Combining the persuasive tactics of narratives and framing to promote daily sunscreen application
Charlotte van de Wijngaert, 11775785 Master’s Thesis
Graduate School of Communication
Master’s Programme Communication Science Supervisor: Birthe Lehmann
Completed 1 July 2022
Sunscreen is not applied daily by most people, while this is a good preventative measure for skin cancer. The current study aims at convincing people through framing in narratives. It was expected that the effect of framing of the narrative on attitude and intention towards daily sunscreen application would be mediated by identification, transportation, counterarguing and perceived vulnerability to skin cancer. An experiment with 120 participants was performed.
None of the expected mediation effects were significant, except for attitude mediating the relationship between counterarguing and intention. The findings suggest that combining the persuasive tactics of framing and narratives is not effective in influencing identification, transportation, attitude and intention, and that more research should be done about counterarguing as a personal factor.
Keywords: Narrative persuasion, gain and loss framing, daily sunscreen use, health prevention
In The Netherlands alone, 82.800 people are diagnosed with skin cancer yearly (KWF, n.d.), making it the second most occurring cancer in both men and women (Integraal
Kankercentrum Nederland, n.d.), and causing both individual suffering and high costs for the healthcare system. One behavior that has shown to be effective in maintaining healthy skin and preventing skin cancer is sunscreen application (KWF, n.d.; Iannacone et al., 2014).
According to an independent study by Dr. Leenarts, a dermatologist, that was published on her blog, most Dutch people are willing to use sunscreen for going to the beach (58% for men, 75% for women), however, they are much less willing to apply sunscreen for their daily outdoor activities such as taking a walk through the city (10% for men, 23% for women), working out outside (40% for men, 61% for women) or going shopping (8% for men, 3% for women) (Leenarts, 2019). Sun damage builds up during the years, making daily sunscreen application a good preventative measure (Perugini, et al., 2019). Therefore, this study will focus on heightening the intention of applying sunscreen on a daily basis for outdoor activities.
Over the years, various skin cancer campaigns have been spread worldwide by health organizations with various strategies, such as fear appeals, expert tips and statistical
information (Vafeiadis & Shen, 2021). In The Netherlands, a campaign called “Hoe ziet je huid er écht uit?” (“What does your skin really look like?”) was launched in 2021 by the Huidfonds (Skin association) and the NVDV (Dutch association of dermatologists) with the goal of informing the public of the risks of unprotected exposure to the sun and tips on prevention measures such as adequate sunscreen use (Integraal Kankercentrum Nederland, 2021). Despite these efforts, skin cancer diagnoses are expected to increase by 5% each year (Integraal Kankercentrum Nederland, 2021).
There has been extensive research about how to persuade people into taking
preventative action for their health, examining various persuasive strategies such as framing, narrative persuasion, and fear mongering to formulate an effective message (Gallagher &
Updegraff, 2012; Graaf et al., 2016; Rothman et al., 2006; Tannenbaum et al., 2015; Witte &
Allen, 2000). Various insights about the effective use of framing and narrative persuasion have been found, as well as evidence of the efficacy of combining the two tactics in one health message (Asbeek Brusse et al., 2016). When framing a health message, the outcome can be framed as either a gain; the benefits of carrying out the desired behavior, or a loss; the negative consequences of not carrying out the desired behavior (Rothman et al., 2006). This can be applied to a narrative, which means a storyline. By combining these two persuasive tactics, two different versions of the same story can be told. A gain framed narrative
emphasizes the good outcomes of performing the desired health behavior in a storyline, while a loss framed message emphasized the negative consequences of not performing the desired behavior in a storyline.
To test whether and how combining framing and narratives structures is effective for the target behavior of daily sunscreen use, the mechanisms and variables are adapted in order to be applicable for sunscreen use. The following research question will be answered:
“To what extent does exposure to a gain-framed narrative (compared to a loss-framed narrative) cause a more positive attitude and a higher intention towards daily sunscreen application, and to what extent is this effect mediated by identification, transportation, counterarguing and perceived vulnerability?”.
The theoretical model used in Asbeek Brusse et al. (2016) looks promising for changing health behavior, but not all expected results were found, as no significant effects were found
for transportation and identification as mediators. Therefore, this study aims at improving the methodology by making the stimulus material more realistic. Based on the Entertainment Overcoming Resistance Model (EORM) by Moyer-Gusé (2008), perceived vulnerability will also be incorporated, as it is described in this study that higher identification with the
character leads to higher perceived vulnerability. Therefore, incorporating this factor into the model may give more insights into the mechanisms of framing the narrative. Based on Moyer- Gusé (2008), the order of the mediating variables is adjusted in the current study. In the research model by Asbeek Brusse et al. (2016), identification, transportation and
counterarguing were three separate factors expected to mediate the relationship between the frame of the narrative and the attitude towards the behavior. In the EORM, it is suggested that one way to lower resistance towards a persuasive attempt, and thereby counterarguing and to heighten perceived vulnerability towards negative consequences, is by having more
identification with the character in a story. Therefore, based on EORM by Moyer-Gusé (2008), identification could be a factor that mediates the relationship between type of framing and counterarguing. Therefore, studying whether identification indeed mediates the
relationship between the type of framing and counterarguing in narrative structures may give new insights into how the mechanisms of the relationship between framing of the narrative and the attitude towards the behavior work.
In the study by Asbeek Brusse et al. (2016), the effects of gain and loss framing in
entertainment-education (E-E) on attitude towards drunk cycling and the intention of drunk cycling are examined, taking the mediating factors of identification, transportation and counterarguing into account. Participants were presented with either a gain- or a loss-framed narrative in the form of a cartoon. In the gain frame condition, a man is shown who decides to
not cycle home drunk after a night of drinking but walks instead and arrives home safely. In the loss frame condition, the same story is shown, except in this version, the man does cycle home drunk and causes an accident. In this study, no significant mediating effects of
transportation and identification were found. Only a direct effect of framing of the narrative was found for attitude, not for intention. Counterarguing was found to be a significant mediating factor in the effect of framing of the narrative on attitude towards drunk cycling, which resulted in a higher intention to refrain from drunk cycling.
Framing & Prospect Theory
Framing is increasing the salience of certain elements of a message (Rothman & Salovey, 1997). The outcome of following the desired behavior can be framed either as a gain, or the outcome of not following the desired behavior can be framed as a loss (Rothman et al., 2006).
Based on the literature, it is known that gain frames are more effective when promoting prevention behaviors, while loss frames are more effective when promoting detection behaviors (Rothman et al., 2006; Gallagher & Updegraff, 2011).
The basis of this mechanism lies in Prospect Theory (Kahneman & Tversky, 1981) which shows that losses loom larger than gains, which means that people are willing to take more risk when it means they have a chance to prevent a loss, compared to risk they are willing to take when they have a chance to gain something. Detection behaviors such as going for a scan or a check-up at the doctor are considered riskier, as there is always a chance of finding an unfortunate outcome. By communicating the loss, people become more risk seeking and are therefore more willing to perform the riskier detection behavior. Prevention behaviors such as brushing your teeth or working out regularly are considered low risk, as performing them will most likely have no negative consequences. When communicating a gain on the other hand, people become more risk averse and are therefore more willing to perform the less risky prevention behavior. This effect was found for the behavior of
sunscreen application in a field-experiment by Detweiler et al. (1999), in which beachgoers who read a gain-framed message requested more sunscreen and had a higher intention of applying sunscreen than people who read a loss-framed message. However, contradicting results were found as well, as in O’Keefe and Wu (2012), no significant differences were found in a meta-analysis concerning the persuasiveness of gain- and loss-frames of skin cancer prevention messages.
Narratives can be defined as “any cohesive and coherent story with an identifiable beginning, middle, and end that provides information about scene, characters, and conflict; raises
unanswered questions or unresolved conflict; and provides resolution” (Hinyard & Kreuter, 2007, p. 778). According to Moyer-Gusé (2008), E-E consists of prosocial messages that are embedded into popular entertainment media content. One example of E-E is a cartoon with a health message, or a tv-series in which health information is provided. Using E-E makes it possible to incorporate an educational message into entertaining media content to raise awareness, increase knowledge, create favorable attitudes about a certain behavior and motivate people to take socially responsive action (Singhal & Rogers, 1999). Part of E-E is often the use of a narrative structure, in which storytelling is used to attract the audience’s attention (Slater & Rouner, 2002). In Appel and Richter (2007), it is shown that factional information embedded in fictional stories caused substantial changes to people’s real-world beliefs, with the effects being even stronger after two weeks. Even though E-E is a persuasive tactic in itself, different persuasive tactics can be embedded in the E-E (Asbeek Brusse et al., 2016). For this study, the persuasive tactic of framing of the narrative is examined.
Narratives for overcoming resistance
Applying sunscreen is considered a prevention measure, as consistent use results in lower chances of getting skin cancer in the future and maintaining a good condition of the skin (Iannacone et al., 2014). Based on the definitions above, it is expected that using a gain- framed narrative when describing the health outcomes of daily sunscreen application results in a more positive attitude towards the behavior and a more positive behavioral intention.
H1: Exposure to a gain-framed narrative results in a more positive attitude towards daily sunscreen application compared to exposure to a loss-framed narrative.
H2: Exposure to a gain-framed narrative results in a more positive intention to apply sunscreen daily compared to exposure to a loss-framed narrative.
Narratives have the unique capability of being able to facilitate an emotional experience for the viewer, which offers the experience of being swept up in the storyline and becoming involved with the characters in the story (Moyer-Gusé, 2008). This unique feature may even be a more effective way of influencing attitudes and behavior than traditional persuasive messages, as the involvement in the storyline and with the characters may help overcome resistance towards the persuasive intent of the message (Moyer-Gusé, 2008, Slater & Rouner, 2006), because the viewer has lower cognitive capabilities to resist the message due to their involvement in the story.
In the study by Murphy et al. (2013), a narrative was more effective than a non-narrative format in increasing cervical cancer-related knowledge and attitudes. In the study, participants were presented with either a narrative in the form of a video in a fictional format about a
family who spoke about cervical cancer, or a non-narrative informational video in a non- fictional format in which doctors speak about cervical cancer. Regressions revealed that transportation and identification with specific characters in the story caused shifts in knowledge, attitudes and behavioral intentions.
Identification with a character occurs whenever a person likes, perceive themselves as similar as the character and feel as if they know the character (Asbeek Brusse et al., 2016;
Slater & Rouner, 2002). Whenever a person truly identifies with a character, they perceive the events happening in a story as if they are happening to themselves (Cohen, 2006).
According to Swann et al. (1987), people are generally motivated to obtain or remain a positive self-view. People prefer to view themselves as responsible beings, so information that fits with this image of themselves is more easily accepted. This positive self-view is more easily maintained if the character in the story with which the viewer identifies shows responsible behavior. Therefore, a positively framed message is more useful to the viewer than a negatively framed message, as the positive message helps maintain the positive self- view. The positive self-view is more easily remained by identifying with a character that shows responsible behavior and experiences positive life outcomes (Asbeek Brusse et al., 2016). Therefore, it is assumed that using the positive message of a gain frame in a narrative is more effective in increasing identification then using the negative message of a loss frame.
Even though this effect was not found by Asbeek Brusse et al. (2016), this study aims at finding this expected relationship by improving the stimulus material. Therefore, the next relationship is proposed:
H3: Exposure to a gain framed narrative compared to a loss framed narrative will increase identification.
According to Green and Brock (2000), when somebody feels fully transported, he or she is no longer focused on the events happening around them, but fully submerged in the events happening in the story. While this effect might seem similar to identification, one essential difference is that transportation does not take into account that the viewer takes the
perspective of one particular character (Moyer-Gusé & Nabi, 2010). According to Asbeek Brusse et al. (2016), a gain frame will lead to more transportation compared to a loss frame, as transportation is an enjoyable and affective process, and the more positive message of a gain frame is more likely to induce this positive affect (Green & Brock, 2000; Green, Brock,
& Kaufman, 2004). Even though this effect was not found by Asbeek Brusse et al. (2016), this study aims at finding this expected relationship by improving the stimulus material.
Therefore, the following is expected:
H4: Exposure to a gain-framed narrative compared to a loss-framed narrative will increase transportation.
Counterarguing is the generation of thoughts that are not consistent with the message that is being portrayed (Slater & Rouner, 2002). Counterarguing can be seen as a form of resistance towards the message (Fransen et al., 2015). In the EORM, it is described that both
transportation and identification have the capability of reducing counterarguing (Moyer-Gusé, 2008). Transportation can reduce the motivation and ability to counterargue, as transportation is an enjoyable as well as an immersive process, which makes viewers less motivated and less able to interrupt this process to generate counterarguments (Moyer-Gusé & Nabi, 2010; Green
& Brock, 2000; Green et al., 2004; Knowles & Linn, 2004; Moyer-Gusé, 2008; Slater &
Rouner, 2002). Identification with the character has a similar effect on counterarguing. As the viewer adopts the perspective, thoughts and feelings of the character, there is lower capacity to generate counterarguments on the advised behavior (Moyer-Gusé & Nabi, 2010; Slater &
Based on the findings above, the following relationships are proposed:
H5: More identification with the character will lead to less counterarguing.
H6: More transportation into the narrative will lead to less counterarguing.
When less counterarguments are generated, it is believed that the attitude towards a proposed behavior is more positive (Asbeek Brusse et al., 2016). Therefore, the following relationship is proposed:
H7: Less counterarguing against the message will lead to a more positive attitude towards daily sunscreen application.
The EORM suggests that identification with the character can heighten perceived
vulnerability towards the consequences of a health behavior (Moyer-Gusé, 2008). Perceived vulnerability is the extent to which a viewer sees themselves as vulnerable to the outcomes that performing or not performing a behavior may induce. People have the tendency towards an optimism bias, which is the belief that they are not vulnerable to the consequences of risky behavior (Goossens et al., 2002). Because of this bias, increasing the perceived vulnerability can be an effective tool in changing health behavior. According to the Health Belief Model,
perceived vulnerability, among other individual factors, leads to the outcome of (preventative) health behavior, which can be described as activities taken by a healthy person before
symptoms are present. (Kasl & Cobb, 1966; Rosenstock, 1974). Whenever a viewer identifies with a character, they are immersed in the worldview of the character and therefore
vicariously experience the emotions of the character. Whenever the character is then exposed to a certain health risk, this also heightens the feelings or vulnerability to this risk to the viewer. This may be helpful whenever the message is about a risk or a behavior that the viewer would not normally think of or see themselves as susceptible to (Moyer-Gusé, 2008).
As described above, higher levels of identification are expected to lead to higher perceived vulnerability of getting the disease in the future. Therefore, the following relationship is expected:
H8: Higher identification with the character will lead to higher perceived vulnerability for getting skin cancer.
According to the EORM, heightening the perceived vulnerability will lead to story-consistent attitudes and behaviors (Moyer-Gusé, 2008). Therefore, the following relationship is
H9: Higher perceived vulnerability to skin cancer will lead to a more positive attitude towards daily sunscreen application.
Based on the Theory of Planned Behavior (Ajzen, 1991), it is expected that a more positive attitude towards daily sunscreen application leads to a more positive intention of daily sunscreen application. The following relationship is proposed:
H10: A more positive attitude towards daily sunscreen application will lead to a more positive intention of daily sunscreen application.
The path model will look as follows:
Method Design and participants
The study consisted of a one-factorial between subjects design, with the factor being framing of the narrative. Several mediators were investigated in the study. An online experiment was carried out in order to establish the causal relationships between the concepts in the model, via the survey program Qualtrics. Participants were recruited through convenience sampling and snowball sampling. There were no exclusion criteria, except that participants had to be older than 18. The survey was in English. Participants were approached through social network apps Whatsapp and Instagram with the following message: “Hi! Would you be willing to fill out a survey for my Master’s thesis? The study is about sunscreen use. Filling out takes approximately 5-7 minutes and your help would be greatly appreciated.”. The message ended with a link, bringing participants to the Qualtrics survey. The link was active from 16 – 24 May 2022. The sample consisted of N=120 participants. Values that were not filled out were marked missing, 5 participants had missing values. The mean age was 38.59, with a minimum age of 20 and a maximum age of 89. Of the sample, 24% was male, 75%
Figure 1: Expected path model.
was female and 1% non-binary/third gender. Most participants had the education level of a Bachelor’s degree (50.8%), followed by a high school degree (19.2%), a Master’s degree (13.3%) and an associate’s degree (12.5%).
After opening the link, it was explained to participants in a factsheet that the research was about sunscreen use, that their answers were recorded anonymously and where they could turn to in case of questions or complaints. Participants had to provide their informed consent to start the survey. First, participants were asked about their demographic information, being age, gender and highest completed education level. The participants then saw a text that explained that they were about to see a short cartoon with six images. It was also explained that it was not possible for participants to go back to a previous image once they had clicked the arrow to the next page. This was done so all participants saw the six images just once and in the correct order, without skipping forward and backward, so the storyline was perceived in the same way for all participants. After seeing either the gain or the loss condition of the story, participants were asked to answer a series of questions about the mediators and dependent variables. Participants first answered questions about the factors transportation, identification, counterarguing, attitude towards daily sunscreen use, intention to use sunscreen daily and perceived vulnerability to skin cancer. Finally, participants filled out the question for the manipulation check. The whole survey took approximately five to seven minutes.
Participants were exposed to either a gain or a loss framed message embedded into a narrative. Since the study of Asbeek Brusse et al. (2016) used quite abstract images where emotions were harder to read and there were no backgrounds to the scenes, a new cartoon was
developed with the goal of the images looking more realistic. The cartoons show a woman in simple and clear scenes with visible emotions on her face. Two versions of a series of six images were created. The execution of the gain and loss frame versions was based on the study by Vafeiadis and Shen (2021). In this study, a gain frame was operationalized as describing the benefits of sun-protection. The narrative in this study shows a person that is often exposed to the sun but took necessary precautions and now has healthy skin as a result.
The loss condition in the study showed the narrative of a character who failed to take the necessary precautions and now battles skin cancer. This operationalization is in line with the earlier explained definitions of a gain and loss frame. (Rothman et al., 2006). The full versions of both conditions can be found in Appendix I.
The operationalization of the concepts was based on the measured used in the study by Asbeek Brusse et al. (2016) where possible.1
Transportation was measured with Green and Brock’s (2000) Transportation Scale. The scale consists of 11 items such as “While I was reading the cartoon, I could easily picture the events taking place” and “I could picture myself in the scene of the events described in the cartoon”, with a 7-point Likert scale ranging from 1 (not at all) – 7 (very much). In the study by
1This was done for the concepts of transportation, identification and counterarguing. Attitude towards the behavior and behavioral intention were based on different literature on sunscreen use, because the Asbeek Brusse et al. (2016) study focused on a different behavior. All scales were computed using the MEAN option in SPSS.
Asbeek Brusse et al. (2016), 10 out of 11 items were used, but since it was not described which item was not incorporated, all 11 items were used in the current study. Items 2, 5 and 9 were reverse coded. A factor analysis with varimax rotation indicated that the scale was not unidimensional but loaded onto 4 factors, Eigenvalue= 3.220, Eigenvalue=1.511
Eigenvalue=1.419 Eigenvalue=1.044, together explaining 65.40% of the total variance. The scale proved to be reliable with Cronbach’s α =.69.Even though the scale did not prove to be unidimensional, because the scale has been validated in previous studies and showed to be reliable with a sufficient Cronbach’s α score, the full scale was still used in the current study. The newly computed scale had a mean of 4.41 (SD=.79, min=2.45, max=6.45). A higher score indicates more transportation into the storyline.
Identification was measured with a scale by Cohen (2001). This scale consists of 10 items measured on a 7-point Likert scale, ranging from 1 (not at all) – 7 (very much). Examples of items are “While reading the cartoon I could feel the emotions Emma portrayed” and “When Emma succeeded, I felt joy, but when she failed, I was sad”. A factor analysis with varimax rotation indicated that the scale was not unidimensional but loaded onto 2 factors, Eigenvalue
= 4.043, Eigenvalue=1.522, together explaining 55.61% of the total variance. The scale proved to be reliable with Cronbach’s α =.82. Even though the scale did not prove to be unidimensional, because the scale has been validated in previous studiesand showed to be reliable with a sufficient Cronbach’s α score, the full scale was still used in the current study.
The newly computed scale had a mean of 4.64 (SD=.94, min=1.70, max=6.60). A higher score indicates more identification with the character.
Measuring counterarguing was based on Nabi et al. (2007). The scale consists of three items:
“During the cartoon, I criticized the message of the cartoon”, “During the cartoon, several counterarguments occurred to me” and “During the cartoon I was looking for things that are not true”. All items were measured on a 7-point Likert scale, ranging from 1 (not at all) to 7 (very much). A factor analysis with varimax rotation indicated that the scale was
unidimensional, Eigenvalue = 2.228, explaining 61.55% of the variance. The scale proved to be reliable with Cronbach’s α =.83. The newly computed scale had a mean of 3.24 (SD=1.51, min=1, max=7). A higher score indicates more counterarguing.
Attitude towards daily sunscreen use
Measuring the attitude towards sunscreen use is based on Jessop, Simmonds, and Sparks (2009). The scale consists of four 7-point word pairs, being bad/good, negative/positive, foolish/wise and unpleasant/pleasant. The scales were originally 9-point scales, this was changed to 7-point scales for the current study to be more in line with the other questions, for the question to be clearer for the participants. A factor analysis with varimax rotation
indicated that the scale was unidimensional, Eigenvalue = 1.945, explaining 51.41% of the variance. The scale first proved to be unreliable with Cronbach’s α =.52, therefore the item unpleasant/pleasant was removed from the scale, which improved the scale to Cronbach’s α
=.71. The newly computed scale had a mean of 6.51 (SD=.81, min=3, max=7). A higher score indicates a more positive attitude towards daily sunscreen use.
Intention towards daily sunscreen application
The scale for measuring the intention towards the behavior was taken from Asbeek Brusse et al. (2016), but adapted for the behavior of daily sunscreen application. The scale consists of
three items, being “I intent to use sunscreen daily”, “I will try to use sunscreen daily” and “I will use sunscreen daily”. The items were measured on 7-point Likert scales ranging from 1 (not all) to 7 (very much). A factor analysis with varimax rotation indicated that the scale was unidimensional, Eigenvalue = 2.736, explaining 86.94% of the total variance. The scale proved to be reliable with Cronbach’s α =.95. The newly computed scale had a mean of 3.99 (SD=1.92, min=1, max=7). A higher score indicates a higher intention to use sunscreen daily.
To measure perceived vulnerability, the scale by Jafar et al. (2013) was used. The scale originally consisted of 5 items, but one item about work that was not relevant and fitting for the current study was removed. The remaining items were “It is possible that I get skin cancer in the future”, “I feel that I am at risk of skin cancer”, “I have healthy skin and I would not probably get skin cancer” (reverse coded) and “My skin health is quite good and I am not concerned with getting skin cancer” (reverse coded). The items were measures with 7-point Likert scales ranging from 1 (strongly disagree) to 7 (strongly agree). A factor analysis with varimax rotation indicated that the scale was unidimensional, Eigenvalue = 2.078, explaining 39.26% of the total variance. The scale proved to be reliable with Cronbach’s α =.75.
The newly computed scale had a mean of 4.18 (SD=.77, min=2, max=6). A higher score indicates a higher perceived vulnerability to skin cancer.
Results Manipulation check
To check whether the manipulation of the stimulus material was perceived by the participants as intended, an independent samples t-test was performed with condition (gain vs. loss) as independent variable and the score for the manipulation check as dependent variable. It was tested whether the gain condition made participants feel better than the loss condition, as
participants in the gain framed condition should perceive the story as more positive than the participants in the loss framed condition (Elbert & Ots, 2018). The item was measured with the question “How good did the outcome of the cartoon make you feel?” which could be answered on a 7-point scale ranging from 1(not good at all) to 7(very good). The 57
participants (M=5.28, SD=1.28) who saw the gain condition compared to the 58 participants in the loss condition (M=3.57, SD=1.33) scored significantly higher on the scale indicating how good the cartoon made them feel, t(113)=-7.046, p<.001. This indicates that the
participants who saw the gain framed cartoon had a more positive feeling than the participants who saw the loss framed cartoon, meaning the manipulation was successful.
Three randomization checks were performed for age, gender, and level of education.
An independent samples t-test showed that there was not a significant difference in age for the participants who saw a loss frame (M=37.48 SD=16.68) compared to participants who saw a gain frame (M=39.70 SD=16.22), t(118)=-.734, p=.464. Two Chi-Square tests showed that there was no significant difference in gender for the participants who saw a loss frame compared to participants who saw a gain frame, X2(2, N=120)=2.04, p=.361, and that there was no significant difference in education level for the participants who saw a loss frame compared to participants who saw a gain frame, X2(6, N=120)=8.42, p=.209. It can be concluded that the randomization was successful for age, gender and education level.
To test the hypotheses that exposure to a gain-framed narrative results in a more positive attitude towards sunscreen use and a higher intention to apply sunscreen daily compared to exposure to a loss-framed narrative, a multivariate analysis of variance (MANOVA) was
performed, as this was done in the study by Asbeek Brusse et al. (2016) as well. Framing of the narrative was included as independent variable, and attitude toward daily sunscreen
application and intention to apply sunscreen daily as dependent variables. Results showed that there were no significant results for the effect of framing of the narrative on attitude towards daily sunscreen application, F(1,113)=1.29, p=.259. A gain frame did not lead to a more positive attitude (M=6.42, SD=.90) than a loss frame (M=6.59, SD=.70). The was also no significant effect of framing on intention to apply sunscreen daily, F(1,113)=.91, p=.342, a gain frame (M= 3.82, SD=1.93) did not lead to a more positive intention than a loss frame (M=4.17, SD=1.91). Both h1 and h2 are rejected.
To test whether the framing of the narrative influenced identification and
transportation, two one-way ANOVA’s were performed. Results show no significant effect of framing of the narrative on transportation, F(1,119)=.72, p=.398, a gain frame (M=4.47 SD=.79) did not lead to more transportation compared to a loss frame (M=4.35, SD=.79) No significant effect of framing was found on identification F(1,116)=1.90, p=.170, a gain frame (M=4.76, SD=.93) did not lead to more identification than a loss frame (M=4.52 SD=.94) Both h3 and h4 are rejected.
To test the relationships between the mediating variables, linear regression analyses were performed. The analysis showed that identification did not significantly predict
b=-.19, t=-1,31, p=.192, rejecting h5. The analysis showed that transportation did not significantly predict counterarguing,
b=-.18, t=5.05, p=.326, rejecting h6.
Counterarguing negatively predicted attitude towards daily sunscreen application,
b=-.12, t=2.45, p=.015. Counterarguing predicted 5.1% of variance in attitude scores, R2=0.051, F(1, 115)=6.13, p=0.015, accepting h7. The analysis showed that identification did not
significantly predict perceived vulnerability,
b=-.05, t=-.61, p=.541, rejecting h8. The analysis
showed that perceived vulnerability did not significantly predict perceived vulnerability, b=- .14, t=1.46, p=.146, rejecting h9. Attitude towards daily sunscreen application positively predicted the intention to apply sunscreen daily,
b=.65, t=3.03, p=.003. Counterarguing predicted 7.5% of variance in intention scores, R2=0.075, F(1, 114)=9.19, p=0.003, accepting h10.
To investigate whether the effect of framing of the narrative is mediated by processes of narrative persuasion, attitude and perceived vulnerability, as well as by the attitude toward daily sunscreen application, a mediation analysis was performed by using the PROCESS macro in SPSS (Hayes, 2012). To be able to create a full model, the path via identification and the path via transportation were analyzed separately. Comparable to the analysis in the study by Asbeek Brusse et al. (2016), 10.000 bootstrap samples and a 95% confidence interval were used.
Path via transportation
In the figure below, the path model via transportation is visualized. Table 1 shows the mediation effects. Significant relationships (3,4) are in bold.
Figure 2: Serial multiple mediation model via transportation.
Table 1: Via transportation. Path coefficients estimated using PROCESS. Bootstrap results for mediation effects.
Path b SE t p
1 .08 .15 .56 .579
2 -.17 .18 -.974 .332
3 -.12 .05 -2.46 .016
4 .55 .22 2.49 .014
It can be seen from the model that effect of counterarguing on intention of applying sunscreen daily is mediated through attitude towards daily sunscreen application. Counterarguing
negatively influenced attitude,
b=-.12, t=-2.46, p=.016, meaning a lower level of
counterarguing leads to a more positive attitude towards daily sunscreen application. Attitude towards daily sunscreen application then has a positive effect on intention,
b=.55, t=2.49, p=.014, meaning a more positive attitude towards daily sunscreen application leads to a higher intention to apply sunscreen daily.
Path via identification
In the figure below, the path model via identification is visualized. Table 2 shows the mediation effects. Significant relationships (3,4) are in bold.
Figure 3: Serial multiple mediation model via identification.
Table 2: Via identification. Path coefficients estimated using PROCESS. Bootstrap results for mediation effects.
Path b SE t p
1 .22 .18 1.26 .211
2 -.04 .08 -.54 .589
3 -.18 .15 -1.18 .241
4 .19 .09 1.973 .051
5 -.13 .05 -2.67 .009
6 .45 .22 2.03 .045
7 -.38 .34 -1.12 .266
It can be seen from the model that effect of counterarguing on intention of applying sunscreen daily is mediated through attitude towards daily sunscreen application. Counterarguing
negatively influenced attitude,
b=-.13, t=-2.67, p=.009, meaning a lower level of
counterarguing leads to a more positive attitude towards daily sunscreen application. Attitude towards daily sunscreen application then has a positive effect on intention,
b=.45, t=-2.03, p=.045, meaning a more positive attitude towards daily sunscreen application leads to a higher intention to apply sunscreen daily.
The aim of this study was to examine whether using a gain-framed narrative compared to a loss-framed narrative in E-E would lead to a more positive attitude and intention towards daily sunscreen application. In the study, the mediating effects of several mechanisms of narratives were examined, being transportation, identification, counterarguing and perceived vulnerability. No direct effects were found from framing of the narrative on attitude towards daily sunscreen application and the intention to apply sunscreen daily. This was not expected,
as based on Rothman et al. (2006) and Gallagher and Updegraff (2011), a gain frame was expected to be more effective when promoting prevention behaviors. Exposure to a gain frame did not cause more identification or transportation compared to exposure to a loss frame. This was not in line with expectations, as according to Green and Brock (2000) and Green et al. (2004), a gain frame would induce a more positive affect, which will lead to more transportation because this is an enjoyable process. According to Asbeek Brusse et al. (2016), identification would occur more when being exposed to a gain frame, as the more positive self-view would be easier to maintain when identifying with a character that shows
responsible behavior and experiences positive life outcomes. This effect was not found in the current study. Based on Moyer-Gusé and Nabi (2010), Green and Brock (2000), Green, et al.
(2004), Knowles and Linn (2004), Moyer-Gusé (2008) and Slater and Rouner (2002) it was expected that more transportation would lead to less counterarguing, as whenever viewers are transported by the narrative, the generation of counterarguments is suppressed. Identification with the character was expected to have a similar effect on counterarguing, as identification with the character makes the viewer adopt the perspective, thoughts and feelings of the character, leading to lower capacity to generate counterarguments (Moyer-Gusé & Nabi, 2010; Slater & Rouner, 2002). Both expected relationships were not found in the current study. This does not necessarily mean that identification and transportation are not important factors in reducing counterarguing, but that these factors were not affected by the framing conditions in the form of cartoons. Based on Moyer-Gusé, it was expected that identification with the character would lead to higher perceived vulnerability, as immersion in the
worldview of the character and the experience of the emotion related to the exposure to a health risk are vicariously experienced by the viewer when identifying with a character, which heightens the feelings of vulnerability. This relationship was not found in the current study.
This would not necessarily mean that identification is not an important factor in heightening
perceived vulnerability, but that the framing conditions in the form of cartoons were not able to influence identification. Based on the EORM by Moyer-Gusé (2008), it was expected that higher perceived vulnerability would lead to a more positive attitude towards daily sunscreen application. This effect was not found in the current study either.
Results showed that one mediating effect was present. Attitude significantly mediated the relationship between counterarguing and intention. Counterarguing negatively influenced attitude towards daily sunscreen application, meaning that somebody who generates fewer counterarguments during exposure to a message, has a more positive attitude towards daily sunscreen application. A more positive attitude towards sunscreen application in turn causes a more positive intention to apply sunscreen daily. These findings are in line with the
expectations that when fewer counterarguments are generated, the attitude towards the behavior is more positive (Asbeek Brusse et al., 2016).
Even though most of the findings were not in line with the expectations, the results show that counterarguing has an important role in influencing the attitude and intention towards daily sunscreen application. More research about the exact mechanisms of counterarguing could provide more knowledge about how to convince people to apply sunscreen daily. Since counterarguing was not affected by either identification or
transportation, it is possible that counterarguing is a personal factor that may be more present in some people. According to the ELM (Petty & Cacioppo, 1986), a persuasive message is judged on the arguments when the involvement of the viewer is high, compared to the heuristic cues for people with low involvement. To understand for which people
counterarguing occurs, investigating whether people with high involvement, for example, people with fair skin or skin cancer in their family, process arguments about sunscreen differently would be interesting.
The findings of the current study are not in line with the expectations that were formulated based on previous studies such as Asbeek Brusse et al. (2016) and Moyer-Gusé (2008). In Asbeek Brusse et al. (2016) however, the expectation of the framing of the narrative being able to influence identification and transportation was also not met. The intention to perform the behavior was only heightened directly by the framing of the narrative.
This result, combined with the current study implies that a gain-framed narrative is simply not effective in heightening identification and transportation compared to a loss-framed narrative.
One of the reasons for building upon the work by Asbeek Brusse et al. (2016) was to improve the stimulus material to a more realistic setting to examine whether this would lead to
significant relationships with identification and transportation. The fact that this was not successful also points to the argument described above that framing of the narrative is not an effective way of increasing identification and transportation.
In the EORM by Moyer-Gusé (2008), the mediating relationships of transportation, identification and perceived vulnerability were described. In the current study, none of these mediating relationships were confirmed, which implies that the mediating factors of
transportation, identification and perceived vulnerability are not present when combining narratives and framing in the form of a cartoon.
One explanation for the fact that different framing of the narrative did not result in different outcomes than expected could be due to the type of behavior that was being tested in this study. In the study done by Asbeek Brusse et al. (2016), the consequences of the target behavior of drunk cycling are something that can be experienced directly, both in the storyline and in real life. The risk associated with the behavior can be reduced right away by
performing the recommended behavior. For sunscreen use, however, the outcome of not performing the recommended behavior is much less immediate, as sun damage builds up over the years and is not always visible immediately (except for sunburns). This could be an
explanation for why persuasive tactics in campaigns regarding sunscreen use are not as effective, as the risky outcome of not performing the recommended behavior may be years from now, if ever. For this reason, it may be more difficult to increase perceived vulnerability to skin cancer and positively influence attitude and intention towards the behavior, as the receiver of the message may simply not be motivated enough to comply with the message when they do not feel the need to. This study shows that the theories used by Asbeek Brusse et al. (2016) may not be successful in predicting the direct and indirect effects for this type of health behavior. For future research, it could be tested whether the immediacy of the result of the behavior is a boundary condition for this mediated effect to take place.
In the study, a cartoon was used as stimulus material. It was found that the stimulus material did not cause differences in the different conditions on identification and transportation. It would be possible that this form of E-E is not effective in causing transportation and identification, as this was also not seen in Asbeek Brusse et al. (2016) and that E-E in the form of a scene in which real people can be seen, like in Murphy et al. (2013), would be more effective in causing changes in identification and transportation. It could be possible that it is easier for people to identify with real people and transport into a more realistic scene,
compared to a drawn image of a person.
Another limitation of the study is the sampling strategy and sample size. Due to limited time, the sample was made up of 120 participants. Since the participants were recruited through convenience and snowball sampling, the sample may not be representative of the whole population. One example is that approximately half of the participants were in possession of a bachelor’s degree, compared to 37% in the Dutch population (CBS, 2018).
The high education level may have influenced the way the stimulus material and questions were understood and processed in both conditions.
Another limitation of this study is not taking into account the factor of fear arousal.
Since both conditions of the stimulus material showed a cancer scare, it could be possible that fear was aroused in the participants. According to the Extended Parallel Process Model (Witte, 1992) whenever someone feels both the severity and susceptibility of a threat and then feels low self-efficacy and response efficacy, meaning they don’t feel like they themselves are able to lower the risk or they don’t feel like their response will be effective in lowering the risk, the message will be rejected. Because the factor of fear was not measured in this study, it is unknown whether the aroused fear in participants made the message less effective, and whether this effect occurred more in one of the conditions. For future research, taking into account the factor of fear may give insights into how and when a narrative is effective for sunscreen use, and whether fear may influence the mediating factors of transportation and identification.
The current study did not show that framing of the narrative is an effective way of convincing people to use sunscreen daily, which means that still no answer was found as to how to convince people to wear sunscreen daily. Since there is evidence that the use of narratives in E-E, as well as evidence that framing of messages about sunscreen use is effective on their own, it is recommended to designers of campaigns for sunscreen use to not combine these two persuasive tactics. It was seen from the results that counterarguing plays an important role in influencing attitude and intention. Designers of campaigns for sunscreen use should therefore consider persuasive tactics that reduce counterarguing towards sunscreen application. It was shown that using a cartoon was not successful in increasing identification and transportation,
which shows that a cartoon might not have been the best version of E-E for attitude and intention change. Therefore, practitioners should consider different forms of E-E and framed narratives, such as a scene in which real people can be seen.
It can be concluded from this study that for a narrative about daily sunscreen application, the type of framed message, gain or loss, does not directly influence the attitude towards applying sunscreen daily or the intention towards applying sunscreen daily. This effect is not
significantly mediated by identification, transportation or perceived vulnerability, which contradicts previous literature. Counterarguing however plays an important role, as this factor does influence attitude towards the daily sunscreen application, which in turn influences the intention to apply sunscreen daily. This shows that it is important to take into account the role of counterarguing when designing a persuasive message with the aim to increase daily
Appel, M., & Richter, T. (2007). Persuasive effects of fictional narratives increase over time. Media Psychology, 10(1), 113-134.
Asbeek Brusse, E. D., Fransen, M. L., & Smit, E. G. (2016). Framing in Entertainment- Education: Effects on processes of narrative persuasion. Health Communication, 1-9.
Cohen, J. (2006). Audience identification with media characters. In J. Bryant & P. Vorderer (Eds.), Psychology of entertainment (pp. 183–198). Mahwah, NJ: Lawrence Erlbaum Associates.
Detweiler, J. B., Bedell, B. T., Salovey, P., Pronin, E., & Rothman, A. J. (1999). Message framing and sunscreen use: Gain-framed messages motivate beach-goers. Health Psychology, 18(2), 189.
Elbert, S. P., Ots, P. (2018). Reading or listening to a gain- or loss- framed health message:
Effects of message framing and communication mode in the context of fruit and vegetable intake. Journal of health communication, 23(6), 573-580.
Fransen, M. L., Verlegh, P. W., Kirmani, A., & Smit, E. G. (2015). A typology of consumer strategies for resisting advertising, and a review of mechanisms for countering them. International Journal of Advertising, 34(1), 6-16.
Gallagher, K. M., & Updegraff, J. A. (2012). Health message framing effects on attitudes, intentions, and behavior: a meta-analytic review. Annals of behavioral
medicine, 43(1), 101-116.
Goossens, L., Beyers, W., Emmen, M., & van Aken, M. A. G. (2002). The imaginary audience and personal fable: Factor analyses and concurrent validity of the ‘‘New Look’’ measures. Journal of Research on Adolescence, 12, 193–215.
Graaf, A. D., Sanders, J., & Hoeken, H. (2016). Characteristics of narrative interventions and
health effects: a review of the content, form, and context of narratives in health-related narrative persuasion research. Review of Communication Research, 4, 88-131.
Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79, 701–721.
Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The role of transportation into narrative worlds. Communication Theory, 14, 311–327.
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from http://www.afhayes.com/public/pro cess2012.pdf
Hinyard, L. J., & Kreuter, M. W. (2007). Using narrative communication as a tool for health behavior change: a conceptual, theoretical, and empirical overview. Health education
& behavior, 34(5), 777-792.
Huidkankercampagne 2020 – Hoe ziet je huid er écht uit? (2020, May 26). Integraal Kankercentrum Nederland. Retrieved June 21, 2022, from
Iannacone, M. R., Hughes, M. C. B., & Green, A. C. (2014). Effects of sunscreen on skin cancer and photoaging. Photodermatology, photoimmunology & photomedicine, 30(2- 3), 55-61.
Integraal Kankercentrum Nederland. (n.d.). Kankersoorten. Retrieved April 1, 2022, from https://iknl.nl/kankersoorten
Jafar, T., Mohtasham, G., & Reza M, R. (2013). Threat appraisal for skin cancer among rural farmers in Ilam, Iran. Iranian Journal of Dermatology, 16(4), 121-127.
Jessop, D. C., Simmonds, L. V., & Sparks, P. (2009). Motivational and behavioural
consequences of self-affirmation interventions: A study of sunscreen use among women. Psychology and health, 24(5), 529-544.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.
Econometrica: Journal of the Econometric Society, 47, 263–291.
Kasl, S. V., & Cobb, S. (1966). Health behavior, illness behavior and sick role behavior: I.
Health and illness behavior. Archives of Environmental Health: An International Journal, 12(2), 246-266.
Kim, M.-S., & Hunter, J. E. (1993a). Attitude-behavior relations: A metaanalysis of attitudinal relevance and topic. Journal of Communication, 43, 101–142.
Knowles, E. S., & Linn, J. A. (2004). The importance of resistance to persuasion. Resistance and persuasion. Mahwah, NJ: Lawrence Erlbaum.
KWF. (n.d.). Huidkanker. KWF Kankerbestrijding. Retrieved April 22, 2022, from https://www.kwf.nl/kanker/huidkanker
Leenarts, M. (2021, April 16). Mijn onderzoek naar zonbescherming onder Nederlanders in 2019. Drs Leenarts. Retrieved April 21, 2022, from
Maslowski, R. (2018, September 11). Onderwijs. De Sociale Staat van Nederland. Retrieved June 13, 2022, from https://digitaal.scp.nl/ssn2018/onderwijs/
Moyer-Gusé, E. (2008). Toward a theory of entertainment persuasion: Explaining the
persuasive effects of entertainment-education messages. Communication theory, 18(3), 407-425.
Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26–52.
Murphy, S. T., Frank, L. B., Chatterjee, J. S., & Baezconde-Garbanati, L. (2013). Narrative versus nonnarrative: The role of identification, transportation, and emotion in reducing health disparities. Journal of Communication, 63(1), 116-137.
Nabi, R. L., Moyer-Gusé, E., & Byrne, S. (2007). All joking aside: An investigation into the persuasive effect of funny social issue messages. Communication Monographs, 74, 29–54.
O’Keefe, D. J., & Wu, D. (2012). Gain-framed messages do not motivate sun protection: A meta-analytic review of randomized trials comparing gain-framed and loss-framed appeals for promoting skin cancer prevention. International journal of environmental research and public health, 9(6), 2121-2133.
Perugini, P., Bonetti, M., Cozzi, A. C., & Colombo, G. L. (2019). Topical sunscreen application preventing skin cancer: Systematic review. Cosmetics, 6(3), 42.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion.
In Communication and persuasion (pp. 1-24). Springer, New York, NY.
Rothman, A. J. & Salovey, P. (1997). Shaping Perceptions to Motivate Healthy Behavior. Psychological Bulletin, 121 (1), 3-19.
Rothman, A. J., Bartels, R. D., Wlaschin, J., & Salovey, P. (2006). The strategic use of gain- and loss-framed messages to promote healthy behavior: How theory can inform practice. Journal of communication, 56(1), S202-S220.
Rosenstock, I. M. (1974). The health belief model and preventive health behavior. Health education monographs, 2(4), 354-386.
Singhal, A., & Rogers, E. M. (1999). Entertainment-education: A communication strategy for social change. Lawrence Erlbaum Associates Publishers.
Slater, M. D., & Rouner, D. (2002). Entertainment—education and elaboration likelihood:
Understanding the processing of narrative persuasion. Communication theory, 12(2), 173-191.
Swann, W. B., Griffin, J. J., Predmore, S. C., & Gaines, B. (1987). The cognitive–affective crossfire: When self-consistency confronts self-enhancement. Journal of personality and social psychology, 52(5), 881.
Tannenbaum, M. B., Hepler, J., Zimmerman, R., Saul, L., Jacobs, S., Wilson, K., &
Albarracin, D. (2015). Appealing to fear: a meta-analysis of fear appeal effectiveness and theories. Psychological Bulletin, 141, 1178-1204.
Vafeiadis, M., & Shen, F. (2021). Effects of narratives, frames, and involvement on health message effectiveness. Health Marketing Quarterly, 1-17.
Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model. Communications Monographs, 59(4), 329-349.
Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior, 27(5), 591-615.
Appendix I: Stimulus material Version 1: Loss condition
Loss condition, scene 1 Loss condition, scene 2
Loss condition, scene 3 Loss condition, scene 4
Loss condition, scene 5 Loss condition, scene 6
Version 2: gain condition
Gain condition, scene 1 Gain condition, scene 2
Gain condition, scene 3 Gain condition, scene 4
Gain condition, scene 5 Gain condition, scene 6
Appendix II: Questionnaire
First, thank you for your interest in participating in this research project! Before the
experiment starts, it is important that you are well-informed about the procedure. Therefore, we would like you to read this information letter carefully.
Goal of the study
The research is about health messages for sunscreen use.
The participant is asked to read the images and questions carefully and answer them to their best notice. The participant is asked to complete the survey fully.
Participation in the study entails no considerable risks or inconveniences. Participation in this study takes approximately 5-7 minutes.
Information about the research
As this research is being carried out under the responsibility of the The Amsterdam School of Communication Research (ASCoR), which is part of the University of Amsterdam (UvA), we can guarantee that:
- Your personal information (about who you are) remains confidential and will not be shared without your explicit consent. Your research data will be analyzed to answer the research question as described above in the goal of this study.
- You can refuse to participate in the research or cut short your participation without having to give a reason for doing so. You also have up to 7 days after participating to withdraw your permission to allow your answers or data to be used in the research. To withdraw your permission, you can contact the project leader directly (see below).
For more information about the research you are welcome to contact the researcher, Charlotte van de Wijngaert via email: email@example.com
Should you have any complaints or comments about the course of the research and the procedures it involves as a consequence of your participation in this research, you can contact the designated member of the Ethics Review Board representing ASCoR via ascor‐secr‐
firstname.lastname@example.org Any complaints or comments will be treated in the strictest confidence.
I would like to take this opportunity to thank you in advance for your assistance with this research, which I greatly appreciate.
Charlotte van de Wijngaert
If you would like to participate in the survey, click on “Yes” below.
With this you declare:
• I am 16 years or older.
• I have read and understood the information.
• I agree to participate in the study and to use the data obtained with it.
• I reserve the right to withdraw this consent without giving any reason.
• I reserve the right to stop the study at any time I wish.
oYes, I participate
oNo, I am not participating
What is your age in years?
What is the highest level of school you have completed or the highest degree you have received?
oLess than highschool degree
oHigh school degree (VMBO/HAVO/VWO)
oAssociate's or technical degree (MBO)
What is your gender?
oNon-binary / third gender
oPrefer not to say
You will now see a short cartoon in six images. Click on the arrow at the bottom to view the next image. View the images carefully, as you are not able to go back once you click on the arrow.
Choose the option that applies most to you:
While I was reading the
cartoon, I could easily
picture the events taking
While I was reading the
cartoon, activity going on in
the room around me was on my mind.
I could picture myself in the
scene of the events described in the cartoon.
I was mentally involved in the cartoon
while reading it.
After finishing the
cartoon, I found it easy
to put it out of my mind.
I wanted to learn how the cartoon
much The cartoon
much I found
myself thinking of
ways the cartoon could have
Choose the option that applies most to you:
turned out differently.
I found my mind wandering
while reading the
The events in the cartoon are
relevant to my everyday
The events in the cartoon have
changed my life.