Key Takeaways
- Surveys that ask about unconventional beliefs often capture a mix of sincere respondents, jokesters, and deliberate trolls.
- In a New Zealand online sample, 8.3 % admitted to answering insincerely, while 7.2 % endorsed a blatantly false “giant raccoon army” conspiracy.
- When insincere responders were removed, belief in more plausible theories (e.g., 5G‑COVID links) fell by roughly half, showing how trolling inflates prevalence estimates.
- People who endorse contradictory conspiracy theories are far more likely to be responding jokingly or dishonestly, suggesting genuine contradictory belief is rarer than earlier work implied.
- Core relationships found in conspiracy‑theory research (e.g., links to perceived danger) remain robust after filtering out insincere answers.
- Researchers should routinely include checks for sincerity and acknowledge the “Lizardman constant” – the baseline noise from joking or trolling – when interpreting survey data on unusual beliefs.
Introduction and the Lizardman Constant
A 2013 public‑opinion poll claimed that roughly four percent of Americans – about twelve million people – believe that “lizard people” secretly run the planet. Psychiatrist Alexander Scott seized on this striking figure to illustrate a broader methodological problem: even when a question concerns an outlandish idea, a non‑trivial share of respondents may answer insincerely, either as a joke, to provoke, or due to cognitive shortcuts. Scott coined the term “Lizardman constant” to denote the inevitable background noise of jokesters, trolls, and mischievous answering that can obscure the true signal of genuine belief. He warned that any source of noise – whether deliberate misbehaviour, humor, or bias – can easily drown out the substantive responses researchers seek. The present study builds on Scott’s insight by directly measuring how much insincere responding contaminates surveys of conspiracy theories and what happens to prevalence estimates when that noise is removed.
Methodology: Testing Insincere Responses
To quantify the Lizardman constant in a real‑world setting, we adapted an approach first used in Australian research and surveyed a representative online panel of 810 New Zealanders. The questionnaire contained two complementary checks for insincerity. First, at the end of the survey we asked a direct yes/no question: “Did you respond insincerely at any earlier point in this survey? In other words, did you give any responses that were actually just joking, trolling, or otherwise not indicating what you really think?” This self‑report provided a transparent metric of admitted dishonesty. Second, we embedded an obviously fictitious conspiracy theory – the claim that the Canadian Armed Forces are secretly developing an elite army of genetically engineered, super‑intelligent, giant raccoons to invade neighbouring countries – and recorded how many participants endorsed it as “probably” or “definitely true.” Because the premise is biologically implausible and lacks any credible evidence, we reasoned that virtually any affirmation of this statement reflected joking or trolling rather than sincere belief. By comparing the two measures and examining their overlap, we could gauge the extent of non‑serious responding and its impact on other conspiracy‑theory items embedded in the same questionnaire.
Results: Prevalence of Insincere and Belief Responses
In our sample, 8.3 % of respondents (≈ 67 people) confessed to having answered insincerely at some point. A slightly smaller proportion, 7.2 % (≈ 58 people), endorsed the giant‑raccoon conspiracy as probably or definitely true. Importantly, there was notable overlap: 13.3 % of the total sample (≈ 108 people) fell into at least one of these two categories, meaning roughly one in eight participants did not appear to be taking the survey seriously. This figure aligns closely with earlier Australian findings, suggesting that a stable baseline of insincere responding exists across Western populations. The presence of these responders matters because they also tended to endorse other, more plausible conspiracy theories, thereby inflating overall belief rates if left unaddressed.
Impact on Conspiracy Theory Estimates
To illustrate the distorting effect, we examined endorsement of the claim that governments worldwide are covering up the fact that 5G mobile networks spread coronavirus. In the full sample, 6.5 % agreed with this statement. After excluding the 8.3 % who admitted insincere responding, the proportion dropped to 2.7 % – a reduction of more than half. Similar patterns emerged across thirteen different conspiracy theories: each estimated believer percentage declined substantially once the insincere respondents were removed. The magnitude of the shift varied by theory, but the direction was consistent: trolling and joking responses systematically elevate apparent belief levels. This demonstrates that even modest levels of non‑serious answering can produce large overestimates when the base rate of genuine belief is low, as is typical for many fringe claims.
Contradictory Beliefs and Insincerity
An additional layer of insight emerged when we looked at participants who endorsed mutually contradictory conspiracy theories – for instance, asserting both that COVID‑19 is a myth and that 5G networks spread the virus. Historically, such pattern‑matching has been taken as evidence of a fragmented or paranoid worldview. In our data, however, nearly three‑quarters of those who held contradictory views also exhibited signs of insincere responding (either by admitting to joking answers or by endorsing the raccoon army myth). This suggests that genuine belief in directly opposing conspiracies may be far less common than earlier surveys implied; instead, many apparent contradictions likely arise from respondents treating the survey as a playground for absurdity rather than a sincere expression of belief.
Genuine Believers Persist
Despite the substantial noise uncovered, our findings do not invalidate the existence of sincere conspiracy‑theory adherents. When we removed insincere respondents, well‑established correlations from the literature remained largely unchanged. For example, the link between conspiracy endorsement and perceiving the world as a dangerous, threatening place persisted, and in some cases grew slightly stronger after cleaning the data. This indicates that the core psychological and social drivers identified in prior research – such as feelings of powerlessness, mistrust in institutions, or a need for explanatory simplicity – are not artefacts of trolling but characterize a real subset of believers. Moreover, a notable minority continued to endorse theories like the flat‑Earth cover‑up even after exclusions, underscoring that genuine, albeit potentially harmless, conviction exists alongside the more frivolous responses.
Implications for Research and Future Surveys
The study carries practical recommendations for anyone measuring unconventional beliefs. First, surveys should incorporate explicit sincerity checks – either self‑report items like the one used here or indirect measures such as absurd “trap” questions. Second, analysts ought to report results both with and without the flagged insincere cases, allowing readers to gauge the sensitivity of conclusions to the Lizardman constant. Third, researchers interpreting prevalence figures must acknowledge that a certain baseline of joking or trolling is inevitable, especially when querying topics that lend themselves to humor or satire. Finally, while correcting for insincerity sharpens estimates of true belief, it is equally vital to retain the broader explanatory focus of conspiracy‑theory scholarship: understanding why people adopt these ideas, how they spread, and what real‑world consequences they may produce.
Conclusion: Balancing Skepticism and Serious Inquiry
The phenomenon of the “Lizardman constant” reminds us that survey data on bizarre claims are always a blend of signal and noise. In our New Zealand sample, roughly one in eight participants failed to treat the questionnaire with earnestness, and their presence inflated belief estimates across a range of conspiracy theories by up to 50 %. Yet, after accounting for this noise, robust relationships between conspiracy endorsement and underlying psychological motives endured, confirming that a genuine subset of the population does hold these views. Good‑quality research must therefore navigate both dimensions: devising methods to filter out jokesters and trolls while preserving the substantive inquiry into why sincere believers fall into conspiracy‑laden rabbit holes and how those beliefs shape attitudes and actions. By acknowledging and measuring the Lizardman constant, scholars can separate the amusement from the alarm, yielding clearer insights into both the fringe and the faithful.