Under the Influence
An online experiment looking into the effects of peers and fear in messages about XTC.
Student Name: Yara de Kwaasteniet Student Number: 11388021
Wordcount: 7006
Supervisor: J.S. Lemmens
Master’s programme Communication Science Graduate School of Communication (GSC) University of Amsterdam
Master’s Thesis 7 July 2017
Abstract
This study looked into the effects of peer presence and fear appeal on viewers’ responses to messages on XTC. Specifically it looked at how presence of peers and fear appeal in messages about XTC influence the positive affect, knowledge and behavioural intention of young adults. In addition this study looked at the mediating power of transportation and appreciation. The research was conducted by creating an online experiment using Qualtrics. Participants were first presented with one of four videos after which they had to answer several questions and propositions. This study found that both the presence of peers had a direct negative influence on participants’ knowledge gain, but that fear appeal had no effect on viewers’ responses. In addition results showed that appreciation mediated both the
relationship between peers and knowledge, and the relationship between fear and knowledge. In both cases appreciation had a positive effect, in some way softening the negative effect of peers and fear on knowledge. This study concluded that perhaps the creating appreciation is more important than the content, and in order to gain appreciation one should really know their audience. For young adults the best practice would be to not use peers, but maybe to present information in text-based formats.
1 Introduction
In the last two years, The Guardian used the headlines: ‘Ecstasy in comeback as new generation discovers dance drug’ (Alan, 2016) and ‘Ecstasy and LSD use reaches new high among young’ (Gayle & Pegg, 2015). Last June, headlines from Dutch news platforms RTL Nieuws, NOS and NRC read: ‘Dutch youth most frequent users of XTC1’, ‘Dutch youth on top of European XTC list2’ and ‘The Netherlands remains in the top of the list on drug use in Europe3’. These reports indicate that over the last few years XTC has gained a renewed popularity among young people. The drug has become increasingly popular with mainstream crowds, whereas before the drug was mostly used within a niche community. And as more young people are using the drug, it becomes increasingly important to educate this age group about how this particular drug works. The big question is however, how can they be reached and how can they best be educated? Reaching this group might not be the biggest problem, as most of young people nowadays use new media throughout their day (Geraci & Nagy, 2004). However, in order to stand out and to actually get a message across, it is important that the content and the message are tailored to the audience. What does an informative video about XTC need to contain that will make young people not only feel attracted to the clip, but will also make them watch it until the end.
One organisation that attempted to create a perfect recipe for an educational message about XTC is BNN. Their YouTube-based series Drugslab presents itself as “an educational YouTube channel about drugs” (BNN Drugslab, 2016). They have put three fresh-looking young people in front of the camera, while they report the pros and cons of XTC use. They seem to focus on two important aspects of reaching young adults: peers and no use fear appeal. Young adults tend to steer away from parental authority and are now more focused on influences from their peer group (Valkenburg & Cantor, 2000). Research into this topic has
1
("Nederlandse jongeren gebruiken vaakst xtc", 2017)
2
("Nederlandse jongeren bovenaan Europese xtc-lijst", 2017)
2
found that videos produced by peers tended to have a bigger influence on changing young adults’ attitudes, than videos produced by experts or organisations (Paek, Hove, Jeong, & Kim, 2011). In addition, BNN attempts to avoid the boomerang effect that a lot of these messages seem to cause (Riet & Ruiter, 2013; Roskos-Ewoldsen, Yu, & Rhodes, 2004), by not using a fear appeal in their videos. A boomerang effect means that the target audience will do exactly that which you were trying to avoid (Moyer-Gusé, 2008), which in case of XTC is not a favourable outcome. Based on these two factors it seems like BNN’s Drugslab should be able to live up their motto of being an educational channel. However, traditionally a lot of health messages about XTC or drugs in general have used fear appeal to educate people about drugs (Laroche, Toffoli, Zhang, & Pons, 2001; Moyer-Gusé, 2008; Robberson & Rogers, 1988).
Drugslab is relatively new and has quickly gained popularity among young people4, but does it actually work? And on another note, how effective is the format comparison to messages about XTC that have used fear appeal, or left out the peers? Might there be other factors that contribute to the success of these types of videos? This study suggests that transportation into the narrative and appreciation of the message could be underlying mechanisms. Both of these concepts have in the past been proven to have predictive power when it comes to changing attitudes, behavioural intent and affecting knowledge (Adler & Fagley, 2005; Murphy, Frank, Moran, & Patnoe-Woodley, 2011). As discussed above, previous studies have revealed conflicting outcomes in terms of fear appeal. However, most studies on the presence of peers and peer influence have shown that they have a positive influence on informing young adults. In addition, most studies have been positive about triggering transportation and appreciation to help engage viewers in the message. This study proposes that these four concepts are all connected when it comes to educating young adults
4
3
and changing their attitude towards drugs. In order to determine how these concepts work together, this study will aim to answer the following research question:
RQ: How do the characteristics of messages about drug use influence viewers’ cognitive, affective and behavioural responses?
4 Theory
Peer Influence
When children are young, their parents are their main source of information and the ones who determine their values (Valkenburg & Cantor, 2000). However, when they grow older and reach adolescence and adulthood, there is a shift. From then on young people tend to attach more value to what their peers say or believe, than to what their parents say or believe. This shift not only occurs from parents to peers, but also from other forms of authority, such as the government, to peers (Valkenburg & Cantor, 2000). The general consensus seems to be that peers play a valuable and influential role in the education of young people. Research by Paek et al. (2011) found that when young adults were presented with a message created by peers these were more effective in changing attitudes towards both the message and the issue, in comparison to a message presented by an adult or an expert. In addition, another study found that young people talking to peers were more willing to inform themselves and to share information than the young people who talked to adults (Wolf & Pulerwitz, 2003).
As research has shown, peers can play an important part when it comes to influencing young people, in many different ways. In order to find out how they play a part in messages about drugs, the following sub question is posed:
SQ1: How does the presence of peers in messages about drug use influence viewers’ responses?
Fear Appeal
In the field of health related persuasion fear appeal is a widely researched topic. A lot of studies have found that fear appeals are positively related to persuasion (Laroche et al., 2001;
5
Robberson & Rogers, 1988). When fear appeal is used this means that the message contains some sort of threat, in this case to someone’s health. And when people feel their health is threatened, this leads to fear. When the target is sufficiently involved, and the message also provides a solution to the threat, the fear appeal can lead to action (Moyer-Gusé, 2008; Riet & Ruiter, 2013; Witte, 1992).
However, not all studies have found that fear appeals are effective when it comes to educating. When a fear appeal is used, but the target is not sufficiently involved or there is no solution provided, the overwhelming fear might cause people to avoid the message or even counteract it (Riet & Ruiter, 2013; Roskos-Ewoldsen et al., 2004; Witte, 1992). In addition the use of fear can lead to the opposite of the intended behaviour when people’s existing knowledge and beliefs do not match with the information provided. Clashing information is a good reason for people to counteract a message, and as a result show behaviour that the message tried to avoid (Roskos-Ewoldsen et al., 2004). In other words, fear can cause boomerang effects, leading people to do exactly the opposite of what the message tried to convey. A format such as Drugslab does not use fear appeal as a persuasive method, and even when they mention harmful effects they balance this with a solution. However, traditional drug related entertainment education is known for using fear appeal as a persuasive method (Cauberghe, De Pelsmacker, Janssens, & Dens, 2009).
Following the information discussed earlier it is expected that the formats that balance their message (i.e. in terms of positive and negative information), will have a more positive effect than formats that focus on the use of fear appeal and negative information. In order to examine the relation between fear appeal and knowledge, behaviour and emotions the following question is posed:
6 Transportation
As mentioned above, involvement plays an important part in fear appeal theory, but is also plays a role in the general field of persuasion (Moyer-Gusé, 2008). In entertainment education there is a lot of space for narrative in the message, which in turn draws people in and makes them interested in the message (Moyer-Gusé, 2008). The main idea of transportation is that viewers are “primarily engaged in the storyline, rather than in one’s immediate environment” (Moyer-Gusé, 2008, p. 409). Because of that, entertainment education can potentially be much more effective than overtly persuasive messages when it comes to delivering prosocial or health related messages. In addition, transportation has been found to be positively related to the acceptation of the message presented, however in this case the only correlation and not causality could be determined (Green, 2004). A study by Murphy et al. (2011) found that transportation or involvement were the best predictors of changes in behaviour, knowledge and emotions of young people.
Taking into account the information presented above, it is expected that higher
amounts of transportation lead to a higher acceptance rate of the message and subsequently to positive changes in behaviour, knowledge and emotions. In order to find out how and if this works, the following sub question is presented:
SQ3: How does transportation affect the relationship between the characteristics of messages and viewers’ responses?
Appreciation
Apart from transportation, appreciation also plays a role in how people interact with narratives and perceive information. Appreciation entails that a person acknowledges the value and meaning of something and that they feel a certain positive emotional connection to
7
it (Adler & Fagley, 2005). So in these terms appreciation is something more than mere pleasure, but rather a more gratifying feeling (Oliver & Bartsch, 2010; Oliver & Raney, 2011). It is important to notice that appreciation can also be linked to more tragic topics (i.e. drugs, health issues), whereas pleasure or fun is generally linked to more topics like sport and comedy (Oliver & Raney, 2011). This can also be described as enjoyment, which in turn can fall under appreciation. According to a study by Reeve “the principle motivational effects of enjoyment are the willingness to continue and persist in the activity” (1989, p. 101). One could argue that because enjoyment is a part of appreciation, appreciation can make people stay focused on the message and finish watching it. In addition, when a person appreciates the material and information presented, it will enhance their mood and connection the material. This as a result leads to a positive reaction to the message, and it becomes more likely the person accepts the information and uses it (Adler & Fagley, 2005).
Following the information discussed above, one would expect that appreciation can cause an increase in knowledge, positive emotions and the desired intended behaviour. In order to find out how appreciation comes into play, the following sub-question is posed:
SQ4: How does appreciation affect the relationship between the characteristics of messages and viewers’ responses?
8
Fig 1. Hypothesised model based on theoretical framework.
The model (see fig. 1) presented above is a helpful tool to put the concepts discussed into perspective and to make the expected relations visible. This framework and model will make it easier to understand and look into the questions at hand, and most importantly will help to find out how peers and fear appeal play a role in educational messages about drug use and how they influence viewers’ responses to the messages.
9 Method Sample
Participants were gathered by posting an anonymous link on different online social platforms. The link sent participants to the online survey tool Qualtrics, which was used to set up the research. A total of 137 participants started the research, but the final data set consisted of 108 participants. The other 29 participants were deleted because they did not finish the research or because they were too young. The aim was to gather participants between 18 and 25 years old, but eventually all participants older than 18 were included in the sample. The goal was to gather a minimum of 120 participants, as an experiment should consist of at least 20-30 participants per condition (Sekaran, 2003). As this research deals with four conditions, this leads to 80-120 participants. Due to complications and time restraints, the final goal was set at a minimum of 100 participants. Qualtrics was set up to randomly and equally distribute participants into the four conditions.
The final sample (N=108) consisted of 30 males (27.8%) and 78 females (72.2%). The average age of the participants was 26.07 years old (SD = 10.48), with the youngest
respondent being 18 and the oldest 62 years old. Of all the participants, 71% indicated they had never used XTC in their life, 7% indicated 1-2 days, 6% said 3-10 days, 12% mentioned 11-25 days and 12% indicated 26-50 days in their life. In terms of educational background 1.9% finished primary education, 12% a form of lower education (i.e. VMBO, MAVO, MBO), 21.3% a form of middle education (i.e. HAVO, VWO) and 64.8% a form of higher education (e.g. HBO, WO). Regarding the stimulus material, 97.2% indicated they had not previously seen the material, 1.9% indicated they did, and 0.9% indicated they did not know.
The characteristics of the conditions can be found in table 1. There were no significant differences in age between the four conditions F (3, 97) = 1.55, p = .21.
10
Table 1. Characteristics of conditions after data clean up.
Condition % of sample Mean age SD age Length of video
Peers/no fear 23.1 (N = 25) 28.7 12.72 1:24 min
No peers/fear 20.4 (N = 22) 21.8 1.92 1:50 min
No peers/no fear 32.4 (N = 35) 26.8 11.38 1:50 min
Peers/fear 24.1 (N = 26) 25.7 10.17 1:17 min
Procedure
The language used in the research was Dutch, as Dutch people were the target group of the research. Once a respondent had clicked on the link, they were first presented with a consent form. In the consent form participants were informed about the general topic, the amount of time the research would take, their rights concerning participation and anonymity and they were given an incentive (i.e. the chance to win a Bol.com gift card after completing the survey). After having accepted the terms of the research, participants were randomly assigned to one of four experimental conditions: fear and peers, fear and no peers, no fear and no peers, no fear and peers. In these conditions participants were presented with a video belonging to their experimental condition. In order to make sure the exposure time to the stimulus material was similar, each of these videos was around 2 minutes long. In order to assure the respondent would watch the video a background timer was set that only showed the ‘next’ button after 1.5 minute. In addition, right after the video participants were asked to identify which one of four pictures could be seen in the video they just watched, and also whether they had
previously seen the video.
After the manipulation check, the measurements followed. Participants were asked about their transportation and involvement with the clip and message. In addition they had to answer questions about their appreciation of the video and message. After this, participants were asked about their feelings towards the use of XTC, what they had learned from the video and what their previous and intended behaviour looked like. Participants were also asked
11
three basic demographic questions, regarding age, gender and educational background. At this point, participants could fill in their email address in case they wanted to participate in the raffle for one of the two €10 Bol.com gift cards. In the final part participants could fill in any questions or comments they had regarding the research. Afterwards participants were thanked for their time and the experiment was closed. The entire experiment took around five minutes. A transcript of the online experiment can be found in appendix A.
Stimulus Material
The stimulus material used in this research was picked based on the presence or absence of peers and fear appeal. The initial video which led to inspiration for the research was a video by Drugslab in which peers are present, but fear appeal is absent. To complete the stimulus material three other videos were found: one with peers and fear appeal (by Drugfree World), one with no pears and no fear appeal (by Unity), and one with no peers but with fear appeal (by Trimbos Institute). All four videos were retrieved from YouTube (see table 1 for the length of the videos).
Peers and no fear. The video (BNN Drugslab, 2016) shows a girl who talks about the good and bad sides of XTC, as well as the quantities, the precautions and the possible
consequences. The information in the video is neutral and balanced, without using a fear appeal. This video was not edited before it was used in the research.
Peers and fear. The Drugfree world video shows a group of young adults (the peers) who talk about their first time using XTC and all the negative, short- and long term,
consequences they faced. In contrast to the BNN video this clip is very one-sided, only taking into account the downsides of XTC use. The video is quite dark, sometimes even using black and white filters. In addition the music used is somewhat haunting heavy, adding to the fear
12
element of the video. This video was originally 10 minutes long (Drug-Free World, 2011), but was edited to be suitable for the research (Drug-Free World, 2017).
No peers and no fear. The Unity video (Unity, 2013) shows no peers, but has a voice over who talks the viewer through the video. In addition an animated character is used
illustrate the information. Just like the BNN video, the video is quite neutral and balanced. To make the video more compliant to the category no peers and no fear, the part with fear appeal was cut during editing. .
No peers and fear. The video by Trimbos institute uses a voice over to present the information and music in the background that creates a dark atmosphere. In addition the information presented was mostly focused on the negative aspects of XTC use. The original video (Trimbos-Instituut, 2014) contained peers talking about their experiences, to make the clip suitable for the no peers and fear condition, the part with the peers was cut during editing. The final video was 1:17 minutes long.
Measures
Transportation. To measure transportation, questions from the Transport Narrative Questionnaire by Green & Brock (2013) were used. From the 12 items, five relevant questions were picked (e.g. I was mentally involved in the narrative while watching the video). All questions could be answered on a 5 point Likert-scale. The reliability test (Cronbach’s α = .84) proved that the scale was reliable.
Appreciation. The questions for appreciation were derived from the Experience Enjoyment Scale by Lin, Gregor and Ewing (2008). Six questions were eventually used in the research (e.g. I had fun watching the video) and could be answered on a 5 point Likert-scale. The reliability test (Cronbach’s α = .89) proved the scale was reliable.
13
Positive affect. Participants were presented with nine feelings and their antonyms (e.g. pleasant – unpleasant, safe – unsafe), which were selected using a list of emotions (List of Emotions, n.d.). On a 5 point scale they could indicate which feeling was most accurate when it came to the perceived use of XTC. The reliability test revealed that one item (i.e. exciting) brought down the Cronbach’s α (= .89). After this item was deleted, the reliability test (Cronbach’s α = .95) proved the scale was reliable.
Knowledge. Propositions about knowledge gained from the videos were created using information present or absent in the video. The six propositions (e.g. I learned about: XTC in general; positive aspects of XTC) presented were based on general knowledge that could have been obtained from the videos. Participants could indicate on a 5 point Likert-scale how much they agreed with the proposition. The reliability test (Cronbach’s α = .86) proved the scale was reliable.
Behavioural intent. The questions regarding past behaviour and behavioural intent were loosely based on tips and a list of behavioural questions ("Behavioral and Hypothetical Interview Questions", n.d.; Rudloff, n.d.). Firstly participants were asked on how many days in their lives they had used XTC with answer options ranging from never to over 50 days on a 6 point scale. Secondly participants were asked about their future intentions through four questions (e.g. Are you planning to use XTC in the future), which could be answered on a 5 point-Likert scale. The reliability test (Cronbach’s α =.91) proved the scale was reliable.
14 Results The Effect of Peers
To determine what the effect of peers is, five independent-samples t-test were run for positive affect, knowledge, behavioural intention, transportation and appreciation. Table 2 shows the means and standard deviations for the peers- and no-peers condition. First of all, participants who were presented with a video with peers turned out to gain less knowledge than the participants who watched a video without peers, t (106) = 2.50, p = .01. In addition the tests showed that participants who were in the condition with peers felt less transported than the participants in condition with no peers, t (106) = 2.28, p = .03. Finally the tests revealed that participants who watched video with peers appreciated the message less than participants who watched the video without peers, t (85.76) = 2.48, p = .02. No other significant differences were found.
Table 2. Mean and standard deviations for t-tests measuring effect of peers.
Peers No Peers M SD M SD Positive Affect 3.27 1.52 3.09 1.29 Knowledge * 2.79 1.06 3.27 0.92 Behavioural Intent 2.19 1.36 2.11 1.36 Transportation * 3.17 1.00 3.57 0.82 Appreciation * 3.28 1.03 3.70 0.69
Note: * indicates significant difference between conditions, at least p < .05
The Effect of Fear Appeal
In order to establish whether fear appeal has any effect on positive affect, knowledge, behavioural intention, transportation and appreciation, five different independensamples t-tests were run. Table 3 shows the means and standard deviations for the fear-and no-fear condition. The t-tests showed that participants who were exposed to the video that used fear
15
appeal appreciated the message less than participants who were presented a video where no fear was used, t (81.27) = 3.60, p = .001. No other significant differences were found.
Table 3. Mean and standard deviations for t-tests measuring effect of fear appeal.
Fear No Fear M SD M SD Positive Affect 3.09 1.18 3.25 1.56 Knowledge 3.00 1.15 3.08 0.89 Behavioural Intent 2.19 1.38 2.11 1.35 Transportation 3.23 1.03 3.50 0.82 Appreciation * 3.17 1.00 3.78 0.70
Note: * indicates significant difference between conditions p = .001
The Effects of the Conditions
Six one-way ANOVA’s with Bonferroni Post-Hoc were conducted to determine a statistically significant impact of condition on the different variables. Table 4 shows the means and standard deviations for all four conditions. The ANOVA for knowledge, F (3,104) = 3.073, p = .03, revealed that participants in the condition with no peers, but with fear gained more knowledge than the participants in the condition with both peers and fear. The ANOVA for transportation, F (3,104) = 5.162, p = .002, showed that participants in the condition with peers and fear were transported less than the participants in the condition with no peers but with fear, the condition with peers but with no fear and the condition with no peers and no fear. When it came to appreciation, F (3,104) = 20.479, p < .001, the ANOVA showed that the participants in the condition with peers and fear appreciated the message less than the participants in the condition with no peers but with fear, the condition with peers but with no fear and the condition with no peers and no fear.
16
Table 4. Mean and standard deviations for one-way ANOVA’s measuring effect of condition. Peers/no fear No peers/fear No peers/no fear Peers/fear
M SD M SD M SD M SD Positive Affect 3.62 1.66 3.26 1.01 2.98 1.44 2.94 1.31 Knowledge 2.97 0.94 3.45a 0.99 3.15 0.86 2.62a 1.16 Behavioural Intent 2.01 1.22 1.98 1.25 2.18 1.44 2.36 1.48 Transportation 3.54a 0.79 3.73b 0.74 3.47c 0.86 2.82a,b,c 1.10 Appreciation 4.01a 0.60 3.61b 0.72 3.86c 0.62 2.58a,b,c 0.87 Note: Rows with identical superscript (a,b,c) differ significantly, at least p <.05
Transportation as a Mediator
In order to establish whether transportation acted as a mediator, six mediation tests were run in SPSS using the PROCESS 2.16 macro. Table 5 shows the coefficients of the different paths within the six models (see fig 1 and fig 2) as well as their levels of significance. The test with peers and positive affect showed that the presence of peers was a significant negative
predictor of transportation; however none of the other paths within that model turned out significant, just like the indirect effects were not significant, b = .07, SE = .08, 95% CI = -.289, .036. The test with peers and knowledge indicated that the presence of peers was a significant negative predictor of knowledge, and that the presence of peers was a significant negative predictor of transportation. However, even after controlling for the mediator the presence of peers was still a significant predictor of knowledge, which indicates partial mediation. Approximately 7% of the variance in knowledge was accounted for by the predictors (R2 = .07) However, when the indirect effect was tested using a bootstrap estimation approach with 5000 samples, it confirmed that the indirect coefficient was not significant, b = -.05, SE = .06, 95% CI = -.210, .039. The test with peers and behavioural intent showed once more that the presence of peers was a significant negative predictor of
17
transportation, but none of the other paths turned out significant, as well as the indirect effects, b = -.01, SE = .07, 95% CI = -.154, .150.
The three tests for fear as predictor in a model mediated by transportation did not provide any significant results, which is consistent with what was found in the independent-samples t-tests for the effect of fear.
Table 5. Coefficients and levels of significance for models with transportation as mediator.
X Y R2 A A’ B C
Peers Positive Affect .02 .18 .25 -.40c .18
Peers Knowledge .001 -.48c -.43c -.40c .11
Peers Behavioural Intent .07 .08 .09 -.40c .01
Fear appeal Positive Affect .01 -.16 -.12 -.27 .14
Fear appeal Knowledge .008 -.08 -.03 -.27 .16
Fear appeal Behavioural Intent .02 .08 .08 -.27 .01 Note: a = p < .001, b = p < .01, c = p < .05
Fig 1. Direct effect of X on Y Fig 2. Mediation model for transportation
Appreciation as a Mediator
In order to establish whether appreciation acted as a mediator, six mediation tests were run in SPSS using the PROCESS 2.16 macro. Table 6 shows the coefficients of the different paths within the six models (see fig 3 and fig 4) as well as their levels of significance.
The test with peers and positive affect showed that the presence of peers was a significant negative predictor of appreciation; however none of the other paths within that model turned
18
out significant, just like the indirect effects were not significant, b = -.04, SE = .05, 95% CI = -.209, .031. The test with peers and knowledge indicated that the presence of peers was a significant negative predictor of knowledge, and that the presence of peers was a significant negative predictor of appreciation. In addition it showed that appreciation was a significant positive predictor of knowledge. However, after controlling for the mediator the presence of peers was no longer a significant predictor of knowledge, which is consistent with full mediation. Approximately 13% of the variance in knowledge was accounted for by the predictors (R2 = .13). When the indirect effect was tested using a bootstrap estimation
approach with 5000 samples, it indicated the indirect coefficient was also significant, b = -.14, SE = .08, 95% CI = -.348, -.021.This means that the path via appreciation is a better predictor of knowledge than the direct path. The test with peers and behavioural intent showed once more that the presence of peers was a significant negative predictor of appreciation, but none of the other paths turned out significant, as well as the indirect effects, b = .10, SE = .08, 95% CI = -.016, .338.
The test with fear and positive affect showed that fear was a significant negative predictor of appreciation; however none of the other paths within that model turned out significant, just like the indirect effects were not significant, b = .05, SE = .10, 95% CI = -.119, .291. The test with fear and knowledge showed once more that fear was a significant negative predictor of appreciation, but it also showed that appreciation was a significant positive predictor of knowledge. For this model no significant results were shown for the predicting power of fear for knowledge. However, when the indirect effect was tested using a bootstrap estimation approach with 5000 samples, it indicated the indirect coefficient was in fact significant, b = -.24, SE = .12, 95% CI = -.527, -.067. Approximately 11% of the variance in knowledge was accounted for by the predictors (R2 = .11). This means that the path via appreciation is a better predictor of knowledge than the direct path. The test with fear and
19
behavioural intent test showed the same results as the previous test, where fear was a
significant negative predictor of appreciation. None of the other paths turned out significant, and neither did the indirect effect, b = .15, SE = .12, 95% CI = -.037, .430.
Table 6. Coefficients and levels of significance for models with appreciation as mediator.
X Y R2 A A’ B C
Peers Positive Affect .004 .18 .17 -.42c -.02
Peers Knowledge .02 -.48c -.34 -.42 c .32b
Peers Behavioural Intent .13 .08 -.01 -.42c -.23
Fear appeal Positive Affect .006 -.16 -.21 -.61a -.09
Fear appeal Knowledge .02 -.08 .17 -.61a .40a
Fear appeal Behavioural Intent .11 .08 .77 -.61a -.24 Note: a = p < .001, b = p < .01, c = p < .05
20 Discussion The Effects of Peers
In order to answer the sub-question whether the presence of peers had an effect on viewers’ responses in terms of positive affect, knowledge and behavioural intent, comparisons were made between the videos with and without peers. It seems like the presence of peers has a negative effect on viewers’ knowledge. Looking at the effects of the specific conditions, both videos with peers scored worse than the videos without peers, however the video with both peers and fear appeal scored the worst of all. These results seem to imply that when you want to teach something about XTC to young adults, it is better not to use peers. This is
contradictory to findings by Wolf and Pulerwitz (2003) who found that young people were much more willing to inform themselves when they got information from peers. The fact that no effects were found for positive affect and behavioural intent stands in contrast with
findings by Paek et al. (2011), because they found that when young adults were presented with information on this type of topic by a peer, there was a bigger chance they experienced changes in attitude and changes in feelings towards the topic.
In addition to testing the effect of peers on the viewers’ responses, the effect of peers on transportation and appreciation were also tested. It turns out that peers have both an effect on transportation and on appreciation. However, once more the results indicated that when no peers were present, the young adults felt more transported and were more appreciative. Once again the video with both peers and fear appeal scored the worst in comparison to the other three videos. These results were unexpected as previous research has proven on multiple occasions that that when young adults are presented information by peers, they are more willing to adopt information (Green, 2004; Wolf & Pulerwitz, 2003), and keep watching (Adler & Fagley, 2005; Reeve, 1989). However, the presence of peers may not be the only reason for these unexpected results. The video with both peers and fear appeal scored worst
21
on all aspects, which could mean that the Drugfree world video had more aspects that
negatively influenced the outcomes. It could be that the viewers saw the peers as unreliable or fake, or maybe the video quality was lacking for it to be taken seriously and for viewers. These aspects could lead to less appreciation of the message and engagement with the message.
Results indicated that transportation did not mediate the effect of peer presence on viewers’ knowledge gain, which was unexpected as previous research has found that that transportation is one of the best predictors of changes in knowledge (Moyer-Gusé, 2008; Murphy et al., 2011). A reason for this may be that the video with peers and fear appeal scored significantly lower on transportation than the video with peers, but without fear appeal. When the two videos with peers were taken together to determine the mediating effect of transportation, that one video influenced the model in such a way that transportation had no effect5. However, appreciation did mediate the effect of peers on viewers’ knowledge gain. At first the presence of peers has a negative effect on appreciation, when appreciation is added to the model peers no longer have that predicting power. This full mediation model shows that when there are no peers there is more appreciation and when there is more appreciation there is an increase in knowledge gain. These results are in line with research by Reeve (1989) and Adler and Fagley (2005), who found that when a message is appreciated there is a greater chance that the viewer will continue looking at the message and take in the information. This means that it might be important to make sure people appreciate the message regardless of its content.
To answer the first sub-question and part of the third and fourth: the presence of peers has a direct negative influence on the knowledge gain of viewers. However, appreciation
5
22
mediates this this relationship in a positive way, as appreciation actually increases knowledge gain.
The Effect of Fear Appeal
In order to answer the sub-question whether the presence of fear appeal had an effect on viewers’ responses in terms of positive affect, knowledge and behavioural intent, comparisons were made between the videos with and without fear appeal. Surprisingly, no significant results for the effect of fear were found. These results were unexpected because there has been a long history of research on the topic of fear appeal, and the effect it can have. And whether the result was in favour (Laroche et al., 2001; Robberson & Rogers, 1988; Witte, 1992) or against the use of fear appeal (Riet & Ruiter, 2013; Roskos-Ewoldsen et al., 2004), fear appeal always seemed to have some sort of influence on the effect. These unexpected results may be caused by the video with peers and fear appeal, as they scored significantly lower than the other videos on most parts. This video seems to have messed with the results, as it brings down the total score of the fear condition in the fear- and no-fear comparison6. In addition these results may be due to incorrect use of fear appeal, or maybe other aspects of the videos, such as quality. These flaws can bring down the viewer’s appreciation and
engagement with the message, which means people will not watch the video and therefore not have a response.
When looking at the mediating power of transportation, results showed that transportation did not media the effect of fear appeal. This was once more against
expectations as previous studies have shown that transportation is one of the best predictors of changes in affect, knowledge and behavioural intent (Moyer-Gusé, 2008; Murphy et al., 2011). The reason may however be related to the reason presented above, the fact that the
6
23
video with peers and fear really was the odd one out. Even though transportation did not mediate the effect of fear appeal on viewers’ responses, appreciation did have mediating power. Results showed that appreciation had a positive effect on the relationship between fear appeal and knowledge gain of viewers. Even though fear did not directly predict knowledge to begin with, the model proved that there was an effect via appreciation. The model shows that when there is no fear appeal there is more appreciation, and when there is more appreciation there is more knowledge gain. Again, these results are in line with research by Reeve (1989) and Adler and Fagley (2005), who found that when a message is appreciated there is a greater chance that the viewer will continue looking at the message and take in the information. As mentioned before, this implies that it is not just the use of fear that we should look at, but that it might be more important to make sure people appreciate the message regardless of its content.
To answer the second sub-question and part of the third and fourth: fear has no direct influence on viewers’ responses. However, when appreciation is taken into account, it does have an indirect effect on viewers’ knowledge gain. Even though fear appeal negatively influences appreciation, appreciation positively influences knowledge gain off viewers.
Limitations
As a lot of the results were either unexpectedly not significant or unexpected in general, the decision was made to address this issue in the limitations section. The reason so many results came back as not significant or as significant but in an unexpected way, might be due to some of the following limitations of this research. First of all, the sample was relatively small for an experiment with four conditions, and the conditions were not equally distributed. Both of these flaws can have a negative impact on the results, because if the sample is bigger a bit of unequal distribution is not necessarily an issue. In addition, the sample was less diverse than
24
planned, which may have made the data quite one sided. Secondly, in hindsight the videos used as stimulus material may not have been completely suitable for the experiment. The videos were from four different platforms, ranging in quality and general appearance. If this research were to be repeated, it would be smarter to either create videos from scratch, or to edit all four conditions from one longer clip. This would make the videos more comparable and the research more likely to succeed.
25 Conclusion
To summarise, several questions were posed at the beginning of this study, but the main question that needed to be answered was: How should young adults be educated on the topic of XTC and drugs in general? It was expected that the format of Drugslab would have the biggest positive effect on the participants’ responses. However, this study found evidence against this expectation, as the presence of peers actually had a negative effect on viewers’ knowledge gain. Interestingly though, this study also did not find proof that traditional formats using fear appeal were a better solution, as fear appeal seemed to have no influence on viewers’ responses at all. When it came to videos with fear appeal however, the video with both fear appeal and peers scored the worst when it came to knowledge gain. The general consensus seems to be that peer presence has a negative impact on viewers’ responses, but that the outcome is even worst when these peers convey the fear appeal aspect. Looking purely at this research, the best practice for creating such videos is the following: if you want to increase young adults’ knowledge do not use peers, especially not when you also want to use a fear appeal. In other words, a solution may be to simply stick to simple text-based messages that provide information. That way the focus is completely on the information and no distraction is created by the presence of peers or the use of fear appeal. In case you do decide to use either of these, at least make sure that you have a good idea of what you audience can appreciate, because this might help to turn the effect around and lead to an increase in knowledge in the end.
In general this research yielded much less results than anticipated. It is important to be careful with generalizing the findings of this research. First of all a lot of it is contradictory to already existing research. This in itself does not have to be an issue; it could mean that there are some unforeseen processes at play that need to be taken into account in future research. However, as can be read in the limitations there might be some other, more pressing,
26
problems with the set-up of the research. For now, to answer the overall research question, the presence of peers does not seem to work when you want to educate young adults about XTC, especially when these young adults use fear appeal to do so. However, results also imply that creators behind these messages should think very carefully what audience they are trying to reach and what they will and will not appreciate in order for them to get their information across. To make sure the Dutch youth are no longer the most frequent users of XTC in Europe providing basic information, which they can appreciate, may be the best way to get them under the influence of the message.
27 References
Adler, M. G., & Fagley, N. S. (2005). Appreciation: Individual differences in finding value and meaning as a unique predictor of subjective well-being. Journal of Personality, 73(1), 79–114. http://doi.org/10.1111/j.1467-6494.2004.00305.x
Alan, T. (2016). Ecstasy in comeback as new generation discovers dance drug. The Guardian. Retrieved from https://www.theguardian.com/society/2016/may/31/ecstasy-comeback-european-drugs-survey
Behavioral and Hypothetical Interview Questions. Harvard Law School. Retrieved from
http://hls.harvard.edu/dept/opia/job-search-toolkit/behavioral-and-hypothetical-interview-questions/
BNN Drugslab. (2016). Drugslab. YouTube. Retrieved from
https://www.youtube.com/channel/UCvRQKXtIGcK1yEnQ4Te8hWQ BNN Drugslab. (2016). Ecstasy (XTC / MDMA) Do's en don'ts. Retrieved from
https://www.youtube.com/watch?v=Rh5q42acRW8&t=1s
Cauberghe, V., De Pelsmacker, P., Janssens, W., & Dens, N. (2009). Fear, threat and efficacy in threat appeals: Message involvement as a key mediator to message acceptance.
Accident Analysis and Prevention, 41, 276–285. http://doi.org/10.1016/j.aap.2008.11.006 Drug-Free World. (2011). Truth About Drugs Documentary: Ecstasy. Retrieved from
https://www.youtube.com/watch?v=2J8ORsBuM5U&t=52s
Drug-Free World. (2017). The truth about XTC - edited. Retrieved from https://youtu.be/zVmBOiRmiqk
Gayle, D., & Pegg, D. (2015). Ecstasy and LSD use reaches new high among young. The Guardian. Retrieved from
28
Geraci, J., & Nagy, J. (2004). Millennials ‐ the new media generation. Young Consumers, 5(2), 17-24. http://dx.doi.org/10.1108/17473610410814111
Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge and perceived realism. Discourse Processes, 38(2), 247–266.
http://doi.org/10.1207/s15326950dp3802
Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The role of transportation into narrative worlds. Communication Theory, 14(4), 311–327. http://doi.org/http://dx.doi.org/10.1111/j.1468-2885.2004.tb00317.x
Laroche, M., Toffoli, R., Zhang, Q., & Pons, F. (2001). A cross-cultural study of the persuasive effect of fear appeal messages in cigarette advertising: China and Canada. International Journal of Advertising, 20(3), 297–313.
http://doi.org/10.1080/02650487.2001.11104895
Lin, A., Gregor, S., & Ewing, M. (2008). Developing a scale to measure the enjoyment of Web experiences. Journal Of Interactive Marketing, 22(4), 40-57.
http://dx.doi.org/10.1002/dir.20120
List of Emotions (p. 1). Retrieved from http://www.therapistaid.com/worksheets/list-of-emotions.pdf
Millennials prefer new media over the TV and radio. (2016). Consultancy.uk. Retrieved from http://www.consultancy.uk/news/3223/lek-millennials-prefer-new-media-over-the-tv-and-radio
Moyer-gusé, E. (2008). Toward a theory of entertainment persuasion: Explaining persuasive effects of entertainment-education messages. Communication Theory, 18, 407–425. http://doi.org/10.1111/j.1468-2885.2008.00328.x
Murphy, S. T., Frank, L. B., Moran, M. B., & Patnoe-woodley, P. (2011). Involved,
29
and behavior in entertainment-edcuation. Journal of Communication, 61, 407–431. http://doi.org/10.1111/j.1460-2466.2011.01554.x
Nabi, R. L., Roskos-Ewoldsen, D. R., & Dillman Carpentier, F. (2008). Subjective knowledge and fear appeal effectiveness: Implications for message design. Health Communication, 23(2), 191–201. http://doi.org/10.1080/10410230701808327
Nederland blijft in top drugsgebruik Europa. (2017). Nrc.nl. Retrieved from
https://www.nrc.nl/nieuws/2017/06/06/nederland-blijft-in-top-drugsgebruik-europa-a1561842
Nederlandse jongeren bovenaan Europese xtc-lijst. (2017). Nos.nl. Retrieved from
http://nos.nl/artikel/2176930-nederlandse-jongeren-bovenaan-europese-xtc-lijst.html Nederlandse jongeren gebruiken vaakst xtc. (2017). RTL Nieuws. Retrieved from
https://www.rtlnieuws.nl/nederland/nederlandse-jongeren-gebruiken-vaakst-xtc Oliver, M. B., & Bartsch, A. (2010). Appreciation as audience response: Exploring
entertainment gratifications beyond hedonism. Human Communication Research, 36(1), 53–81. http://doi.org/10.1111/j.1468-2958.2009.01368.x
Oliver, M. B., & Raney, A. A. (2011). Entertainment as pleasurable and meaningful:
Identifying hedonic and eudaimonic motivations for entertainment consumption. Journal of Communication, 61(5), 984–1004. http://doi.org/10.1111/j.1460-2466.2011.01585.x Paek, H., Hove, T., Jeong, H. J., & Kim, M. (2011). Peer or expert? International Journal of
Advertising, 30(1), 161–188. http://doi.org/10.2501/IJA-30-1-161-188
Reeve, J. (1989). The interest-enjoyment distinction in intrinsic motivation. Motivation and Emotion, 13(2), 83–103. http://doi.org/10.1007/BF00992956
Riet, J. Van, & Ruiter, R. A. C. (2013). Defensive reactions to health- promoting information: an overview and implications for future research. Health Psychology Review, 7(1), 104– 136. http://doi.org/10.1080/17437199.2011.606782
30
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, 18(3), 277–287. http://doi.org/http://dx.doi.org/10.1111/j.1559-1816.1988.tb00017.x Roskos-Ewoldsen, D. R., Yu, J. H., & Rhodes, N. (2004). Fear appeal messages affect
accessibility of attitudes toward the threat and adaptive behaviors. Communication Monographs, 71(1), 49–69. http://doi.org/10.1080/0363452042000228559
Rudloff, A. Complete List of Behavioral Interview Questions. Retrieved from
https://dornsife.usc.edu/assets/sites/780/docs/01_aje_schw_oys_behavior.pdf
Ruiter, R. A. C., Abraham, C., & Kok, G. (2001). Scary warnings and rational precautions: A review of the psychology of fear appeals. Psychology & Health, 16(6), 613–630.
http://doi.org/10.1080/08870440108405863
Sekaran, U. (2003). Research methods for business: a skill-building approach (4th ed.). New York, USA: John Wiley & Sons.
Trimbos-Instituut. (2014). Ongelukken met XTC, wat gaat er mis?. Retrieved from https://youtu.be/_0D9ou7e3hA
Unity. (2013). Wat is XTC? Animatie van Unity. Retrieved from https://youtu.be/kvabSJrKIhw
Valkenburg, P. M., & Cantor, J. (2000). Children’s likes and dislikes of entertainment
programs. In D. Zillmann & P. Vorderer (Eds.), Media entertainment: The psychology of its appeal (pp. 135-152). Hillsdale, NJ: Erlbaum.
Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model. Communication Monographs, 59, 329–349. Retrieved from
https://msu.edu/~wittek/fearback.htm
Wolf, C. R., & Pulerwitz, J. (2003). The influence of peer versus adult communication on AIDS-protective behaviors among Ghanaian youth. Journal of Health Communication,
31 8(5), 436–474. http://doi.org/10.1080/713852119
32
Appendix A: Online Research Drug related entertainment education
Q15 Hallo daar! Wat fijn dat je tijd hebt genomen om deel te nemen aan mijn onderzoek naar voorlichtingsfilmpjes over drugs. Het onderzoek duurt niet langer dan 5 minuten en je
deelname wordt zeer gewaardeerd. De antwoorden die je geeft, worden alleen gebruikt voor dit onderzoek en zullen anoniem blijven. Er zijn geen verkeerde antwoorden, zolang je eerlijk antwoord geeft, is alles goed. Verder kun je ook elk moment stoppen met het onderzoek. Wanneer je vragen of opmerkingen hebt, kun je contact opnemen door een mail te sturen naar yara.dekwaasteniet@student.uva.nl. Als je deelneemt aan dit onderzoek, doe dit dan
alsjeblieft op een computer of een tablet, je zult namelijk een kort filmpje (max 2 minuten) moeten kijken. Als bedankje verloot ik twee €10 Bol.com bonnen onder de deelnemers! Aan het eind kan je je email adres achter laten. Ik heb het bovenstaande gelezen en begrepen, en door op de 'verder' knop te klikken, geef ik aan vrijwillig deel te nemen aan deze studie. Ja, start het onderzoek. (1)
Nee, stop het onderzoek. (2)
Condition: Nee, stop het onderzoek. Is Selected. Skip To: End of Survey.
Q32 Timing First Click (1) Last Click (2) Page Submit (3) Click Count (4)
Trimbos Neem de tijd om naar onderstaande video te kijken. Zorg dat je geluid aan staat.
Drugslab Neem de tijd om naar onderstaande video te kijken. Zorg dat je geluid aan staat.
Unity Neem de tijd om naar onderstaande video te kijken. Zorg dat je geluid aan staat.
33
Q26 Welk van onderstaande beelden kwam voor in het filmpje dat je net hebt gezien? 1 (1)
2 (2) 3 (3) 4 (4)
exposure Had je het zojuist getoonde filmpje al eerder gezien? Ja (1)
Nee (2)
34
Q19 [transportation] The volgende vragen gaan over je betrokkenheid bij het filmpje. Heel erg mee
oneens (1) Beetje mee oneens (2) Niet mee eens, niet mee oneens (3) Beetje mee eens (4)
Heel erg mee eens (5) Tijdens het kijken van het filmpje was ik snel afgeleid. (1) Tijdens het kijken van het filmpje voelde ik me mentaal betrokken. (2) Wat er besproken werd in het filmpje zorgde ervoor dat ik wilde blijven kijken. (3) Ik voelde me betrokken bij wat er in het filmpje besproken werd. (4) Ik kon me inleven in wat er verteld werd in het filmpje. (5)
35
Q30 [appreciation] De volgende vragen gaan over je waardering van het filmpje. Heel erg mee
oneens (1) Beetje mee oneens (2) Niet mee eens, niet mee oneens (3) Beetje mee eens (4)
Heel erg mee eens (5) De inhoud in het filmpje vond ik waardevol. (1) Ik heb met plezier naar het filmpje gekeken. (2) Het filmpje vond ik waardevol. (3) Ik vond het een leuk filmpje. (4) Het filmpje interesseerde me niet. (5) Het filmpje was goed gemaakt. (6)
36
Q16 [authenticity] De volgende stellingen gaan over de authenticiteit van het filmpje. Heel erg mee
oneens (1) Beetje mee oneens (2) Niet mee eens, niet mee oneens (3) Beetje mee eens (4)
Heel erg mee eens (5) Het filmpje kwam authentiek over (1) Het filmpje leek oprecht. (2) Het filmpje was geloofwaardig. (3) Het filmpje kwam professioneel over. (4) Het filmpje kwam objectief over. (5)
37
Q25 [emotions] Het volgende deel gaat over je gevoel bij het gebruik van XTC. Na het kijken van het filmpje, lijkt het gebruik van XTC mij:
1 (1) 2 (2) 3 (3) 4 (4) 5 (5) 6 (6) 7 (7) Onplezierig:Plezierig (1) Eng:Niet eng (2) Onverantwoord:Verantwoord (3) Spannend:Niet spannend (4) Gevaarlijk:Niet gevaarlijk (5) Onverstandig:Verstandig (6) Beangstigend:Niet beangstigend (7) Onveilig:Veilig (8) Laf:Stoer (9)
Q22 [knowledge] De volgende stellingen gaan over wat je geleerd hebt na het kijken van het filmpje. Na het kijken van het filmpje, weet ik meer over:
Heel erg mee oneens (1) Beetje mee oneens (2) Niet mee eens, niet mee oneens (3) Beetje mee eens (4) Heel erg mee eens (5) XTC in het algemeen (1) De werking van XTC (2) Verantwoordelijk gebruik van XTC (3) Onverantwoordelijk gebruik van XTC (4) De positieve gevolgen van XTC (5) De negatieve gevolgen van XTC (6)
38
Q28 Op hoeveel dagen in je leven heb je XTC gebruikt? Nooit (1)
1-2 dagen (2) 3-10 dagen (3) 11-25 dagen (4) 26-50 dagen (5) Meer dan 50 dagen (6)
Q27 [behavior] De volgende vragen en stellingen gaan over toekomstig XTC gebruik. Zeker niet
(1)
Misschien niet (2)
Weet niet (3) Misschien
wel (4) Zeker wel (5) Ben je van plan in de toekomst XTC te gebruiken? (1) Als mijn vrienden mij XTC zouden aanbieden, zou ik het aannemen. (2) Ik heb zin om XTC te gebruiken. (3) Als mijn vrienden het mij zouden vragen, zou ik ze afraden om XTC te gebruiken. (4)
39 age Wat is je leeftijd in jaren?
Gender Wat is je geslacht? Man (1)
Vrouw (2) Anders (3)
Q11 Wat is je hoogst voltooide opleiding? Basisonderwijs (1)
Voorbereidend beroepsonderwijs (VMBO) (2)
Middelbaar algemeen voortgezet onderwijs (MAVO) (3) Middelbaar beroepsonderwijs (MBO) (4)
Hoger algemeen voortgezet onderwijs (HAVO) (5) Voorbereidend wetenschappelijk onderwijs (VWO) (6) Hoger beroepsonderwijs (HBO) (7)
Wetenschappelijk onderwijs (WO) (8)
Anders, namelijk (9) ____________________
Q12 Als je mee wil doen aan de loting van de €10 Bol.com bon (2x), vul dan hier je email adres in.
Q13 Nogmaals bedankt voor je deelname. Als je nog vragen of opmerkingen hebt, of wanneer je meer wilt weten over mijn onderzoek, kan je hier een bericht achterlaten of een mailtje sturen naar yara.dekwaasteniet@student.uva.nl.
