The influence of message framing, intention to quit smoking, and nicotine dependence on the persuasiveness of smoking cessation messages
Introduction
Following Kahneman and Tversky's prospect theory (1981), a substantial amount of literature has explored the different effects of positively and negatively framed health communication messages on individual decision making. A framing approach to studying health communication effects focuses on whether people respond better to messages that emphasize the benefits of adopting certain health behavior (positive frame) or to messages that emphasize the disadvantages of not adopting this health behavior (negative frame). Examples of health behavior on which framing effects have been studied include detection of breast cancer in an early stage (Banks et al., 1995, Meyerowitz and Chaiken, 1987, Rothman et al., 1993), smoking cessation (Wilson et al., 1990, Wong and McMurray, 2002), healthy nutrition (Assema et al., 2001, Donovan and Jallah, 1999), regular exercise (Jones et al., 2003, Robberson and Rogers, 1988), use of sunscreen to prevent skin cancer (Detweiler et al., 1999, Rothman et al., 1993), clinical skin examination to detect skin cancer in an early stage (Rothman, Salovey, Antone, et al., 1993), use of infant car restraints (Treiber, 1986), and diagnostic blood tests to determine cholesterol level (Maheswaran & Meyers-Levy, 1990). Despite the amount of research on the effect of framing in a health communication context, there has not been an unequivocal conclusion that one type of frame is better than the other. Some studies have found positively framed messages to be more persuasive, while other studies have found the opposite.
Two main reasons have been suggested for these conflicting findings. First, the relative effectiveness of positively or negatively framed messages may depend in part on the type of behavior (Rothman and Salovey, 1997, Rothman et al., 1993, Rothman et al., 1993). If the behavior has an illness-detecting function (e.g., doing breast self-examination or using mammography to detect breast cancer) a negative frame is preferable, but if the behavior is illness-preventing (e.g., smoking cessation, sunscreen use) a positive frame works best. Rothman and Salovey (1997) put forward that this is in line with prospect theory (Kahneman & Tversky, 1981), which assumes that people are more risk-seeking (i.e., taking the risk to find out that one has cancer) when confronted with losses (i.e., negative frame that communicates negative consequences), instead of gains (i.e., positive frame that communicates positive consequences), but more risk-averse (i.e., using sunscreen to avoid the risk of getting skin cancer) when confronted with gains instead of losses.
The second reason that has been suggested for the conflicting findings is the degree to which people engage in detailed message processing. In line with studies on modes of information processing and persuasion (e.g., Petty, Cacioppo, & Schumann, 1983), negative frames are considered to be more persuasive in high elaboration situations, because negative information receives greater weight during processing than positive information (Maheswaran & Meyers-Levy, 1990). When processing is minimal, on the other hand, it is believed that positively valenced peripheral cues will attract more attention and consequently be more persuasive. In line with this, studies exploring the influence of message processing on the effect of framing have shown that negative frames are more persuasive when people process health communication messages in depth, while positive frames are better when people process the message superficially (e.g., Block and Keller, 1995, Maheswaran and Meyers-Levy, 1990, Donovan and Jallah, 1999).
We postulate that message processing also accounts for the different outcomes for preventive and detective health behaviors. Messages promoting preventive behaviors probably elicit less message processing than messages for detective behaviors, and therefore a positive message frame will generally work better for preventive behaviors, and a negative frame will work better for detective behaviors A first rationale for the assumption that messages for different types of behavior elicit different levels of processing can be found in the immediacy of the health threat. Preventive behaviors are conducted to avoid an illness people might (or might not) get in the future, while detective behaviors are conducted to discover an illness that may already be present. The latter would tend to cause more anxiety about one's present health status and consequently elicit more elaborate message processing.
A second reason why information on illness-detecting behaviors might elicit more elaborate message processing than information on illness-preventing behaviors can be found in people's resistance toward adopting the promoted behavior. Messages for illness-detecting behaviors advise people to take new actions (e.g., breast self-examination, clinical skin examination, diagnostic blood tests), while messages for illness-preventing behaviors predominantly advise people to change existing health-damaging behaviors (e.g., stop smoking, eat more healthily, exercise more regularly). People are frequently unwilling to change existing behaviors, and therefore they tend to pay less attention to information that advises them to do so, especially when they do not perceive their current behavior as damaging to their health. On the other hand, information that urges people to take new actions will generate less resistance, and consequently more elaboration.
In the field of smoking cessation, several studies have shown that the majority of smokers do not intend to quit smoking in the near future (e.g., Engels et al., 1998, Hennrikus et al., 1995, John et al., 2003). Additionally, a substantial number of smokers have a low nicotine dependence, with a low intensity and frequency of smoking, which is positively related to a lower perceived health risk. Smokers who do not intend to quit smoking and those with a low nicotine dependence will not be strongly motivated to process information about smoking cessation. Together, these groups encompass the majority of smokers, and it is therefore likely that, in general, a positive frame will outperform a negative one, as seen in studies which contrast the effects of differently framed smoking cessation messages (e.g., Schneider et al., 2001, Steward et al., 2003). However, this does not mean that a positive frame will outperform a negative frame for each smoker. It is conceivable that a negative frame works better for people who have a strong intention to quit smoking, as well as a high nicotine dependence. After all, these people are already willing to change their behavior and are the ones most at risk, and therefore they will be the most motivated to process smoking cessation messages.
In this study, we examine the different influence of negatively and positively framed smoking cessation messages on smokers with varying levels of nicotine dependence and varying intentions to quit smoking. We study the influence of message framing on respondents' intentions to quit smoking, their attitude toward smoking cessation, and their perceived behavioral control, in other words, their belief whether they actually can quit smoking. Attitude and perceived behavioral control have frequently been mentioned in the literature as the most important predictors of intention to quit smoking. Subsequently, intention to quit is often found to be the most important predictor for future quit attempts (e.g., Hu and Lanese, 1998, Norman et al., 1999).
Section snippets
Hypotheses
We postulate that the relative effectiveness of a positive message frame and a negative message frame depends on the level of quit intention and level of nicotine dependence. Both these characteristics will influence the persuasiveness of message framing on intention to quit, attitude toward smoking cessation, and perceived behavioral control. The rationale is that these two variables influence how motivated the message receiver will be to actively process the message. If the level of message
Participants and procedure
The study was conducted in the Netherlands. The study's sample consisted of smokers recruited from an internet panel hosted by the Amsterdam School of Communications Research, ASCoR. The members of this panel are all students of the University of Amsterdam, who have indicated their willingness to participate in online research. Joining the panel and participating in studies is voluntarily. An e-mail was sent to all panel members (N = 959), inviting them to participate in a study on smoking
Descriptive statistics and correlations
In the introduction we argued that the majority of smokers has a low intention of quitting, or a low level of nicotine dependence. We suggested that this might explain why a positive frame in smoking cessation messages has usually been found to outperform a negative frame. Table 1 shows that this expectation was confirmed by our sample. In the first wave, the mean score for intention to quit smoking was 2.85 on a scale from 1 to 7. The mean score for nicotine dependence was 1.18 on a scale from
Discussion
The rationale for this study was our contention that smokers' nicotine dependence and quit intention would moderate the persuasiveness of positively and negatively framed smoking cessation messages. This contention was based on studies on modes of information processing and persuasion (e.g., Petty et al., 1983) that have shown that positively framed information works better when people process the information peripherally, while negatively framed information is more persuasive when people
References (28)
- et al.
Antecedents of smoking cessation among adolescents: Who is motivated to change?
Preventive Medicine
(1998) - et al.
The smoking cessation process: Longitudinal observations in a working population
Preventive Medicine
(1995) - et al.
The applicability of the theory of planned behavior to the intention to quit smoking across workplaces in southern Taiwan
Addictive Behaviors
(1998) - et al.
Relation among stage of change, demographic characteristics, smoking history, and nicotine dependence in an adult German population
Preventive Medicine
(2003) - et al.
The influence of message framing on intentions to perform health behaviors
Journal of Experimental Social Psychology
(1993) - et al.
Framing of nutrition education messages in persuading consumers of the advantages of a healthy diet
Journal of Human Nutrition and Dietetics
(2001) - et al.
The effects of message framing on mammography utilization
Health Psychology
(1995) - et al.
When to accentuate the negative: The effects of perceived efficacy and message framing on intentions to perform a health-related behavior
Journal of Marketing Research
(1995) - et al.
The impact of feelings on ad-based affect and cognition
Journal of Marketing Research
(1989) - et al.
Message framing and sunscreen use: Gain-framed message motivate beach-goers
Health Psychology
(1999)