|
Article Excerpt
There is a large body of evidence showing that individuals respond differently when the same information is framed as either a gain (e.g. lives saved) or a loss (e.g. deaths; see Ferguson, Bibby, & Leaviss, 2003; Kuhberger, 1998; Kuhberger, Schulte-Mecklenbeck, & Perner, 1999; Levin, Schneider, Gaeth, 1998; Rothman & Salovey, 1997 for reviews). Within the domain of health behaviours, Rothman and Salovey have argued that the type of behaviour under consideration moderates framing effects. Initially they argued that loss frames are more effective in influencing cognitions and behaviours for detection behaviours (e.g. breast self-examination) and gain frames for prevention behaviour (e.g. using sun-block). Recently, the role played by perceived risk as a moderator of framing within a single category of behaviour, either prevention or detection, has been highlighted (Apanovitch, McCarthy, & Salovey, 2003; Rothman, Kelly, Hertel, & Salovey, 2003;). It is argued that when a behaviour is perceived as risky a loss frame advantage should be observed and for behaviours that are perceived as less risky or safe, a gain frame advantage. Here this hypothesis is tested with respect to a prevention behaviour: the male hormonal contraceptive. Unlike other prevention behaviours it is argued that this will be perceived as risky and a loss frame advantage should be observed. Finally, as the need to identify potential mediators and moderators of the relationship between frames and intentions has been stressed (Steward, Schneider, Pizarro, & Salovey, 2003), two classes of psychological model (theory of planned behaviour and stress appraisals) are explored.
The male contraceptive pill
A clinically-available hormonal male contraceptive is likely to be available in the very near future (Anderson & Baird, 2002). Recent research has reported that up to 30% of couples worldwide use a male method of contraception and that over 70% of couples agreed that men should take more responsibility for contraception (see Anderson & Baird, 2002). Moreover, given there is no alternative to vasectomy or condom use for men, the acceptability and demand of male hormonal methods is likely to be high (see Martin et al., 2000). To date, psychological research has only examined the behavioural effects of different formulations of the male hormonal contraceptive (O'Connor, Archer, Hair, & Wu, 2001, 2002; O'Connor, Archer, & Wu, 2004). Little or no research has investigated the psychological factors that may be important in predicting male and female intentions to using male hormonal methods of contraception (pill or injection).
The male hormonal contraceptive was chosen for this study for four reasons. First, it is a prevention behaviour with high perceived health risk. Second, it is a new piece of biomedical technology, which is not yet available, therefore, the base rate of past behaviour is zero. Third, few studies in the framing literature have focused on new biomedical technology (see Kuhberger, 1998; Kuhberger, Schulte-Mecklenbeck, & Perner, 1999). Finally, the male contraceptive will soon be available for clinical use and information on people's psychological responses to it would be of great practical significance with respect to health promotion campaigns (cf. Anderson & Baird, 2002; Martin et al., 2000).
Frames, health behaviours and perceived risk
Prospect theory (Kahneman & Tversky, 1979, 1982) provides a useful framework to guide the predictions of the current study. Kahneman and Tversky's postulate predicts that risky behavioural choices will be more likely when information is framed in terms of relative disadvantages (or losses) compared with advantages (or gains). Underlying this prediction is the notion that exposure to persuasive messages highlighting the potential losses resultant in not performing a risky behaviour are likely to result in a shift in one's reference point from a situation of relative optimism and certainty to one of relative uncertainty and doubt. Consequently, when a message is framed as a loss, individuals are likely to encode the message as a relative loss from their original point of reference. As a result, risk seeking is likely to be enhanced, and their motivation to engage in the behaviour should be greater in order to return to their original reference point.
Rothman and Salovey (1997) developed and applied these ideas to health-related behaviours arguing that framing effects are modified by the type of behaviour (prevention or detection) due to the levels of perceived risk associated with each of these. They initially define risk as reflecting '... the subjective perception that to perform a behaviour may involve an unpleasant outcome' (p. 5). Specifically, detection behaviours (e.g. breast self-examination) are seen as risky, and therefore, loss frames (which are associated with risk seeking) are more effective. Conversely, for safer prevention behaviours (e.g. using a condom), gain frames are more effective as they emphasize certainty and risk aversion (see also Detweiler, Bedell, Salovey, Pronin, & Rothman, 1999).
However, recent theoretical and empirical work on framing effects and perceived risk has extended this original model (Apanovitch et al., 2003; Rothman et al., 2003). It is argued that even within a category of health-related behaviour (e.g. prevention behaviours), the relative effectiveness of a frame will depend on the level of perceived risk associated with that behaviour (Rothman et al., 2003). With the category of detection behaviours these authors define risk as the 'possibility of learning undesirable information about one's health' (Rothman et al., 2003). This has been operationalized in terms of the perceived uncertainty associated with test results--uncertainty about the future outcome (see Apanovitch et al., 2003; Rothman et al., 2003). Specifically, when a behaviour is perceived as having high risk (uncertainty), there should be a loss frame advantage, and for behaviours with perceived low risk (certainty), a gain frame advantage should prevail. This argument has recently received partial empirical support with respect to the uptake of HIV screening (Apanovitch et al., 2003). These latter authors demonstrated a gain frame advantage for those who perceived their test results to be less risky, but there was no significant loss frame advantage for those who perceived their results to be more risky.
These authors further argue that for prevention behaviour to be considered risky or safe depends on the extent that it is perceived as effective in maintaining health (Rothman et al., 2003). Therefore, risk in the context of this study on the male hormonal contraception (a prevention behaviour), was defined in terms of perceived health-outcome effectiveness. That is, the extent to which the male hormonal contraceptive is seen to be risky with respect to health.
The male contraceptive may be viewed as a potentially risky prevention behaviour for the following reasons. First, there may be concerns about side effects, efficacy, and unwanted pregnancy associated with using hormonal male contraception. Second, new advances in bio-technology and medicine are often seen as risky (Gaskell, Bauer, Durant, & Allum, 1999). Third, there is empirical evidence that hormonal contraception is viewed as risky (in terms of health-related side-effects) for people who are about to use it for the first time (Emmett & Ferguson, 1999). Therefore, it is hypothesized that (a) compared with other prevention behaviours, the male hormonal contraceptive would be viewed as risky, and (b) for this prevention behaviour, a loss frame rather than a gain frame will be observed.
Frame, intentions and the theory of planned behaviour (TPB)
The framing literature has consistently shown that frames influence intentions to behave, and there is evidence that frames also influence other parameters of the theory of planned behaviour (TPB) (see Levin et al., 1998 for a review). Therefore, it may be the case that attitudes, subjective norm, and/or perceived behavioural control mediate the relationship between frame and intentions. To the authors' knowledge, the potential mediating role of all TPB variables with respect to frame and intentions has not been examined before. Previous studies have investigated the mediating effects of attitudes and intentions to explain effects of frame on behaviour, but they have failed to demonstrate any mediation effects (Banks et al., 1995). However, these authors did show that framing effects still accounted for additional variance once social cognitive variables had...
|
