A wide array of operationalizations of perceived vulnerability (possibility, probability, likelihood, etc.) exists,
and no one definitive measurement scale or strategy has emerged. However, a number of methods of assessment, and distinctions between these methods have emerged.
Absolute Perceived Vulnerability: Many measures
of perceived vulnerability focus on absolute measures of risk
der Pligt, 1998). Absolute measures
refer to the perceived likelihood a negative event
will occur, e.g., "How likely is it that you will
get lung cancer?", "What do you think is the risk that you
will get AIDS?" (Gerrard,
Gibbons, & Bushman, 1996; Weinstein
& Nicolich, 1993). Likert-type response scales are usually utilized
for these questions, e.g., 1= "almost certainly will not"
to 5 = "almost certainly will" (Joseph,
Montgomery, Emmons, Kirscht et al., 1987).
Many studies, however, employ scales that call for a numerical
probability estimate such as percent likelihood estimates,
e.g., "What is the likelihood that you will have an unplanned
pregnancy in the next year?"; response scale = 0 to 100%).
A serious weakness of simple absolute vulnerability questions
such as these is that they confound expectation, intentions,
and current risk behavior, creating two related problems.
First, interpretation of estimates made in response to simple
absolute assessment of vulnerability requires that knowledge
of the respondents' current behavior. For example, a low risk
estimate can be interpreted as optimistic from a heavy smoker
and accurate from a respondent who has never smoked
Lane, Gerrard, Pomery & Lautrup, 2002).
Second, respondents who are anticipating quitting a risk behavior
or increasing a precautionary behavior frequently report less
risk than their current behavior would suggest. For example,
the question "How likely is it that you will get lung cancer?"
will elicit a different response from smokers who expect to
quit smoking soon than from those who do not anticipate that
they will have the motivation or ability to quit in the foreseeable
future. Recognition of these problems has lead most researchers
to abandon simple absolute measures of personal vulnerability
and adopt conditional measures.
Conditional Perceived Vulnerability: Conditional
vulnerability measures are designed to elicit consideration
of expected or intended future behavior, thus avoiding
the confounding of expectations, intentions, and current risk
behaviors with perceptions of susceptibility (Ronis,
1992). These measures are often phrased in
the subjunctive (e.g., "Imagine that you had six bottles of
beer at a party. What is the chance that you will get sick
from the beer and throw up?"; Halpern-Felsher,
Millstein, Ellen, Adles, Tschann, & Beihll, 2001), and can include consideration of frequency
of the risk behaviors, preventive behaviors, etc. (e.g., "What
would be the likelihood of pregnancy if you had intercourse
more than 3 times per week and used no birth control method?";
& Luus, 1995). Thus, conditional measures also allow researchers
to assess risk perceptions among people who are not currently
engaging in the behavior but may do so in the future. They
can also be employed to distinguish between perceived
vulnerability when preventive action is taken and when it
is not taken (e.g., "If you brush your teeth daily,
how likely do you think it is that you will develop gum disease?"
followed by "If you do not brush your teeth daily, how likely
do you think it is that you will develop gum disease?").
Comparative Perceived Vulnerability: (see
& Weinstein, 1997). A number of
studies have indicated that when asked about their vulnerability,
respondents often make automatic comparisons of their
own health behavior and characteristics with those of others
& Weinstein, 1997). Thus, some recent
work has included both absolute perceived vulnerability measures
as well as measures of comparative (or "relative") risk (e.g.,
"Compared to others your age, how likely is it that you will
have a smoking-related illness [e.g., lung cancer) at some
time in the future?"; Gerrard,
Gibbons, Benthin & Hessling, 1996).
Affective Heuristic: Loewenstein and colleagues
proposed the risk-as-feelings hypothesis,
which suggests emotional reactions often drive risk and precautionary
behavior, and that these reactions are the result of a variety
of factors that are not necessarily associated with cognitive
evaluations of risks, e.g., the vividness of imagined negative
consequences, personal experience with outcomes (Loewenstein,
Weber, Hsee, & Welch, 2001). Similarly, Slovic and his colleagues
suggest that risk decisions stem more from how people
feel about the behavior than what they think
about the behavior (Slovic,
Finucane, Peters, & McGregor, 2003). They argue that people often refer to this
"affective pool" when making a decision because it is easier
and quicker than weighing the costs and benefits, or recalling
specific objective information.
Weinstein and colleagues have developed a 2-item scale designed
to assess this feeling component of risk perceptions ("With
no flu shot, I would feel that I'm going to get the flu this
year," and "With no flu shot, I would feel very vulnerable
to the flu"; Weinstein,
Kwitel, McCaul, Magnan, Gerrard, & Gibbons, in press).
Windschitl (2003) makes the same distinction,
but suggests a somewhat different procedure that assesses
both the cognitive evaluation and the feeling component of
risk with two separate questions:
"What is the objective likelihood that you will get skin cancer?" [with detailed instructions on how to use the numeric scale]
"You just indicated your beliefs about how objectively likely it is that you will get skin cancer. However, at a gut-level, you might feel somewhat more or less vulnerable than your response above suggests. Place a mark on the scale below to indicate how you feel about your chances of getting skin cancer."