions of risk. This
finding led to a simple and straightforward recommendation to those interested in
getting consumers to accept higher risk: highlight that any of the symptoms men-
tioned could place them at risk.

10.14 Concluding Thoughts

The goal of this chapter was to present an overall model of how consumers assess
risk and the biases that could creep into the process. The thesis was that understand-
ing the process used to construct a risk estimate and identifying the source of the
bias would present managerially actionable ways to de-bias risk estimates, irrespec-
tive of whether they were over- or underestimates of risk. Prior literature in the
domain of risk perception was reviewed in the context of the presented model.
However, in this section, we conclude with some thoughts on the open questions
that remain unanswered in this field. Some of these are summarized below.

10.14.1 Symptom Typology

A typology of symptoms and how they could affect the manner in which symptoms
are aggregated into a bottom-up risk estimate has only just begun (Raghubir and
Menon 2005 ). There is a genre of physiological health problems that are diagnosed
using self-reported psychological inventories. Besides depression, these include
alcoholism and the attention-deficit syndrome (ADD). Alcoholism is defined by
Alcoholics Anonymous as allergic physiological reaction to the consumption of
alcohol with the consequence of an inability to stop drinking once the first drink has
been consumed (Alcoholics Anonymous World Services, Inc., 1998). ADD is
another psychological disease with a physiological basis, relying on self- and other-
inventories. These inventories invariably rely on a set of behaviors characteristic of
the malady. In the context of alcoholism, binge drinking would be a behavior with
extreme consequences that is often used in a self-diagnostic inventory together with
a behavior such as daily drinking which has less extreme consequences. Behaviors
associated with alcoholism are unambiguous, but differ in terms of their extremity.
On the other hand, behaviors associated with ADD differ in terms of their ambigu-
ity, but are less extreme. Examining whether the effect of including/excluding
different behaviors from self-diagnosis inventories replicates to these contexts
would help disentangle whether it is the extremity of the consequences of a behav-
ior, its lower likelihood of being engaged in, or its relatively lower ambiguity that
affects perceptions of risk.
Other diseases also rely on self-diagnosis at an initial stage. For example, the
symptoms of Type I diabetes include “increased thirst and urination, constant
hunger, weight loss, blurred vision, and extreme tiredness.” Note that these