To panic, or not to panic
Moral panics, political campaigns, and their long-lasting effects on electoral support: the case of the classic psychedelics
Mathilde Kamer 11263601
MSc. Political Communication Thesis supervisor Dr. Rachid Azrout
June 29th, 2022
9711 words (excl. tables, figures and footnotes)
Additional words approved by Dr. Azrout due to the novel operationalization of the theoretical concept of ‘moral panic’, in order to elaborate on the operationalization and discuss the
substantial contribution to the scholarship of moral panic theory
I would firstly like to thank my thesis supervisor Dr. Rachid Azrout. I deeply appreciate your constructive feedback and the way that you pushed me to do research to the very best of my abilities. I am grateful for the feedback and support that I have received from you over the past few months. You motivated me to go the extra mile and allowed my creativity to shape this MSc thesis into the final product that it has become. I want to furthermore thank the friends I made this year who made this program into such a joy, and helped me when I lost perspective during the writing of this thesis – specifically Sajal, Charlie, Gabby and Reinout. Our days (and nights) spent in the library remain some of the fondest memories I have of this past year.
I also must thank my very best friend Freek Wallagh, for once again inspiring me to pursue a fiery passion, the flame of which you lit personally. While my passion for American politics and intrigue in political power and regulation were already there, your birthday gift for my 23rd birthday – Food of the Gods, A Radical History of Plants, Psychedelics and Human Evolution by Terence McKenna – introduced me to a whole new field of (politicized) research that I was previously unaware of.
Your genuine belief in me consistently leads me to new and uncharted territories, and I enjoy the exploration of those with you by my side immensely.
This year, I submerged myself in psychedelic research through podcasts, books (notably the one by McKenna as mentioned above, but also the book How to Change your Mind, The New Science of Psychedelics written by Michael Pollan, which will make multiple appearances in the thesis that lies in front of you) and lectures. I have become a member of the Amsterdam Psychedelic Research Association (APRA), where I found a group of likeminded, enthusiastic young researchers who warmly welcomed me and tried to help me further my research wherever possible. I explicitly want to thank Paloma David, who, as head of research within the organization, helped shape my study and embed it in existing psychedelic research through inspiring personal meetings. APRA furthermore allowed me to use their extensive network to spread my survey, an act of kindness for which I am still grateful. I am excited to see what my future as Graduate Researcher at APRA will bring.
Lastly, my wonderful family cannot remain unmentioned. I especially want to thank my parents for opening themselves up to a field of study which may not have been one that you expected me to become interested in. Thank you for your openness, your willingness to learn and for always being my sounding board and helping me along. It is because of you that I was able to come into contact with Robert Chesal, the fantastic journalist who agreed to be my voice over and fictional politician, and who wanted to take the time to create a modern psychedelic panic with me.
I am deeply grateful for his enthusiasm and kindness. I must also thank Michel Brands for his guidance and help with the video editing of my campaign clips.
I have learned more in the past year than I will be able to summarize in these acknowledgements. One of the main takeaways must be that while the plans that you make sometimes fall through, the opportunities that arise in their absence can be just as magnificent.
Who knows? Maybe the 2-year MSc Political Communication at American University in Washington DC would not have led me to want to become a psychedelic researcher. All in all, the work that lies in front of you has brought me great joy and personal growth over the past months.
I am proud to, at least for now, end my time at the University of Amsterdam with a work that is so uniquely my own.
Research into moral panics is widespread. An exploration of the long-lasting political effects of moral panics and their institutional legacies, however, is lacking. Using the case of the moral panic surrounding classic psychedelics in the 1960s and 1970s, this study attempts to fill this gap in the scholarship by conducting a 5 × 3 × 2 (moral panic × age × lifetime experience with classic psychedelics) quasi-experimental experiment, where the effects of traditional moral panic rhetoric (IV) will be tested on respondents’ electoral support (DV). This study finds that using
‘traditional moral panic rhetoric’, when talking about the topic of medical use of classic psychedelics, leads to a sharp decrease in electoral support compared to using non-moral panic (neutral) rhetoric. This finding holds for respondents who were socialized during the moral panic’s institutional legacy and during the psychedelic renaissance. Respondents who were socialized during the moral panic in the 1960s respond more positively to the traditional moral panic rhetoric, but this finding is not statistically significant. The negative effect of traditional moral panic rhetoric is significantly stronger for respondents who have lifetime experience with classic psychedelics.
The implications of these findings are broad for moral panic theory: this study shows that there is indeed a moral panic legacy which can be triggered decades later; public opinion towards the (subject of the) moral panic can change over time and generations; and politicians as agents of facilitation or exacerbation of moral panics are shown to be vulnerable to a backlash effect similar to one that can also be found in negative campaigning. Further research should be conducted into the question if the concept of moral panic can be interpreted as being hierarchical.
On the 3rd of February, 1966, Dutch anarchist protest movement Provo spread rumors that LSD would be administered to police horses and the water supply of the city of Amsterdam to protest the marriage between (Queen) Beatrix of the Netherlands and Claus van Amsbergen1. The panic this created directly led to the criminalization of hallucinogenic substances under the Dutch opium law on February 8th, 1966 – a mere five days after the spread of the rumor (Roes, 2016).
This exemplifies the extremely high anxiety surrounding the perceived threat of psychedelics in society at the time. In the past years, psychedelic research is going through a ‘renaissance’ (Sessa, 2018; Pollan, 2019). Groundbreaking results suggest that hallucinogenic substances like psilocybin2 can revolutionize the mental health industry within the next decade (Stichting GGZ Groep, n.d.).
These findings are in stark contrast with the arguments given during the criminalization of classic psychedelics in the 1970s; they were claimed to be unfit for medical use and told they had no medicinal benefits (Tupper et al., 2015; Richert & Dyck, 2020).
This study will attempt to explore a specific phenomenon in political communication surrounding controversial political topics. Specifically, controversial topics that have been deemed as ‘deviant’ and have been subject to ‘moral panics’3. Moral panics are short periods of widespread anxiety among citizens about a perceived threat, which in hindsight is not as threatening as was
1 Provo was known for ‘playful’ (ludieke) political actions through which they attempted to provoke authorities to break their own rules. They were prodemocratic and anti-authoritative, and part of the counterculture movement which spread from the US to other Western countries in the 1960s. The forthcoming marriage between Beatrix der Nederlanden and Claus von Amsbergen was controversial, because Von Amsbergen was a former member of the Hitlerjugend and NSDAP-Jungvolk. The marriage was only allowed after the Dutch parliament established that Van Amsbergen had committed no war crimes during World War II, since he served in the German army as a German citizen. Within Dutch society the forthcoming marriage created a lot of social unrest and protest. For example, protestors painted orange swastikas in cities across the Netherlands with the text “Claus, go back to your Heimat”
(German for “hometown”), and “Raus Claus” (“Get out Claus”) could be seen painted on buildings across Amsterdam (Nieuwsblad van het Noorden, 1965). Provo played into this social unrest as well as the psychedelic moral panic. They created their own fake “Orange Committee” to disseminate false rumors about the spread of hallucinogenic substances in the water supply of the city of Amsterdam, as well as sugar cubes laced with LSD which would be given to police horses by unsuspected children on the royal wedding day. Five days later, on February 8th, 1966, LSD was made illegal in the Netherlands. This is an excellent example of the zeitgeist surrounding psychedelics at the time and the general
‘psychedelic panic’ of the Dutch cabinet.
2 Psilocybin is the active component in ‘magic mushrooms’ or ‘shrooms’ (fungi).
3 E.g. gay marriage, the Red Scare, and more recently, the anxiety surrounding Critical Race Theory in the United States.
4 claimed and perceived at the time of the moral panic (Cohen, 1972; Goode & Ben-Yehuda, 2009).
While moral panics are characterized by a volatility that means that they come up and disappear quickly (Goode & Ben-Yehuda, 2009, p. 41), the subjects of moral panics seem to occasionally return to the political stage in a new light. So, too, is the case of classic psychedelics; subject to what many scholars deem a moral panic in the 1960s4 (Goode, 2008; Goode & Ben-Yehuda, 2012;
Vitiello, 2020). This inspires the question: what kind of legacies can a moral panic leave, and what kind of contemporary effects can they have when the subject re-enters the political field at a later time as a campaign topic?
Research in the field of political psychology suggests that not everyone may respond the same to being presented with traditional moral panic rhetoric5. They show, for example, the importance of respondents’ earlier life experiences (Bower, 1981), and how they moderate the neurological process of emotions (Hebb, 1949; Collins & Loftus, 1975), which political campaigns often try to trigger in order to persuade citizens’ electoral support (Brader, 2005). In relationship to moral panics reentering the political field, it is interesting to see how respondents’ lifetime experience with the substance in question influences their receptibility to the traditional moral panic rhetoric and their subsequent electoral support. For this specific case, it is therefore expected that lifetime experience with classic psychedelics (LECP) is a predictor for the influence of the moral panic rhetoric on electoral support. Because moral panics are widespread and volatile, they naturally influence people in the time that they are active. However, moral panics can leave a temporary6 or
4 Pollan (2019) illustrates why psychedelics created such a (moral) panic: “[…] Psychedelics introduced something deeply subversive to the West that the various establishments had little choice but to repulse. LSD truly was an acid, dissolving almost everything with which it came into contact, beginning with the hierarchies of the mind (the super ego, ego and unconscious) and going on from there to society’s various structures of authority and then to lines of every imaginable kind: between patient and therapist, research and recreation, sickness and health, self and other, subject and object, the spiritual and the material. If all such lines are manifestations of the Apollonian strain in Western civilization, the impulse that erects distinctions, dualities, and hierarchies and defends them, then psychedelics represented the ungovernable Dionysian force that blithely washes all those lines away” (Pollan, 2019, p. 214).
5 Traditional moral panic rhetoric refers to rhetoric that is similar to the rhetoric that was used at the time of the original moral panic. In this specific case, that means the same or very similar (false) stereotypes about users, misinformation about the potentially harmful effects of the substances and the (exaggerated) negative effect psychedelic users and the substances themselves supposedly had on society.
6 For example the imposing of “fines of as much as $1000 and sentences of up to one year behind bars for possession of the hallucinogenic drug LSD” (Time Magazine, 1966) in Nevada and California in 1966.
5 long-lasting institutional legacy (Goode & Ben-Yehuda, 2009, p. 247) – for example the ongoing criminalization of classic psychedelics to this day. Diving into the moderating effects of age and how different age groups respond to being presented with traditional moral panic rhetoric, can help create an understanding of the effects of the moral panic, and the institutional legacy of a moral panic. To explore these facets of moral panic theory, the following two research questions have been formulated:
RQ1: “To what extent does falling back on ‘traditional moral panic rhetoric’ in campaigns influence citizens’
electoral support compared to using neutral rhetoric, and to what extent is this process moderated by lifetime experience with classic psychedelics?”
RQ2: “How does age moderate citizens’ electoral support when being presented with ‘traditional moral panic rhetoric’, and how does lifetime experience with classic psychedelics influence this relationship?”
These questions will be answered by conducting a 5 × 3 × 2 (moral panic × age × lifetime experience with classic psychedelics) quasi-experimental experiment, where the effects of the moral panic rhetoric (IV) will be tested on respondents’ electoral support (DV). The academic relevance of this study is twofold. This study will help deepen moral panic theory by both operationalizing the concept of moral panic and manipulating it in an experimental study, as well as applying it to a political campaign of a singular political actor. Both are new approaches to applying moral panic theory in the field of political communication. Moral panic theory tends to focus on mass-media and culture as the most influential actors in the eruption of moral panics (Béland, 2005, p.2), even though Goode and Ben-Yehuda – influential scholars in the field of moral panic – identify politicians as key actors in the elite-engineered creation of moral panics (2009, p.54−55). Little research has been conducted into the role of individual politicians and their campaigns surrounding topics that are, or have been, subject of moral panic. This study attempts to fill this gap in the
6 scholarship by exploring the effects of a politician using traditional moral panic rhetoric when discussing a controversial topic, which reappears on the political stage after lying dormant for a significant time. The case of psychedelics lends itself to understand this process through a deductive approach, and add to the scholarship since the moral panic surrounding it in the 1960s and 1970s was widespread and still has lasting effects on citizens’ perception of the substances (Sessa, 2012). Furthermore, this study attempts to explore an under researched aspect of moral panic theory: the moral panics’ (institutional) legacy and its long-term effects in society. Goode and Ben-Yehuda (2009) distinguish two potential outcomes of a moral panic; they can either leave an institutional legacy, or no trace at all (p.247). When it does leave a trace, Goode and Ben-Yehuda say that this institutional legacy can help stimulate ‘early concerns’ at a later point in time. How this works exactly is unclear. The second research question of this study takes an inductive approach to this subfield of moral panic theory and attempts to translate the findings of this study into an initial thought on how this process could potentially work.
Because moral panics are generally characterized by their rapid rise and decline, the case of psychedelics being destigmatized enough to be approved for medical use offers fruitful ground for research in the context of moral panic theory. Generally, the moral panics in our history leave our collective minds and don’t come back as time (and often, knowledge) progresses7. Only occasionally does a moral panic return to the political arena. The psychedelic research renaissance that is currently happening in countries across the world is a presage of a political discussion that is bound to be forthcoming; the stigmas and preconceptions of what psychedelics do to society and their users, which led to the criminalization of the substances, ensure that the decriminalization of psychedelics is a controversial topic. Combined with the financial interest that the pharmaceutical lobby (in especially the United States) has in psychedelics8, this case may very well
7 Examples of this are the (racist) moral panics surrounding former President Barack Obama’s election (Scheitle et al., 2019) or the entire genre of Jazz music being “the devil’s music” (Peters, 2019).
8 It is expected that Big Pharma will attempt to profit off of these promising new therapies (Carhart-Harris, 2021).
Especially the potential that psychedelics have to replace (or supplement) the highly profitable SSRI (selective serotonin reuptake inhibitors – antidepressants) medication whose efficacy and safety are questioned. While SSRIs are
7 be (re)politicized soon. Understanding the lasting effect of decades old moral panic rhetoric on people’s contemporary perception of psychedelics, their potential medical uses and people’s subsequent voting behavior allows us to understand the complexity of politically controversial issues, and how to effectively communicate about them. This is extremely relevant to the field of political communication. It is known that fear appeals and negative framing influence citizens’
voting behavior (see e.g. Brader, 2005). Implied in the name, fear appeals and negative frames are fundamental in moral panic rhetoric. While the case of psychedelics is the example chosen in this study, the findings can hopefully contextualize political campaign- and communication effects on other controversial topics that have been subject to moral panics such as the ones mentioned above.
Goode and Ben-Yehuda’s criteria and Cohen’s agents
Goode and Ben-Yehuda (2009) take a social constructivist approach to the theory of moral panic. Moral panics, according to them, “are not a simple reflection of the objective harm of threats, but are socially and culturally constructed by partly or wholly unfounded fears” (p.63).
They identify five subsequent stages of moral panic (p.37−43); based on the original ones posed by Cohen in 19729. These five stages are also described as criteria, which together create a fully-
prescribed to generally be taken at least once a day, a revolutionary study conducted by John Hopkins University shows that the positive and anti-depressant effects of just two doses of psilocybin spread out over a 5-week period are still present six month after use for almost 8 in 10 respondents (Johns Hopkins Medicine, 2016). A study published in 2022 shows that the treatment effect for people with major depression is effective for up to a year for most patients (Johns Hopkins Medicine, 2022). Patents on many conventional SSRIs are about to expire (Carhart-Harris, 2021), so Big pharma has strong financial stakes in patenting psychedelics and turning them into a medication that they can attempt to make as profitable as SSRIs.
9 Stephen Cohen (1972) defines a moral panic as a period of widespread fear among citizens which is disproportional to the actual threat the thing poses to a community or country. Moral panics are complex and hard to grasp as a concept: “[the conditions] are new (lying dormant perhaps, but hard to recognize; deceptively ordinary and routine, but invisibly creeping up the moral horizon) – but also old (camouflaged versions of traditional and well-known evils). They are damaging in themselves – but also merely warning signs of the real, much deeper and more prevalent condition. They are transparent (anyone can see what’s happening) – but also opaque: accredited experts must explain the perils hidden behind the superficially harmless (decode a rock song’s lyrics to see how they led to a school massacre)” (Cohen 1972, p. viii). Cohen identifies five subsequent stages in a society that lead to a moral panic (1972, p. 1). First, a “condition,
8 fledged moral panic. According to Goode and Ben-Yehuda, moral panics start with a heightened level of concern about the behavior of a group of people, and especially the “consequences that that behavior presumably causes for one or more sectors of society” (p.37). This concern assumes that the behavior or condition is inherently threatening. The second criterion, hostility, builds on this presumed threat and extends the concern into hostility against the group that presumably engages in, or aligns themselves with, the threatening behavior. This is the basis for the division between
“us – the good, decent, respectable folk – and ‘them’ or the ‘Other’ – the deviants, bad, […]
disreputable folk” (p.38). The third criterion is consensus; the widespread agreement within a society that the threat is “real, serious, and caused by the wrongdoing group members and their behavior”
(p.38). Goode and Ben-Yehuda make sure to distinguish different types of moral panics here – according to them, they come in different ‘sizes’ and are a “matter of degree” (p.39). Consensus need not mean the entire population, but it needs to be widespread to be classified as a moral panic.
The fourth criterion is disproportionality: the societal, legal, media and political response to the threat is not proportionate to the actual threat it poses. The fifth and final criterion of moral panic is volatility. According to Goode and Ben-Yehuda, moral panics erupt and subside suddenly and can leave either no trace, or result in institutionalized or routinized changes in the law, social policy or cultural norms and values (p.41).
Cohen (1972) identifies four key agents that facilitate and exacerbate moral panic: mass media, who through deviance amplification10 redefine the problem to become bigger than it actually is;
episode, person or group of persons” (p. 1) emerges which is experienced as potentially posing a threat to a society’s values and interests. Their behavior is seen as deviant from the norm and untraditional. Secondly, the mass media confirms the notion of deviance from the norm by defining the behavior as such, creating a ‘folk devil’ and an
‘innocent’. This leads to the third stage; a response from ‘moral entrepreneurs’, who man the “moral barricades” (p.
1); bishops, priests, politicians, and other ‘right-thinking’ people claim to be the voice of reason and morality. The moral panic, in this stage, is widespread enough to create a consensus among citizens that the problem is an actual threat; causing the ‘moral entrepreneurs’ offer their opinions and solutions. This, in turn, leads to stage four; a widespread anxiety, resulting in new laws or social policy based on the diagnosis given by ‘experts’ on the issue. These institutional legacies are based on the increased risk perception at the height of the moral panic. The last stage is reached when the condition disappears, and the panic dies down (p. 1−2).
10The focus on the initial condition invites other events that confirm a ‘pattern of deviance’ to suddenly become newsworthy (p. 58) – based on news values such as follow-up, (perceived) magnitude or drama (Harcup & O’Neill 2016, p. 1471, 1482; Goode & Ben-Yehuda 2009, p. 79). This creates highly sensationalized news reports which feed citizens’ anxiety and distorts their perceived size of the problem (Goode & Ben-Yehuda 2009, p. 77).
‘societal control culture’, which refers to the “laws, procedures, programs and organizations which in the name of a collectivity help, rehabilitate, punish or otherwise manipulate deviants” (Lemert 1952 as cited in Cohen, 1972, p.78); the public, whose concern leads to the creation of a “new fragment of the moral constitution of society” (Becker 1966 as cited in Cohen, 1972, p.3); and moral entrepreneurs, who spring up to morally shape the deviant condition. Politicians fall within this last category. They place the condition in a much larger context, alluding to the anxiety- inducing idea that the condition is actually a symptom of a much bigger problem (Cohen, 1972, p.141).
Fear appeals and moral panic in political campaigning
Goode and Ben-Yehuda (2009) furthermore identify three different explanatory models of the eruptions and creations of moral panic: interest-group, grass-roots and elite-engineered. While the grass-roots and interest-group models argue that moral panics erupt in either society’s middle strata (the media, religious groups or social movements – the interest-group model) or at the hands of regular citizens that are concerned about a specific topic (the grass-roots model), the elite- engineered model argues that it is the ruling class who orchestrate moral panics for their own material, political or financial gain (2009, p.54−55). Because this current study attempts to broaden the moral panic scholarship by diving into the relationship between individual politicians, their campaigns and moral panic, it is particularly this elite-engineered model that is relevant to this current study. It is typical for the elite-engineered model to see moral panics as diversion from societies ‘real problems’ (p.62) – such as for example the growing anti-war protests and the civil rights movement in the 1960s in the United States, which coincided with the hippie movement and the country’s widespread use of classic psychedelics (Belouin & Henninfield, 2018, p.9). The elite- engineered model argues that the ruling elite have great power in shaping laws and regulations and influencing public opinion.
10 The scholarship on moral panic does elaborate on political campaigns, but defines them in their relation to the political creation of (political) deviancy: moral-political campaigns. Ben-Yehuda (1990) conceptualizes moral-political campaigns as “revolv[ing] around negotiations of power and morality [which aim] to shame, vilify and deviantize other specific subpopulations” (p.114). In the moral panic scholarship, campaigning thus seems to refer to a broader, elite-engineered (Goode &
Ben-Yehuda, 2009, p.97), general and normatively charged condemnation of the condition that is cause for the moral panic, which translates to a (political) power struggle. Within the scholarship, the focus lies generally on mass-media and culture as the actors in moral panics (Béland, 2005, p.2).
The scholarship lacks an exploration of the effect of moral panic rhetoric in a politicians’ individual campaign, which is not necessarily an extension of the elites’ more general normative standpoint on the subject. Acknowledging the lack of theory on this specific subject, Béland (2005) researched the specific role political actors and strategies have on the formation of collective threat perception.
According to Béland, politicians inhibit the space between personal anxieties and the political agenda. They function as actors that influence agenda-setting; by defining perceived threats as
“collective problems affecting a significant segment of the population” (p.4).
Research has shown that voters can be persuaded through political campaigning by appealing to their emotions (Brader, 2005; Kamer, 2021). Especially fear cues have been shown to have great power of persuasion, because they trigger neurological associations in the brain (Bower, 1981). These neurological associations are made through the parallel affective system, which links concepts in the brain based on their affective valence; good or bad (Bower, 1981). These concepts are stored together in the brains’ long-term memory (LTM). Research has shown that a trigger of one concepts’ activation, triggers the activation of related concepts in the brains’ LTM (Hebb, 1949; Collins & Loftus, 1975, p.419). Research has furthermore shown that the brain has been taught to associate stimuli with three different types of potential outcomes: desirable, undesirable
11 or downright dangerous (Damasio, 1994). When presented with cues, the brain attempts to embed the stimulus within already existing associations, emotions and outcomes.
The research conducted by Brader (2005) shows that fear appeals in political campaign ads are effective in changing behavior of the viewer and stimulating voting behavior. Fear cues result in an increase in recalling related news and have a very strong persuasive power: respondents presented with fear cues were more likely to support the politician in the fear-cued ad, while they were initially indifferent or even leaning more towards the opponent (p.399). Cohen (1972) acknowledges that people, when presented with moral panic, “respond differently to the ‘same’
message” (Cohen, 1972, p.xx). The literature introduced above explains why: if the brain functions by triggering associated concepts, feelings and emotions when presented with stimuli, someone’s life experience and the stimuli that they experienced in the past will decide what kind of emotional response is triggered – desirable, undesirable or downright dangerous.
The emotional response that is triggered in people with psychedelic experience will thus be different to those without. The effect of fear cues on electoral support is always moderated by ones’ previous life experience (such as, in the case of this study, experience with psychedelics).
Additionally, people who are presented with blatant and contradictory misinformation about an experience they had, tend to reject it (Loftus, 1979, p.368). Frustration with the misinformation may even create feeling of resentment towards the source of the misinformation. This is similar to the ‘backlash effect’ within negative campaigning: the evaluation of the source is worsened and political support decreases (Carraro & Castelli, 2010, Lau & Rovner, 2009 and Kahn & Kennedy, 2004 as cited in Galasso et al., 2021). Literature furthermore indicates that (policy and partisan) similarity is a significant predictor for candidate preference (Bailenson et al., 2008, p.950−952), suggesting that respondents with no LECP may favor the moral panic based campaign clips compared to respondents with LECP. It is likely that respondents with lifetime experience will not recognize their experience, or themselves, in the candidate and will reject the politician, potentially
12 even triggering the backlash effect. Furthermore, advocacy to legalize psychedelics predominantly stems from the psychedelic community itself11. Accordingly, it is deemed plausible that respondents with LECP are less likely to agree with the politicians’ moral panic-based standpoints and may thus score lower on electoral support. The first hypothesis of this study will therefore be:
Hypothesis 1a (H1a): For those who have lifetime experience with classic psychedelics, being presented with traditional moral panic rhetoric will lead to lower electoral support than not being presented with traditional
moral panic rhetoric.
On the other hand, research has shown that being presented with attitude-consistent stimuli leads to issue agreement and attitude reinforcement, which leads to higher levels of electoral support (Knobloch-Westerwick et al., 2015). The preference for attitude-consistent over attitude- discrepant messages is also known as confirmation bias (Taber & Lodge, 2006) and in times of political elections, confirmation bias stimulates political participation (Knobloch-Westerwick et al., 2015, p.490). If people who are presented with blatant and contradictory misinformation about an experience they had tend to reject it (Loftus, 1979, p.368), people without LECP may be more susceptible to the fear appeals in the moral panic rhetoric and instead of rejecting the message, may even agree with it – the traditional moral panic rhetoric may trigger their deeply embedded fear for psychedelics, resulting in higher electoral support. Therefore, hypothesis H1b of this study is as follows:
Hypothesis 1b (H1b): For those who have no lifetime experience with classic psychedelics, being presented with traditional moral panic rhetoric will lead to higher electoral support than not being presented with
traditional moral panic rhetoric.
11 See for example the Multidisciplinary Association for Psychedelic Studies (MAPS), one of the worlds’ leading organizations in the attempt to revive the psychedelic research and decriminalize the substance. Their founder, is mentioned on the website, had his first LSD experience in 1972 (MAPS, n.d.).
13 Socialization theory and the moderating effect of age
Socialization theory states that “normative and deviant behaviors are learned social behaviors [and] norms for social behaviors, including drug use, are learned predominantly in the context of interactions with the primary socialization sources” (Oetting & Donnermeyer, 1998, p.995). Socialization theory furthermore identifies adolescence12 as the time where this learning takes place. It is therefore expected that age moderates respondents’ response to traditional moral panic rhetoric, but the different stages of the psychedelic panic (the moral panic itself, its legacy and the psychedelic renaissance) may result in different effects.
The disproportional characteristic of moral panics increases the perceived risk of the subject of the moral panic during the height of the moral panic (late 1960s). Research into risk perception shows that a higher perceived risk is a predictor for support for a policy that aims to reduce said risk (Stoutenborough et al., 2015). The risk perception of psychedelics at the time was high: people were told by mass-media that LSD caused birth defects in unborn babies, made users instantly mentally ill, changed youth into “psychedelic zombies”13 (Goode, 2008, p.540), led users to jump to their deaths thinking they could fly, or stop cars without getting hurt, thinking they were invincible (p.539−540). Michael Pollan, writer of the popular scientific book How to Change Your Mind, the New Science of Psychedelics (2019) said: “By the early 1970s […] everything you heard about LSD seemed calculated to terrify. It worked on me: I’m less a child of the psychedelic 1960s than of the moral panic that psychedelics provoked” (p.5). The height of the moral panic led to the criminalization of psilocybin and LSD by classifying them as a Schedule I drug (like heroin) through the Comprehensive Drug Abuse Prevention and Control Act of 1970 (Comprehensive Drug Abuse Prevention and Control Act, 1970), which can be seen as a turning point in the moral panic
12 Adolescence starts at 10 years old.
13 In moral panics surrounding synthetic (or synthesized) drugs, the ‘zombie’ angle is a common one (Jenkins, 1999).
In moral panics, these drugs are “said to represent the worst hazards of uncontrolled scientific experiment on innocent human victims” (p. 82), which would turn users into ‘monsters’ or ‘zombies’. As Jenkins says: the narratives of “…
zombie drug users […] are wonderful grabbers for media stories” (p. 185).
14 surrounding classic psychedelics – institutionalizing it. In 1971, the United Nations Convention on Psychotropic Substances followed the U.S. and also placed psychedelic drugs in their most controlled category – Schedule I – above drugs such as cocaine (UNODC, 1971).
While respondents who were adolescents during the criminalization of classic psychedelics (born in or before 1961) internalized the moral panic surrounding psychedelics, it is the generation that is born after 1961 which was socialized in a world where psychedelics were illegal and portrayed to be just as dangerous as heroin. It can also be argued that people who grew up internalizing the moral panic’s institutional legacy would respond positively to traditional moral panic rhetoric, since the fear appeals could trigger their internalization of psychedelics being extremely controlled and highly illegal. According to Goode and Ben-Yehuda (2009), the institutionalization of moral panic can be effective in “help[ing] stimulate incipient concerns later on” (p.159); implying that a moral panics’ (institutional) legacy can have long term effects on the societal response to the subject of the moral panic. However, neither Goode and Ben-Yehuda or the scholarship have elaborated on what this means exactly. Specifically researching the effects of traditional moral panic rhetoric on people who were socialized during the moral panic’s legacy can be an initial start in filling this gap in the scholarship.
As mentioned in the introduction, psychedelic research is experiencing a renaissance. The start of this renaissance has been argued to be 2006, when three subsequent events happened which broke with the four decade long stigma on psychedelics. First, Albert Hoffman – the Swiss scientist who synthesized LSD back in 1938 and who remains an icon in the psychedelic community – turned 100 years old. In his honor, a three-day research symposium was organized which was attended by Albert Hoffman himself and about two thousand psychedelic enthusiasts:
neuroscientists, healers, pharmacologists and psychiatrists from around the world flocked to Basel, Switzerland (Pollan, 2019, p.22). The second event was a U.S. Supreme Court ruling which allowed the importing of Ayahuasca, a classic psychedelic, into the United States for religious ceremonies
15 under the Religious Freedom Restoration Act of 1993 (p.27). This started a relatively steady supply chain of ayahuasca into the United States and a lack of federal prosecutions for possession of the psychedelic (p.28). The third, and according to Pollan the most influential event of the three happening in 2006, was the publication of the first “rigorously designed, double-blind, placebo- controlled study in more than four decades – if not ever – to examine the psychological effects of a psychedelic” (p.29): Griffiths et al.’s “Psilocybin Can Occasion Mystical Type Experiences Having Substantial and Sustained Personal Meaning and Spiritual Significance” paper (Griffiths et al., 2006). And, surprisingly, the publication received “a small torrent of press coverage, most of it so enthusiastic as to make you wonder if the moral panic around psychedelics that took hold in the late 1960s might finally have run its course” (Pollan, 2019, p.29). People who were socialized during this time, were socialized in a world where public opinion towards psychedelics and their potential benefits was shifting. It is expected that people who were socialized during this time do not respond as strongly to the traditional moral panic rhetoric as the other two groups, because it does not trigger related negative internalizations.
These three different phases of the moral panic – the moral panic itself, its legacy and the psychedelic renaissance – inspire the question of how age moderates respondents’ responses to being presented with traditional moral panic rhetoric. While the people who were socialized during the ‘psychedelic renaissance’ did not live through the moral panic itself and were even socialized in a world where the public opinion towards psychedelics is changing, it is unclear how the traditional moral panic rhetoric would affect respondents who lived through and were socialized during the moral panic versus respondents who grew up and were socialized in a world where psychedelics were criminalized – and for which group the effect would be strongest. Clearly, these different phases all come with their own influence on socialization of citizens’ moral evaluation of psychedelics. Because previous research has not been conducted into this process, a second, inductive, research question has been formulated to explore this gap in the scholarship:
16 Research Question 2: “How does age moderate citizens’ electoral support when being presented with
‘moral panic’ rhetoric, and how does respondents’ lifetime experience with classic psychedelics influence this relationship?”
See figure 1 for a visualization of the conceptual model of this study.
Visualization of the conceptual model of this study
To answer both research questions and test the hypotheses as formulated above, this study makes use of a 5 × 3 × 2 (moral panic × age × lifetime experience with classic psychedelics) quasi- experimental survey experiment. Experiments are known for their strong internal validity and the fact that they allow for a causal relationship to be established (Halperin & Heath, 2020, p.236). The random assignment of respondents to conditions (manipulated or control) allows for a non- representative sample to suffice in an experimental design (p.237). See figure 2 for a visualization of this study’s factorial design. The two moderators of this study are age and lifetime experience with classic psychedelics (LECP), which are both quasi-experimental factors since they cannot be
IV: Moral panic cue
DV: Electoral support Lifetime experience
17 Figure 2
3D design model illustrating the 5 × 3 × 2 factorial design of this study, with the moderators ‘age’ and ‘lifetime experience with classic psychedelics’ (LECP)
manipulated among respondents. The independent variable is the level of moral panic, which is manipulated in true experimental fashion (1 control, 4 manipulated levels). Having five levels in the moral panic factor is ambitious, but the reasoning behind Mokken scaling analysis has been applied on the independent variable to argue that it can be seen as a hierarchical scale.
Theoretically, Goode and Ben-Yehuda argue (2009, p.35−50) that the concept of moral panic is defined by five criteria: concern, hostility, consensus, disproportion, and volatility. Mokken scaling requires a hierarchical scale in items (Watson et al., 2008, p.575). Determining if the concept of moral panic can be interpreted as a hierarchical scale, depends on whether or not the items measuring the concept of moral panic can be argued to be “ordered relative to one another and, by implication, ordered along the latent trait that is being measured” (ibid.). The definitions of the criteria set by Goode and Ben-Yehuda (2009, p.37−40) indicate that the different items in this study – concern, hostility, consensus and disproportion – can indeed be treated as hierarchical
18 items that together measure the extent of the moral panic present14. This means that the manipulation can be interpreted as an interval variable, which allows for the testing of the three- way interaction effect through regression analysis. Using the Mokken scaling reasoning and subsequently regression analysis, means that the necessary sample size is drastically reduced, which increased the feasibility of this study.
The survey experiment and the manipulation were created in English as well as Dutch, so that a manipulation targeted specifically at Dutch citizens could be created15 and a broad (international) sample could be reached without language being a barrier. Two surveys were created (one Dutch, one English), where the Dutch survey was simply translated from English to be completely the same content-wise. Data was collected over a 3-week period in May of 2022. The survey was spread on various social media platforms (Instagram, Reddit, WhatsApp, LinkedIn) and the Amsterdam Psychedelics Research Association forwarded the survey to its members. Both surveys were closed on May 22nd, 2022, with the final number of responses consisting of 313 Dutch and 1317 English-speaking ones.
14 Disproportion, the “sense on the part of many members of the society that a more sizeable number of individuals are engaged in the behavior in question than actually are, and the threat, danger or damage said to be caused by the behavior is far more substantial than, is incommensurate, above and beyond what a realistic appraisal could sustain” (p. 40, emphasis added) could not exist without consensus, the “substantial or widespread agreement or consensus […] that the threat is real, serious and caused by the wrongdoing group members and their behavior” (p. 38, emphasis in original quote). This consensus could not be reached without hostility; the existence of “an increased level of hostility toward the group or category regarded as engaging in the behavior or causing the condition in question” (ibid., emphasis in original quote) and a prerequisite for hostility is the basic feature of concern, the “heightened level of concern over the behavior of a certain group or category and the consequences that that behavior presumably causes for one or more sectors of the society” (p. 37, emphasis in original quote). If disproportion cannot exist without consensus, which stems from hostility which in turn is based on concern, these items together form a hierarchical scale and the reasoning of Mokken Scaling Analysis allows the variable to be interpreted as continuous, which means that an moderation regression analysis can be performed to test the hypotheses.
15 Since this study measured respondents’ electoral support as dependent variable, it was decided to create a manipulation in which a Dutch fictional politician presented his views. If Dutch respondents were presented with an English speaking politician, it is likely that they would have indicated their electoral support to be significantly lower due to it being unrealistic that they would ever vote for a politician that did not speak Dutch.
19 Data cleaning led to a final sample size of 111916. The two datasets (Dutch and English-speaking) were merged in SPSS. Assumptions for linear regression analysis were tested and met17. The final sample (N=1119) consisted of 67,6% respondents with LECP (n=756) and 32,4%
of respondents without LECP (n=363). In this sample, respondents with LECP are overrepresented; research has shown that less than one percent of Americans used LSD in 2018 (Yockey et al. 2020). A percentage of 67,6% is thus a clear overrepresentation. The sample is furthermore skewed to the left. On a scale of 0 (left) to 10 (right), respondents mean political ideology was strongly leftist (Mpolitical ideology = 3.22, SDpolitical ideology = 2.28).
After consenting to participating in this study, respondents were asked some general demographic questions such as age (one of the moderators) gender, location, religious beliefs, education level and political ideology. They were then presented with a set of questions about their psychedelic experience (or lack thereof) – the second moderator in this study. Respondents who indicated to have lifetime experience with classic psychedelics were then asked to elaborate on how often they consumed specific substances (LSD, Psilocybin, Mescaline, DMT). They were then asked to fill in a battery of questions which measured the level of ego dissolution experienced during their psychedelic experience (Nour et al., 2016)18. Respondents without lifetime experience were asked if they ever saw themselves using classic psychedelics19, and all respondents were asked
16 Respondents who did not fill in the question measuring the dependent variable were removed from the sample (n=342), as well as 167 respondents who opened the survey but did not answer any questions. Data was checked for flatliners and outliers, which led to the removal of two respondents who religiously identified as “jedi” (fictional knights in the Star Wars universe).
17 Error terms are statistically independent, have a constant variance and are generally normally distributed. A scatterplot showed that the IV and DV have a linear and additive relationship (Flatt & Jacobs, 2019).
18 Respondents with psychedelic experience were asked about their psychedelic experience using the ‘ego dissolution inventory’ as operationalized by Nour et al. (2016). This scale measures the compromised sense of ‘self’ that is typical for a strong psychedelic trip and includes statements such as “I experienced a disintegration of my ego” (Mego dissolution= 3.79, SDego dissolution= 1.34). Respondents with psychedelic experience were furthermore asked if their experience had a positive or negative impact on their quality of life, ranging from −3 to +3 (Mpost psychedelic quality of life= 1.80, SDpost psychedelic quality of life= 1.13).
19 Respondents without psychedelic experience were asked if they would ever consider consuming psychedelics, ranging from 1 (definitively not) to 5 (definitely yes) (Mpsychedelic possibility= 2.29, SDpsychedelic possibility= 1.34). These factors could potentially predict respondents’ support for the approval of psychedelics for medical use.
20 to indicate their general opinion towards psychedelic, recreational and medical use20. Respondents were then randomly assigned to one of the five conditions21. Each condition was presented to between 219 and 227 respondents22. After being presented with the manipulation, respondents were asked to indicate their opinion on the fictional politician and the level to which they agreed with the approval of classic psychedelics for medical use in their national healthcare system.
Respondents were then asked what the likelihood was of ever voting for the politician. Finally, respondents were debriefed by being explained that the clip they just saw was fake, fictional, and based on misinformation. Factual resources were provided so that respondents could inform themselves correctly.
For this survey experiment, the theoretical concept of moral panic has been operationalized. This has been done based on the literature by Goode and Ben-Yehuda (2009).
Considering criteria five – volatility – is not a communication technique or something that is easily manipulated, it was decided to focus on criteria one through four. These are, in increasing order, concern; hostility; consensus; and disproportionality. Because disproportionality cannot exist without concern, the manipulated campaign ads were created as a cumulative scale. From the manipulations, video one included only concern, video two included concern and hostility and so on, with video four including all the four criteria as set by Goode and Ben-Yehuda. For each criterion, cues were written for a fictional political ad on the potential approval of classic psychedelics for medical use. The most crucial content was used to build up the manipulation. In
20 See table 16 in appendix 6 for an overview of the mean scores.
21 Randomization appeared to be successful, as age (F(4,1114) = .70, p = .590), education level (F(4,1114) = 1.56, p = .182), psychedelic attitude (F(4,1114) = .10, p = .979), political ideology (F(4,1114) = .11, p = .979), Dutch/English survey (Chi^2 (4) = .44, p = .979), gender (Chi^2 (12) = 16.41, p = .173), lifetime experience with classic psychedelics (Chi^2 (4) = 2.25, p = .689), race (Chi^2 (12) = 7.42, p = .829) and religion (Chi^2 (8) = 11.19, p = .191) did not significantly differ across condition. The variables race and religion were recoded to be able to get reliable Chi^2 results that could be interpreted by forming four (race) and three (religion) broad categories which contained enough respondents.
22 ncontrol/absent = 219, nconcern = 224, nhostility = 219, nconsensus = 227, ndisproportionality = 223.
21 the other videos, there was more elaboration on the crucial cue. The crucial cues of the criteria were as follows:
CONCERN: It is expected that within 5 to 7 years, psychedelics can be used in our national healthcare system. But we know too little about the effects. In the United States, federally funded research has only been conducted since 2021.
HOSTILITY: We should not want a new drug such as this one in our national healthcare system. Administering psychedelic drugs, which make you hallucinate, to people who can be mentally vulnerable, is simply dangerous. History teaches us, after all, that taking psychedelics motivates people to oppose the status quo.
CONSENSUS: Polls have shown that substantial amount of voters share my concerns and have their own doubts about this.
DISPROPORTIONALITY: Between 17 to 20% of youth admits to having used psychedelics recreationally, and a substantial share of this group becomes addicted. Per year, dozens of people die because of psychedelics: hallucinations that cause accidents, or are so horrifying that people decide to end their own life23.
A control clip was created which was absent of moral panic cues. This clip included only general information about psychedelics, and was neutral in tone. The clip was aimed to be neutral by using sentences such as “psychedelics are used recreationally, but recently published research indicates that there could potentially be benefits to using psychedelics in mental healthcare. Some
23 These numbers and claims are completely fictional and false. Research from the Drug Policy Alliance shows that only 0.1% of U.S. citizens across all ages are current users of psychedelics (Drug Policy Alliance, n.d.). Furthermore, psychedelics are non-addictive and the body quickly builds a tolerance, meaning that repeated usage becomes increasingly difficult (Drug Policy Alliance, n.d.a). Additionally, psychedelics are non-toxic and “continuing today, sensationalized media coverage of psychedelic-related deaths misattributed the role of psychedelics like psilocybin in causing suicide or accidental death” (Drug Policy Alliance, n.d.b). Research has actually shown an opposite effect of classic psychedelics such as psilocybin: users show lower rates of suicide or depression than people without psychedelic experiences (Drug Policy Alliance, n.d.b).
22 scientists say the results are promising, but others have their doubts,” and called for a broad societal discussion on the topic. See appendix 1 for the full script and link to the videos.
To determine whether the manipulation was perceived as intended by the respondents in this study, a manipulation check was conducted. Manipulation checks can indicate the strength of a study’s construct validity (Sigall & Mills, 1998, p.218-226); the extent to which an operationalization actually constructs the theoretical concept it is intended to measure (Halperin & Heath, 2020, p.310). The manipulation check consisted of two items. The distinction between question one and two was made to create a more nuanced manipulation check; people who were presented with the concern video could have just as easily said that Smith was “extremely concerned” as people who were presented with the disproportionality clip. The second manipulation check should yield a more nuanced distinction between the different manipulation conditions. Respondents could indicate their opinion on a 7-point Likert scale for the following statements:
1. To what extent was Robert Smith concerned about the potential use of classic psychedelics in the national healthcare system?
2. To what extent was what Robert Smith said cause for panic?
Manipulation checks were conducted by method of independent sample t-tests (N=1119), and post hoc Bonferroni tests were not statistically significant. See table 1 (concern) and 2 (panic).
While respondents significantly perceived the control clip to be different in amount of concern and panic compared to the manipulations, they did not perceive a significant difference among each condition. The only exceptions here are the significant difference between the concern and disproportionality condition (p=.041) and between the consensus and disproportionality condition (p=.003) when it comes to the amount of perceived panic. Looking at the means of each condition
23 in the concern manipulation check question (table 3), a gradual increase can be seen. While the conditions did not all differ significantly between one another, the perceived level of concern did increase among respondents between conditions and the ‘panic’ manipulation check shows that generally, people understood the moral panic to increase steadily except for the ‘consensus’
condition. A (semi) failed manipulation check can indicate a low construct validity, which, in this case, is explicable by the fact that this is an initial attempt at operationalizing the theoretical concept of moral panic. The fact that the manipulation check is significant for each condition when compared to the control variable, indicates that respondents did perceive a meaningful difference between neutral rhetoric and traditional moral panic rhetoric. This inspires the re-running of the main analyses using a binary independent variable (neutral vs traditional moral panic rhetoric), the results of which can be found in the appendices of this paper.
Post-hoc Bonferroni results for concern manipulation check, mean differences
Condition Absent Concern Hostility Consensus Disproportionality Absent
— −1.00*** −1.06*** −1.08*** −1.09***
(.13) (.13) (.13) (.13)
— −.06 −.08 −.09
(.13) (.13) (.13) (.13)
Hostility 1.06*** .06
— −.01 −.03
(.13) (.13) (.13) (.13)
Consensus 1.08*** .08 .01
(.13) (.13) (.13) (.13)
Disproportionality 1.09*** .09 .03 .01
(.13) (.13) (.13) (.13)
Note: *p < 0.05, **p < 0.01, ***p < 0.001.
24 Table 2
Post-hoc Bonferroni results for panic manipulation check, mean differences
Condition Absent Concern Hostility Consensus Disproportionality Absent
— −1.04*** −1.22*** −.91*** −1.50***
(.16) (.16) (.16) (.16)
— −.18 .12 −.46*
(.16) (.16) (.16) (.16)
Hostility 1.22*** .18
— .30 −.29
(.16) (.16) (.16) (.16)
Consensus .91*** −.12 −.30
(.16) (.16) (.16) (.16)
Disproportionality 1.50*** .46* .29 .59**
(.16) (.16) (.16) (.16)
Note: *p < 0.05, **p < 0.01, ***p < 0.001.
Descriptive statistics for manipulation check based on condition
Concern manipulation check Panic manipulation check
Mean SE Mean SE
Control/absent 4.50 .09 1.83 .12
Concern 5.50 .09 2.87 .11
Hostility 5.56 .09 3.05 .11
Consensus 5.58 .09 2.74 .11
Disproportionality 5.59 .09 3.33 .11
25 Construction of variables
Dependent variable – electoral support
To measure respondents’ levels of electoral support for the fictional politician Robert Smith (English) or Robert Smits (Dutch), respondents were asked: “How likely are you, given the chance, to ever vote for Robert Smith?”. They could indicate the likelihood on a 1−10 scale, with 0 meaning
“extremely unlikely” and 10 meaning “extremely likely” (Melectoral support=3.30, SDelectoral support=2.59).
Moderators – lifetime experience with classic psychedelics & age
To measure respondents’ lifetime experience with classic psychedelics, they were asked:
“during your lifetime, have you ever consumed any of these [classic psychedelic] substances?”.
Answer categories were 1) No, never (n=363); 2) Yes, only once (n=60); 3) Yes, occasionally (n=435); and 4) Yes, frequently (n=261). Respondents who answered that they consumed classic psychedelics at least once, were coded as having LECP (n=756). Respondents who answered “No, never”, were coded as having no LECP (n=363).
The second moderator in this study is ‘age’. Respondents were asked to fill in their year of birth in the beginning of the survey. See appendix 2 for the full questionnaire in English. Based on the theoretical framework, three age groups have been created and age has been recoded into a categorical variable. Of the full sample, 7.1% of respondents were categorized as ‘moral panic’ age group, 54.4% as ‘moral panic legacy’ age group and 38.4% as ‘psychedelic renaissance’ age group.
See figure 3 for a visualization of the age groups.
26 Figure 3
Visualization of different age group categories within the sample
Moral panic Moral panic legacy Psychedelic renaissance
1932 1961 1996 2004
n=80 n=555 n=484
Results and analyses
The effect of (no) lifetime experience To test H1a24
, a multiple regression analysis was used to predict respondents’ electoral support from traditional moral panic rhetoric, moderated by respondents’ LECP. The theoretical groundwork of different criteria laid by Goode and Ben-Yehuda (2009) allow for the hierarchical (scale) interpretation of the moral panic variable. The predicting of continuous independent variables’
effect on continuous dependent variables is done through regression (Sarstedt & Mooi, 2019). To model the moderating effect of LECP, an interaction term was added to the simple moral panic model (see model 1 in table 4) by using Hayes (2017) PROCESS Macro in SPSS, which probes for conditional effects. These coefficient estimates in multiple regression models are conditional on the inclusion and values of all other variables (Denters & Van Puijenbroek, 1989), meaning that the effect of traditional moral panic rhetoric on electoral support changes based on each value of the moderator variable (LECP no/yes).
24 Hypothesis 1a (H1a): For those who have lifetime experience with classic psychedelics, being presented with the moral panic cue will lead to lower electoral support than not being presented with the moral panic cue.
27 Table 4
Output of multivariate regression analysis, with and without LECP as a moderator
Model 1: Moral panic cue Model 2: LECP
F (1, 1109) = 203.94*** F (3, 1107) = 75.02***
(Constant) 5.950*** 4.539***
Moral panic cue (1−5) −.725*** −.418***
LECP (no/yes) 1.407***
LECP*Moral Panic cue −.457***
R2 .155 .169
Note: *p < 0.05, **p < 0.01, ***p < 0.001.
Respondents’ LECP as a moderator was a significant predictor of electoral support25, t(1107)=−4.26, p=.000. The conditional effects as shown in table 5 indicate that for respondents with LECP, an increase of one step on the hierarchy of moral panic corresponds to a decrease in electoral support of .87, b=−0.87, SE=0.06, p=.000, 95%CI [−0.99, −0.75]. For respondents without LECP, an increase of one step on the hierarchy of moral panic corresponds to an average decrease in electoral support of .42 points, b=−0.42, SE=0.09, p=.000, 95%CI [−0.59, −0.24]. The regression coefficient for the interaction term presented in table 4 represents the difference
25 Reporting standardized beta coefficients when using PROCESS is not recommended, according to the PROCESS FAQ page (Hayes, n.d.).
28 between conditional effects (the b as described above) as being ±.46 points, meaning that the effects differ significantly per group, b=−0.46, SE=0.11, p=<.001, 95%CI [−0.67, −0.25].
The statistical significance of these findings means that H1a is confirmed and can be accepted; for those respondents who have LECP, being presented with traditional moral panic rhetoric does indeed lead to lower electoral support than not being presented with traditional moral panic rhetoric. While it was hypothesized in H1b that respondents without LECP would respond positively to being presented with traditional moral panic rhetoric, the conditional effects of the multivariate regression analysis shows an opposing effect. This means that the null-hypothesis has to be accepted and hypothesis H1b must be rejected; for respondents without LECP, being presented with traditional moral panic rhetoric does not lead to increased electoral support, but actually has the opposite effect – a decrease in electoral support. While the manipulation decreased electoral support among both groups, the effect was twice as strong for respondents with LECP.
The interaction effect is visualized in figure 4 based on ANOVA output26.
Conditional effects on electoral support for groups based on lifetime experience
Effect Estimate SE 95% CI p
No lifetime experience with classic psychedelics −.418 .088 −.591 −.245 .000 Lifetime experience with classic psychedelics −.875 .062 −.995 −.755 .000
26 As a robustness check, the same interaction regression was ran using a binary moral panic variable: simply control versus manipulation. See the multivariate regression output, the conditional effects and the visualization of the effect (based on ANOVA) in appendix 1. These findings are directly comparable to output using the hierarchical IV (nonbinary).
29 Figure 4
Visualization of group mean scores, for respondents with and without LECP (based on ANOVA output)
The effect of age
To understand how age influences respondents’ response to traditional moral panic rhetoric, a multiple regression was run using the Hayes PROCESS Macro in SPSS (2017), see table 6. This multivariate regression analyses allows for comparisons between the oldest group (socialized during the moral panic) and youngest group (socialized during the psychedelic renaissance), as well as the oldest group and middle group (socialized during the moral panic’s legacy). On average, compared to the moral panic group, respondents in the legacy group scored 1.7 points higher on electoral support when presented with the neutral rhetoric, b=1.70, SE=0.67, p=.108, 95%CI [0.39, 3.01].
Compared to the moral panic group, an increase of one step on the hierarchy of moral panic led respondents in the legacy group to decrease their electoral support by 0.69 points, b=−0.69, SE=0.21, p=.001, 95%CI [−1.10, −.29]. When presented with the control condition, younger