Media Message Judgment and Decision Making: Severe Acute Respiratory Syndrome (SARS)
Wanda Siu, Ph.D
School of Journalism and Communication
The Chinese University of Hong Kong
Online Publication Date: March 13, 2007)
Journal of Media Psychology, Volume 12, No. 1, Winter 2007
Severe acute respiratory syndrome (SARS) is a highly contagious virus that presents a challenge to global health. A preventive task that can be accomplished through communication is to design effective media messages that facilitate the adoption of preventive behavior. In this study, respondents first completed a survey (response-scale alternatives) rating their susceptibilities to illness and then were presented with a public health message on SARS. Results showed that the survey increased the risk that was perceived to exist and affected subsequent judgment of a public health message.
Keywords: judgment, intention, decision making, SARS, response alternatives, message framing
Severe acute respiratory syndrome (SARS) is a highly contagious virus that is spread by close contact. It presents a challenge to both medicine and health communication. The first index case appeared in southern China in November 2002, and the virus was quickly dispersed through international air travel. The World Health Organization declared a global emergency after the outbreak of the virus in Southeast Asia, which saw cases in Hong Kong, Vietnam, Singapore, and Taiwan, with other cases reported in Canada. Many studies of SARS have focused on the mode of disease transmission, and have highlighted the role of preventive measures in minimizing exposure to the virus. Well-documented preventive measures include wearing a surgical mask, washing hands frequently, and avoiding public places. A preventive task that can be accomplished through communication is to design effective messages that facilitate the adoption of preventive behavior.
Research in health advertising has found that the context of information (the message frame) either increases or decreases the risk that is perceived to exist (Rothman & Salovey, 1997, 1999). For example, a positively framed advertising message emphasizes the benefits of adopting recommended health behavior (i.e. physical activity reduces the chance of heart disease). A negatively framed message emphasizes the costs of failing to adopt recommended health behavior (i.e. a lack of physical activity increases the chance of heart disease). Also, media research has shown that response alternatives are treated as information that guides the interpretation of questions (Gaskell, O’ Muircheartaigh, & Wright, 1994). However, the influence of message frames and response alternatives has generally been explored independently. This research examines these lines of research simultaneously in the context of decision making on an acute infection.
Current approaches to the study of health advertising pay much attention to the relationship between risk perception and health behavior (Salovey, Rothman, & Rodin, 1998; Weinstein, 1993). The underlying premise of these studies is that people who recognize that their behavior is risky and acknowledge their vulnerability are more likely to adopt recommended health behavior (Rothman, Haddock, & Schwarz, 2001). Research has shown that personal testimonies are an effective means of enhancing people’s awareness of their own risky behavior (Gerrand, Gibbons, Benthin, & Hessling, 1997; Sherman, Nelson, & Steele, 2000). For example, heavy coffee drinkers were enrolled in an experimental session during which they acknowledged the harmful effects of coffee. In a three-month follow up survey, study participants were found to reduce their coffee consumption.
However, it is difficult to make people acknowledge their vulnerability to illness or disease. Research has shown that people engage in a variety of cognitive strategies to deal with unfavorable health information (Ditto & Lopez, 1992; Croyle, Sun, & Louie, 1993). In one study of pancreatic disease, participants were tested for the presence or absence of thioamine acetylase enzyme (TAA), which is related to pancreatic disorders (Ditto & Lopez, 1992). Some of the participants were led to believe that the presence of TAA had positive consequences for health, whereas others were led to believe that the presence of TAA had negative consequences for health. The results showed that the participants who were told that the diagnosis was bad generated alternative explanations for the unwanted outcome, such as questioning the accuracy of the test. Overall, the participants made use of differential decision criteria by displaying skepticism about information that was inconsistent with their preferences but readily accepting information that was consistent with their preferences. This strategy helps to mitigate the threat of unwanted health information by downplaying the validity of medical tests and the severity of the threats that are associated with an illness.
There is much debate about what motivates people to use differential decision-making criteria. People often attempt to see themselves in a positive light by discounting their own vulnerability (Taylor, Kemeny, Reed, Bower, & Gruenewald, 2000), and in fact a sense of optimism and personal control is associated with better mental health (Aspinwall & Taylor, 1997). The acceptance of one’s vulnerability may become a threat to one’s self-esteem. For example, because screening for breast cancer involves an uncertain outcome, the risk of detecting cancer has been found to be a deterrent to self-examination among women (Meyerowitz & Chaiken, 1987).
Message Framing. Message framing studies suggest that the presentation of problems as resulting in either a gain or a loss affects people’s health decisions (Kahneman & Tversky, 1979, 1982, 1984). Interventions that highlight gain and loss have the same objective outcome, or the same probability of a given outcome. In the main, people will opt for a sure or probable gain when an intervention demonstrates salient benefits, but are more likely to opt for a probable loss when the intervention emphasizes cost. For example, in one study, participants were given information concerning an epidemic that would affect 600 individuals, and were offered the choice of a certain gain (“if program A is adopted, then 200 people will be saved”) and an uncertain gain (“if program B is adopted, then there is a one-third probability that 600 people will be saved and a two-thirds probability that no-one will be saved”). The participants preferred a sure gain to an uncertain gain, even though the expectation in both scenarios was the same (1/3 x 600).
In another scenario in the same study in which the intervention was presented as a loss, the participants preferred the risky option (“if program D is adopted, then there is a one-third probability that nobody will die and two-thirds probability that 600 people will die”) over the sure loss option (“if program C is adopted, then 400 people will die”). The frame was thus changed for both the gain and loss prospects. The preference for a sure gain and under a gain-framed condition and a risky option under a loss-framed condition, given equivalent expected utility, indicates that behavioral decisions are affected by the framing of choices (Kahneman & Tversky, 1979, 1982). In studies of health behavior, researchers have found that people’s willingness to take risks depends on the way in which a message is framed and the function of the health behavior that is being advocated (Rothman & Salovey, 1997). There are two main types of message frames (gain and loss) and two main types of health behavior (preventive and detective). Preventive behavior consists of actions that are undertaken by the healthy to reduce health risks. A gain frame that emphasizes the benefits of adopting a certain type of health behavior (for example, wearing a surgical mask to prevent infection with SARS) has been shown in a series of studies to be more persuasive in encouraging the adoption of preventive behavior. Some other examples are using sunscreen (Rothman, Salovey, Atone, Keough, & Martin, 1993), wearing a seatbelt (Christophersen & Gyulay, 1981), and engaging in physical activity (Robberson & Rogers, 1988; Hsiao, 2002; Jones, Sinclair, & Courneya, 2003). Detective behavior consists of actions that are undertaken by the healthy to detect illness.
People who display detective behavior run the risk of discovering bad news about their health (a loss, such as finding a tumor). A loss frame is found to be better for motivating people to engage in detective behavior, such as undergoing a mammography, performing self-examination (Meyerowitz & Chaiken, 1987; Banks, Salovey, Greener, Rothman, Moyer, Beauvais, & Epel, 1995; Levin & Chapman, 1993), and having an HIV test (Apanovitch, McCarthy, & Salovey, 2002).
Based on the premise that a gain-framed message is more persuasive in the promotion of preventive behavior, the framing hypothesis that is tested in this study is that a message that emphasizes the benefits of undertaking preventive measures against SARS will attain more agreement than a message that emphasizes the costs of not undertaking preventive measures against SARS. People may perceive preventive behavior such as wearing a surgical mask and washing hands to be an inconvenience. However, engaging in preventive behavior promotes a desirable health outcome that wards off the danger of being infected by a deadly virus. Therefore, highlighting the benefits of preventive measures will be more persuasive in health promotion efforts to combat SARS. The following hypothesis is therefore proposed.
Hypothesis 1: A gain-framed message is associated with more agreement with the message than is a loss-framed message. Emphasizing the benefits of preventive health behavior is conducive to the promotion of behavior to combat SARS because risk perception is largely a context-specific judgment process (Fischhoff, 1991). People seldom consider all of the available information before arriving at a decision, and utilize decision aids to simplify the task of judgment (Tversky, 1988). Task and context effects often become the cues for simplifying the judgment process (Payne, Bettman, & Johnson, 1993). For example, the emphasis of certain salient concepts (the benefits of preventive behavior) will make the benefits of engaging in preventive behavior more accessible in a subsequent risk judgment.
Salient concepts are more accessible in the memory (Van der Pligt & Eiser, 1984), and thus have a greater impact on the judgment of risk. For example, if people read a message that emphasizes that wearing a surgical mask prevents SARS infection, then this will render the advantages of wearing a surgical mask more accessible to them in making a judgment about whether to wear a mask at a later time. Alternatively, if people read a message that highlights that failure to wear a surgical mask increases the risk of contracting SARS, then this will render the disadvantages of not wearing a surgical mask more accessible in a later judgment decision. This shows that the context of information (the message frame) either increases or decreases the risk that is perceived to exist.
Response Alternatives. The way in which context effects, such as rating scales, affect people’s perception of a question and their judgment is still debatable. Research has shown that changes in the numeric value of a rating scale affect people’s judgment. For example, when asked to report how successful they were in life on a rating scale that ranged from -5 to 5, 34% of respondents reported numbers below the midpoint. However, only 13% of respondents did so when the scale ranged from 0 to 10 (Schwarz, Bless, Bohner, Harlacher, & Kellenbenz, 1991). This shows that response alternatives affect subsequent judgment.
The context effects provided by response alternatives are of theoretical interest as they might shape respondents’ perception of public service announcement and become an important antecedent in persuasion. Response alternatives are often treated as information that guides the interpretation of questions. In a number of split ballot experiments on national survey data, researchers found that people who were presented with a low-frequency scale reported fewer instances of the target behavior than respondents who were presented with a high-frequency scale. For example, a low-frequency scale may range from “never” to “more than twice a month,” or a high-frequency scale that ranges from “twice a month or less,” to “several times a day.”
The context of information, including the context of a question, affects both the way in which respondents interpret what the question is asking and the answer that they give (Schwarz & Oyserman, 2001). Furthermore, response scales provide participants with information about their own behavior and the behavior of the average person (Rothman, 1997; Schwarz & Oyserman, 2001). A high-frequency scale depicts more frequent instances of a certain type of behavior in the population, and respondents in studies that use such scales are more likely to site their responses at the lower end of the scale to reflect that their behavior falls at the lower end of the distribution. Conversely, a low-frequency scale suggests that a given type of behavior occurs less frequently, and thus respondents are more likely to report behavior that falls at the higher end of the scale. It follows that the context of a question, such as the response alternatives, may be utilized to enhance the salience of risk information and therefore increase or decrease the subsequent level of risk that is perceived by respondents. For example, 62 percent of respondents in a study reported having psychosomatic symptoms more than twice a month or less on a high-frequency scale, but only 39% reported having psychosomatic symptoms more than twice a month or less on a low-frequency scale (Schwarz & Scheuring, 1992).
A study of disease perception showed that medical students and physicians were affected by response scale alternatives in their assessments of patients (Schwarz, Bless, Bohner, Harlacher, & Kellenbenz, 1991). On a low-frequency scale they rated the illness of patients who reported that a physical symptom occurred twice a week to be more severe – which meant that they were more likely to receive medical attention – than when a high-frequency scale was used. This shows that a response scale shapes people’s perceptions by providing a standard of comparison for risk assessment, because people assume that values at either end of the scale denote extremes of behavior in the population distribution, whereas values in the middle of the scale represent the behavior of the average person (Schwarz & Oyserman, 2001).
Based on the assumption that people use the range of response alternatives to estimate the frequency with which they manifest a type of behavior, it is proposed that the perception of risk can be altered by the rating scale that is provided. For example, in a study of perceptions of the risk of contracting HIV among college students, the participants were first asked to report the number of sexual partners that they had had on either a low- or a high-frequency scale and then to report their assessment of their risk of contracting HIV and their use of condoms. The participants stated their risk to be higher when their own placement on a low-frequency scale was relatively higher than that of the average college student. In contrast, the participants judged their risk to be lower when their own placement on a high-frequency scale was relatively lower than that of the average college student (Rothman, Haddock, & Schwarz, 2001). There seems to be two potential pieces of information derived from available response alternatives: 1) the fact that one has fewer partners than the average, and 2) if one has fewer partners than average it means that others have had more partners. Because these people would be at high risk, the risk of contracting HIV would also be high. Therefore, future behavior would be differentially shaped following the information response alternatives provide on others’ stance on a particular behavior. With any illness of which the risk would go up with human interaction, such as SARS and HIV, this is an important research issue addressed in this study.
A high-frequency scale informs respondents that their behavior is less common than that of the average person, which may make them feel more secure about their health condition. One should note that manipulating scale formats should not change whether or not one is above or below the average. That should be an immutable fact. It is the emphasis conveyed by the response scale, which affects a psychological set. It is a case of the glass being half full or half empty. Your father has a 40% chance of dying from the operation vs your father has a 60% chance of surviving from the operation. Whether someone’s score is at a low end of a high frequency scale or high end of a low frequency scale shouldn’t affect whether that person is factually above or below the average. Affected by the context effects of a response scale, when people give responses that fall at the higher end of a low-frequency scale, they may draw the conclusion that their personal risk of contracting an illness are higher than that of the average person. They may therefore feel more vulnerable to illness and become more receptive to health promotion messages. This analysis leads to the following hypothesis.
Hypothesis 2: A low-frequency scale is associated with more agreement with the message than is a high-frequency scale.
The Possible Integration of Response Alternatives and the Message Framing
Based on the premise that the order of questions, or the preceding question, has an effect on the subsequent question, many studies have indicated that the preceding question activates relevant concepts in the minds of respondents and thus affects the way in which they answer subsequent questions (Bishop, Oldendick, & Tuchfarber, 1986; Gaskell, O’Muircheartaigh, & Wright, 1994; Schuman & Presser, 1981; Strack, Schwarz, & Wanke, 1991). For example, in a national survey, respondents were asked how annoyed they felt about an advertisement. Those respondents who were first asked about their opinions on the programming time that was allotted to news, current affairs, sports, comedy, and information programs before being asked the target question concerning annoyance. Respondents reported significantly higher levels of annoyance with the advertisement than those respondents who were not asked to indicate their opinions about programming. This indicates that previous questions enhance the accessibility of relevant concepts that are then incorporated into the answer to the target question.
Research has also shown that media messages, such as political advertisements, activate particular concepts in the memory and affect subsequent interpretations (Iyengar & Kinder, 1990; Wyer & Ottati, 1993). This accessibility hypothesis has been well tested and has received much support in social cognition studies (Fiske & Taylor, 1991; Price & Tewksbury, 1997). For example, it has been found that people tend to use shortcuts in information processing, rather than carefully considering all of the relevant information (Tversky & Kahneman, 1973). This means that an audience bases its judgment of a media message on temporarily accessible information (Zaller, 1992; Higgins, 1996; Wyer & Scrull, 1989). The ease with which concepts come to mind may account for systematic biases in the probability estimates of events and occurrences, such as judging the likelihood that a target person has a particular occupation by the degree to which that person’s appearance fits the occupational stereotype (Tversky & Kahneman, 1973).
This analysis intimates that previous questions enhance the accessibility of relevant concepts in the memory and affect the responses to subsequent questions (Rothman & Schwarz, 1998; Schwarz, 1991; Weinstein & Klein, 1995), which means that concepts that have recently been activated will remain temporarily accessible and thus affect subsequent judgment. In this study, it is assumed that respondents will draw upon a subset of information that is made temporarily accessible by the response alternatives when they are asked to make a subsequent judgment of a promotional message about SARS.
Studies have shown that response alternatives accentuate the perception of risk (Colasanto, Singer, & Rogers, 1992; Rothman, 1997; Schwarz & Oyserman, 2001). This implies that people draw risk information from response alternatives, which makes risk-related information temporarily more accessible for the judgment task. Research in social cognition has shown that when people were asked to make judgment of a particular target, their judgment is affected by information that is most accessible at the time of judgment (Higgins, 1996 & Schwarz, 1995). An example would be the information that comes to the respondent’s mind in the preceding question. In the present study, response alternatives that precede the framing message may enhance respondent’s perceived disease vulnerability.
Feeling more vulnerable to a possible illness might move people’s psychological reference point into the possibility of loss (the possibility of illness), and inclines them to restore this reference point to a positive domain because of the inherent desire to see oneself in a positive light (Taylor, Kemeny, Reed, Bower, & Gruenewald, 2000). A message that emphasizes the benefits of SARS prevention will move the psychological reference point into the positive domain. Thus, a low-frequency response alternative might be expected to enhance feelings of vulnerability and improve the reception of a gain-framed message on SARS prevention more than does a cost-framed message. Conversely, a high-frequency response alternative might be expected to improve the reception of a cost-framed message. This is quite counter-intuitive: respondents in the high-frequency scale condition are first made to feel less vulnerable to illness, thus anchoring their psychological reference point to the positive domain. The subsequent reception of a cost-framed message provides negative information on their health prospects, which makes them more alert to possible illness. As people are conservative in gain prospects, respondents who receive a high-frequency response alternative may try to maintain their positive health prospects by becoming more receptive to the cost-framed message.
Finally, this study also explores whether message judgment mediates the effects of response alternatives and message frame on behavioral intention, as research has generally shown that judgment is positively related to intention (Petty, Priester, & Wegener, 1994). Specifically, a low-frequency scale condition is associated with better message judgment that affects behavioral intention. Moreover, a gain-framed message is associated with better message judgment that affects intention to obtain more information on SARS prevention. The following hypotheses are proposed.
Hypothesis 3: A gain-framed message is more positively related to message judgment than a cost-framed message in the low-frequency condition and a cost-framed message is more positively related to message judgment than a gain-framed message in the high-frequency condition.
Hypothesis 4: A gain-framed message is more positively related to behavioral intention than a cost-framed message in the low-frequency condition and a cost-framed message is more positively related to behavioral intention than a gain-framed message in the high-frequency condition.
Hypothesis 5: Message judgment mediates the effects of response alternatives and frame on behavioral intention. A low-frequency scale is more positively related to message judgment that leads to behavioral intention. A gain-framed message is more positively related to message judgment that leads to behavioral intention.
The participants in this experiment were 340 undergraduate students of a Mid-Western university who participated voluntarily and anonymously. The sample consisted of all students who were enrolled in undergraduate communication courses and whose ages ranged from 18 to 23. Students received a course credit for their participation.
The participants were informed that they were taking part in a communications study on their perceptions of health issues in the lab They were given a questionnaire instrument that contained the following sections: response scales, a copy of a public service announcement (PSA), and the PSA evaluations of SARS. The distribution of the questionnaires was randomized. A 2 x 2 design was used. The participants were instructed to complete the questionnaire individually, and were told that their participation was completely voluntary and anonymous. They were given 15 minutes to complete the questionnaire and were debriefed at the end of the exercise.
Response Scale Condition.
A between-subjects variable with two levels – a high-frequency response scale (n = 170) and a low-frequency response scale (n = 170) – was used. The high-frequency scale ranged from experiencing a physical symptom “less than twice a week” to “about once every 24 hours” to “more often.” The low-frequency scale ranged from experiencing a physical symptom “less than once a month” to “about twice a week” to “more often” (Schwarz, Bless, Bohner, Harlacher, & Kellenbenz, 1991, p. 43). The participants were asked to indicate how often they experienced physical symptoms such as headaches, fever, indigestion, sore throat, and nasal congestion on average. These symptoms were chosen as they were explicitly linked to SARS.
In line with the framing postulate of prospect theory, the frame was operationalized as the benefits of adopting a certain preventive measure and the costs of not adopting this measure. Information about SARS was excerpted from the transcript of a television interview with Dr. Justin Wu, a medical doctor in Hong Kong who was in the frontline of the fight against the virus.
The frame was a between-subjects variable. After the response scale manipulation, some of the participants read and evaluated the gain-framed PSA and the others read and evaluated the loss-framed PSA. The gain- and loss-framed PSAs presented the same information, but in different ways. The gain-framed PSA described the benefits of a certain preventive measure (wearing a surgical mask to reduce the risk of contracting SARS), whereas the loss-framed PSA described the same information but was reworded to emphasize the costs of non-adoption, rather than the benefits of adoption (not wearing a surgical mask increases the risk of contracting SARS).
Evaluation of Messages about SARS. The message evaluation was measured by using an eight-point scale that ranged from “highly disagree” (1) to “highly agree” (8), with higher values indicating a more positive message evaluation. There were eight items to measure message judgment (convincing, important, attention grabbing, interesting, relevant, informative, accurate, and credible). Confirmatory factor analysis showed that these eight items formed a single factor of message judgment that accounted for 54 percent of the total variance, with α = 0.87.
Behavioral Intention. Intention was operationalized as seeking further information after reading the PSA and it was measured by using a 8-point scale ranging from Highly Disagree (1) to Highly Agree (8). The PSA include a phrase to encourage readers to obtain more information on SARS: “So, protect yourself from SARS by talking to your health care professionals and clicking the following website http://www.health.org”. Three items were included to assess intention: will seek more information, click on the internet and talk to health care professionals. Confirmatory factor analysis showed that the items on intention formed a single factor (will seek more information, click on internet, talk to health care professionals), with a Cronbach’s alpha of 0.85, accounting for 77 percent of the total variance. A composite index on behavioral intention was thus formed by averaging these three items.
Analysis of variance revealed that the response-scale manipulation affected study participant's risk perceptions (F (1, 332) = 4.2, p < 0.05). Participants in the low- frequency group perceived greater risk (M = 4.2, SD = 0.1) than those of the high-frequency group (M = 3.5, SD = 0.2).
Hypothesis 1: A gain-framed message is associated with more agreement with the message than is a loss-framed message.
As predicted, the analysis of variance showed that the gain-framed PSA was more persuasive (F (1, 332) = 12, p = 0.001, η2 = .1). The participants in the gain-framed condition (M = 5.4, SD = 0.1) assigned a higher rating to the PSA than the participants in the loss-framed condition (M = 3.6, SD = 0.2), which confirms Hypothesis 1.
Hypothesis 2: A low-frequency scale is associated with more agreement with the message than is a high-frequency scale.
As predicted, those participants who responded using the low-frequency scale made a more positive judgment of the message (M = 4.0, SD = 0.2) than those who answered using the high-frequency scale (M = 2.9, SD = 0.1), with (F (1, 332) = 4.2, p = 0.05, η2 = .1), which supports Hypothesis 2.
Hypothesis 3: A gain-framed message is more positively related to message judgment than a loss-framed message in the low-frequency condition and a loss-framed message is more positively related to message judgment than a gain-framed message in the high-frequency condition.
The influence of the response scale on message judgment was reflected by the significant interaction of the response scale manipulation and the message frame, with F (1, 332) = 6.9, p < 0.05, η2 = .1, which supports Hypothesis 3. The participants in the low-frequency/gain-framed condition gave the PSA a higher rating (M = 4.5, SD = 0.1, n = 96) that was significantly different from the low-frequency/loss-framed (M = 2.5, SD = 0.1, n = 73), the high-frequency/gain-framed (M = 2.1, SD = 0.1, n = 84), and the high-frequency/loss-framed (M = 4.1, SD = 0.1, n = 89) conditions at the 0.05 level in the post hoc analysis.
Hypothesis 4: A gain-framed message is more positively related to behavioral intention than a loss-framed message in the low-frequency condition and a loss-framed message is more positively related to behavioral intention than a gain-framed message in the high-frequency condition.
As predicted, there was a significant interaction of response scale manipulation and the message frame on behavioral intention, with F (1, 332) = 5.6, p < 0.05, η2 = .1 which supports Hypothesis 4. The participants in the low-frequency/gain-framed condition indicated a greater behavioral intention (M = 5.6, SD = 0.1, n = 96) that was significantly different from the low-frequency/loss-framed (M = 5.1, SD = 0.1, n = 73) and the high-frequency/gain-framed (M = 5.1, SD = 0.1, n = 84) conditions at the 0.05 level in the post hoc analysis.
Hypothesis 5: Message judgment mediates the effects of response alternatives and frame on behavioral intention. A low-frequency scale is more positively related to message judgment that leads to behavioral intention. A gain-framed message is more positively related to message judgment that leads to behavioral intention.
Regression analysis showed that there was a positive association between message judgment and behavioral intention (β = 0.3, p < 0.001). Analysis of covariance shows that message judgment mediated the influence of a response scale (F (1, 332) = 4.2, p < 0.05) and a message frame (F (1, 332) = 5.3, p < 0.05) on study participants’ behavioral intention. Respondents in the low-frequency scale condition had better message judgment that affected their intention to obtain more information on SARS (M = 5.3, SD = 0.1) than did the high-frequency scale condition (M = 5.1, SD = 0.2). Moreover, study participants in the gain-framed condition had better message judgment that affected their intention to seek more information (M = 5.6, SD = 0.2) than did those in the cost-framed condition (M = 5.1, SD = 0.1). No other effects were significant at the 0.05 level. So, mediation effects were expressed in terms of message judgment and behavioral intentions.
During a global pandemic of a contagious disease, it is important to arouse people's awareness of the disease to encourage them to adopt preventive behavior. This study extends the work of Schwarz on response scales and explores the influence of response scale manipulation and message framing on message judgment in the context of the SARS epidemic.
Consistent with the predictions, a response scale manipulation was found to affect message judgment. In general, those participants who were asked to respond using a low-frequency scale made a more positive judgment of the message. This suggests that people are more likely to adopt preventive behavior if they feel vulnerable to the disease. In line with the message framing studies, the gain-framed PSA in this study was more persuasive. Other studies have found gain-framed messages to be more persuasive in encouraging preventive health behavior (Rothman, Salovey, Atone, Keough, & Martin, 1993).
The results of this study show that a response scale may enhance perceptions of disease susceptibility, yet individuals are more motivated to adopt preventive behavior if the health message talks about the benefits of undertaking preventive measures. This suggests that the persuasiveness of a gain-framed message lies in its ability to shape perceptions of the importance of attaining the benefits of recommended measures when preceded by the sensitization of respondents to their disease susceptibility. Alternatively, individuals who felt less vulnerable to illness were more receptive to a loss-framed message. This finding adds to the framing and decision making literature and gives insight on the communication of risk information in health advertising. It demonstrates that an effective promotional message should emphasize the salience of susceptibility to a disease and offer a promising prospect of its prevention. Campaigns should involve two methods of framing messages, one targeting those who might feel more vulnerable and those who might feel less vulnerable. This would reach a larger target audience if the dual message campaign is done either simultaneously or in seriation.
This study explores the utility of response alternatives and message frames on the judgment by college students of a PSA on SARS. Response scale manipulation provides an indirect way of persuading people to accept health recommendations, because people are more receptive to conclusions that they draw by themselves than to arguments that are provided by others (Petty, Priester, & Wegener, 1994). Self-persuasion has been shown to be an effective method of arousing attention to undesirable health information. A study on AIDS prevention showed that respondents who publicly advocated the importance of protected sex had a stronger behavioral intention of purchasing condoms (Stone, Aronson, Crain, Winslow, & Fried, 1994). Other studies on coffee drinking found that a response scale manipulation aroused the awareness of college students of the addictive effects of coffee drinking (Rothman, 1997). Those participants who were asked to respond using a scale which conveyed that they drank more coffee than the average college student indicated a greater desire to reduce their coffee consumption than those participants who were led to believe that they drank less coffee than the average student. Given the importance of improving message acceptance in health advertising, response alternatives provide a viable option to guide people’s perception of a disease make them more inclined to behavioral change.
This study is the first attempt to integrate response alternatives and message frames to probe the promotion of measures to prevent SARS. It may suffer from limited generalizability, as the sample consisted only of undergraduates. Moreover, the response scale manipulation only provided information on the health of the participants in comparison to their peers, rather than the general population. It remains uncertain whether national statistics of disease susceptibility may be a better approach to shaping perceptions of disease.
The design of the PSA was guided by message framing, but as the effects of the message frame on judgment were obtained in a U.S. college sample, it remains unclear whether similar effects would appear in the general population or in other countries. Future research on SARS prevention could focus on the message frame to study populations in other countries to shed light on any possible cross-national differences in message judgment as predicted by response scale manipulations and message frames.
Apanovitch, A. M., McCarthy, D., & Salovey, P. (2002). Using message framing to motivate HIV testing among low-income, ethnic minority women. Unpublished manuscript, Yale University.
Aspinwall, L. G., & Taylor, S. E. (1997). A stitch in time: Self-regulation and proactive coping. Psychological Bulletin, 121, 417-436.
Avendano, M. (2003). Clinical course and management of SARS in health care workers in Toronto: A case series, CMAJ, 168(13), 1649-1662.
Banks, S. M., Salovey, P., Greener, S., Rothman, A. J., Moyer, A., Beauvais, J., & Epel, E. (1995). The effects of message framing on mammography utilization. Health Psychology, 14, 178-184.
Bishop, G. F., Oldendick, R. W., & Tuchfarber, A. J. (1986). Opinions on fictitious issues: The pressure to answer survey questions. Public Opinion Quarterly, 50, 240-250.
Bukreyev, (2004). Mucosal immunisation of African green monkeys (Cercopithecus aethiops) with an attenuated parainfluenze virus expressing the SARS coronavirus spike protein for the prevention of SARS. The Lancet, 363, 2122-27.
Christophersen, E. R., & Gyulay, J. E. (1981). Parental compliance with car seat usage: A positive approach with long term follow-up. Journal of Pediatric Psychology, 6, 301-312.
Colasanto, D., Singer, E., & Rogers, T. F. (1992). Context effects on responses to questions about AIDS. Public Opinion Quarterly, 56, 515-518.
Croyle, R. T., Sun, Y. C., & Louie, D. H. (1993). Psychological minimization of cholesterol test results: Moderators of appraisal in college students and community residents. Health Psychology, 12, 503-507.
Ditto, P. H., & Lopez, D. F. (1992). Motivated skepticism: Use of differential decision criteria for preferred and nonpreferred conclusions. Journal of Personality and Social Psychology, 63, 568-584.
Drazen, J. M. (2003). Case clusters of the severe acute respiratory syndrome. The New England Journal of Medicine, 348, e6-7.
Fischhoff, B. (1991). Value elicitation: Is there anything in there? American Psychologist, 46, 835-847.
Fiske, S. T., & Taylor, S. E. (1991). Social Cognition. New York: McGraw-Hill.
Gaskell, O’Muircheartaigh, & Wright. (1994). Survey questions about the frequency of vaguely defined events. Public Opinion Quarterly, 58, 241-254.
Gerrard, M., Gibbons, F. X., Benthin, A. C., & Hessling, R. M. (1997). A longitudinal study of the reciprocal nature of risk behaviors and risk cognitions in adolescents: What you do shapes what you think and vice versa. Health Psychology, 15, 344-354.
Hawkey, P. M. (2003). Severe acute respiratory syndrome (SARS): Breath-taking progress. Journal of Medical Microbiology, 52, 609-613.
Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins, & A. W. Kruglanski (Eds), Social psychology: Handbook of basic principles (pp. 133-168). New York: Guilford.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263-291.
Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American Psychologist, 39, 341-350.
Meyerowitz, B. E., & Chaiken, S. (1987). The effect of message framing on breast self-examination attitudes, intentions and behavior. Journal of Personality and Social Psychology, 52, 500-510.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1993). The adaptive decision maker. New York: Cambridge University Press.
Peiris, J. S. M. (2003). Current concepts: The severe acute respiratory syndrome. New England Journal of Medicine, 349(25), 2431-2441.
Petty, R. E., Priester, J. R., & Wegener, D. T. (1994). Cognitive processes in attitude change. In R. S. Wyer, & T. K. Srull (Eds), Handbook of social cognition (2nd ed., Vol. 2, pp. 69-142). Hillsdale, NJ: Lawrence Erlbaum.
Price, V., & Tewksbury, D. (1997). News values and public opinion: A theoretical account of media priming and framing. In G. Barnett, & F. J. Boster (Eds), Progress in the Communication Sciences (pp. 173-212). New York: Ablex.
Rainer, T. H. (2003). Evaluation of WHO criteria for identifying patients with severe acute respiratory syndrome out of hospital: Prospective observational study. British Medical Journal, 326, 1354-1358.
Robberson, M. R., & Rogers, R. W. (1988). Beyond fear appeals: Negative and positive persuasive appeals to health and self-esteem. Journal of Applied Social Psychology, 43, 236-247.
Rothman, A. J. (1997). Influencing perceptions of personal risk: The utility of direct and indirect appeals. Paper presented at the annual meeting of the American Psychological Society, Washington DC.
Rothman, A. J., Haddock, G., & Schwarz, N. (2001). How many partners is too many? Shaping perceptions of personal vulnerability. Journal of Applied Social Psychology, 31, 2195-2214.
Rothman, A. J., Martino, S. C., Bedell, B. T., Detweiler, J. B., & Salovey, P. (1999). The systematic influence of gain-and loss-framed messages on interest in and use of different types of health behavior. Personality and Social Psychology Bulletin, 25, 1355-1369.
Rothman A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121(1), 3-19.
Rothman, A.J., Salovey, P., Antone, C., Keough, K., & Martin, C. D. (1993). The influence of message framing on intentions to perform health behaviors. Journal of Experimental Social Psychology, 29, 408-433.
Rothman, A. J., & Schwarz, N. (1998). Constructing perceptions of vulnerability: Personal relevance and the use of experiential information in health judgments. Personality and Social Psychology Bulletin, 24, 1053-1064.
Salovey, P., Rothman, A. J., & Rodin, J. (1998). Health behavior. In D. Gilbert, S. Fiske, & G. Lindzey (Eds), Handbook of social psychology (4th ed., Vol. 2, pp. 633-653). New York, NY: McGraw-Hill.
Schuman, H., & Presser, S. (1981). Questions and answers in attitude surveys. New York: Academic Press.
Schwarz, N. (1995). Social cognition: Information accessibility and use in social judgment. In E. E. Smith & D. N. Osherson (Eds.), Thinking: an invitation to cognition science (2nd ed., vol. 3, pp. 345-376). Cambridge, MA: MIT Press.
Schwarz, N., Bless, H., Bohner, G., Harlacher, U., & Kellenbenz, M. (1991). Response scales as frames of reference: The impact of frequency range on diagnostic judgements. Applied Cognitive Psychology, 5, 37-49.
Schwarz, N., & Oyserman, D. (2001). Asking questions about behavior: Cognition, communication, and questionnaire construction. American Journal of Evaluation, 22, 127-160.
Schwarz, N., & Scheuring, B. (1992). Frequency reports of psychosomatic symptoms: What respondents learn from response alternatives. Aeitschrift fur Klinische Psychologie, 22, 197-208.
Schwarz, N. & Scheuring, B. (1988). Judgments of relationship satisfaction: Inter- and intra-individual comparison strategies as a function of questionnaire structure. European Journal of Social Psychology, 18, 485-496.
Sherman, D. A. K., Nelson, L. D., & Steele, C. M. (2000). Do messages about health risks threaten the self? Increasing the acceptance of threatening health messages via self-affirmation. Personality and Social Psychology Bulletin, 26, 1046-1058.
Strack, F., Schwarz, N., & Wanke, M. (1991). Semantic and pragmatic aspects of context effects in social and psychological research. Social Cognition, 9, 111-125.
Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower, J. E., & Gruenewald, T. L. (2000). Psychological resources, positive illusions, and health. American Psychologist, 55, 99-109.
Tversky, A. (1988). Discussion. In D. E. Bell, H. Raiffa, & A. Tversky (Eds), Decision making: Descriptive, normative, and prescriptive interactions (pp. 599-612). Cambridge: Cambridge University Press.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207-232.
Van der Pligt, J., & De Vries, N. K. (1984). Dimensional salience, judgment, and attitudes. In J. R. Eiser (Ed.), Attitudinal judgment. NY: Springer-Verlag.
Weinstein, N. D. (1993). Testing four competing theories of health-protective behavior.
Weinstein, N. D., & Klein, W. M. (1995). Resistance of personal risk perceptions to debiasing interventions. Health Psychology, 14, 132-140.
Wyer, R. S., & Srull, T. K. (1989). Memory and cognition in its social context. Hillsdale, NJ: Erlbaum.
Zaller, J. R. (1992). The nature and origins of mass opinion. New York: Cambridge University Press.
Zhao, Z. (2003). Description and clinical treatment of an early outbreak of severe acute respiratory syndrome (SARS) in Guangzhou, PR China. Journal of Medical Microbiology, 52, 715-720.
Return to home page