About Us:  Bryan Nelson and Andre Nakkab are both seniors at New York University studying psychology. Ward Pettibone is also a senior at NYU studying neural science.


Abstract: The False Consensus Effect is an overestimate of the prevalence of one’s own behaviors and beliefs in a population. Previous research on the False Consensus Effect has shown that it is almost universal to the human condition, whether it involves underage drinking, video game use, etc. Whereas the false consensus effect has also been studied in a general risk behavior and risk-taking framework, there have been no studies on this effect with regard to steroid use and bodybuilding specifically. Here, we investigate the false consensus effect in a sample of 441 male bodybuilders that we recruited in online fitness fora. We divided participants based on whether they had ever used anabolic steroids as well as if they had taken fat burners. We asked them what percentage of frequent exercisers they believed used steroids, whether or not they supported the legalization of steroids in the United States, and what age they believed bodybuilders should first consider steroid use. Participants that had taken anabolic steroids reported higher estimates of steroid use prevalence, were more likely to support steroid legalization, and believed athletes should consider taking steroids at a younger age than nonusers. Similarly, fat burner users gave higher prevalence estimates of steroid use and were more likely to support steroid legalization compared to fat burner nonusers. We interpret these relationships as evidence that the false consensus effect exists in anabolic steroid users and that there is a halo-effect in the general bodybuilding community regarding steroid use.




16 thoughts on “

  1. Greetings! Thank you for sharing your research with Sigma Xi! My questions for this research are:

    1. Due to legal issues surrounding the use of AAS, and considering that information about AAS use was gathered using self-report, do you think that your reported number of AAS users might underestimate the total number actually represented in your sample?

    2. How was the recruiting performed? What did survey participants know about the study that they were in? They almost certainly were not informed about the focus on the FCE, correct?

    3. In the slides, do the error bars refer to standard error of the mean?

    4. Both AAS and fat burner users display evidence of a FCE for the first question (estimated prevalence of AAS users). For the second question, there is a difference (youngest age to first consider using AAS). How would you explain that difference? Is that evidence of a FCE for AAS and not fat burner users, or does that question not directly address the FCE but rather more of policy opinion?

    Thank you again for sharing your research!


    1. Hello, thanks for visiting our website!

      1) I do think that is a possibility. However, we recruited participants from anonymous online fitness forums where they likely already discuss their AAS use, so I think the risk of misclassifying AAS users as AAS nonusers is minimal. Fortunately, that misclassification would bias us toward a null result which may imply that our result is an underestimation of the true difference.

      2) The survey covered three domains: FCE, risk factors for AAS use, and outcomes of AAS use. I posted the following message on reddit: “I am an undergraduate student at NYU doing my senior thesis on the relationship between steroid use and impulsiveness. I’m looking for frequent exercisers (both steroid non-users and users) to take my anonymous survey (most people finish in less than 10 minutes).
      If you choose to participate, you can enter our raffle for one of twenty $50 Amazon gift cards.
      If you would like to participate and do not have any questions, you can go to bit.do/exercisestudy to begin. Thanks!”

      3) Yes, they do. My apologies for not indicating on the slides.

      4) I would not necessarily consider the age to consider AAS use or the question indicating support for legalization to be indicative of the FCE. Rather, these two questions are more policy-related and focus on more AAS approval than AAS prevalence. The difference may indicate that fat burner users perceive similar amounts of AAS users as AAS users do, and similarly are in favor of legalization, but that they perceive higher risk (since they want potential users to wait until they are older). In a follow-up, it would be worthwhile to ask about perceived risk of AAS use.

      Thanks again!


  2. Thank you for the replies! Your reasoning regarding the risk of misclassifying AAS users as nonusers makes sense. Thanks also for the further details about your methods, and for clarifying how the second question relates to the FCE. It’s very interesting research!


  3. Congratulations on very interesting research!
    My questions are:
    1) Slide 12. What is 100%? 94% and 84%? Please read this picture by words.
    2) Slide 11. (Considering your replies to the previous questions). If AAS users perceive higher risk – why you do not consider the age to consider AAS use to be indicative of the FCE? How about halo-effect?
    3) You mentioned: “The survey covered three domains: FCE, risk factors for AAS use, and outcomes of AAS use.” No results regarding second and third ones within your presentation?
    Thank you for sharing your research results!


    1. Hi and thanks for the comment!

      1) 100% would imply that everyone supports AAS being legal. So slide 12 shows that 94% of AAS users support legalization compared to 84% of AAS nonusers. Similarly, 94% of fat burner users support AAS legalization compared to 82% of fat burner nonusers. Lastly, 91.5% of fat burner users with no history of AAS support AAS legalization compared to 81% of fat burner nonusers with no history of AAS use.

      2) The FCE conventionally refers to prevalence estimates, so I am hesitant to add to an existing definition. I think age is indicative of perception but not necessarily related to perceived commonality of AAS use. As for the halo-effect, I would say that the fat burner prevalence estimates indicate that someone does not need to use AAS to have the FCE about AAS; just being exposed to AAS users can lead to FCE.

      3) Unfortunately, we don’t have all the results analyzed yet.

      Thanks again!


      1. Thank you for the replies and explanations!
        In fact, slide 12 shows very important result of your research. But it is not clear without you. There are no presented “non-users”. There are only “users”. Normally “no” and “yes” at the upper corner could be understood as a kind of reply.
        In case you are sure that the age to consider AAS use is not indicative of the FCE – what for this index is analyzed and presented ? In case you could assume that this index could be indicative with the negative score (do not changing an existing definition), perhaps, in future it would be worth to reflect this very much important result within the discussion ?
        Thank you again for sharing of your interesting, actual and important research results!


      2. re: quest

        I think the age to consider using AAS is indicative of perceived risk (since some want potential users to wait until they are older). The three questions — estimated percentage of AAS users, age potential user should consider starting AAS, and if they support AAS legalization — combine to give a broader idea of how steroids are perceived. In the paper, I will be sure to take your suggestion and highlight this in my discussion. I agree that each result needs consideration and discussion.

        Thanks again!


  4. Interesting research and also good replies to previous questions. I am interested in your reply to Quest: did you have any conclusions regarding risk factors for AAS use and outcomes of use (other than halo effect and false consensus effect)?
    Also, you mentioned that in your recruiting, you told potential participants that you were interested in steroid use and impulsiveness. Did you study impulsiveness?


    1. Hello and thanks,

      We unfortunately are not finished analyzing the other results yet, so we will hopefully have conclusions about risk factors ASAP. We did measure impulsiveness as a risk factor. Interestingly, many commenters on the posts mentioned how impulsive they were and expressed interest in knowing how the results come out.


  5. Thanks for sharing this interesting data. Do you have any thoughts on to what degree the assumptions are based on the bodybuilding culture as a stand-alone, or whether there is a stereotype among non-bodybuilders about steroid use that the bodybuilders internalize as true?


    1. That’s a very interesting thought. I think the relationship is likely a dynamic one. Bodybuilding tends to foster a hyper-masculine culture, which is certainly aided by AAS use. (After all, synthetic testosterone would by definition make you hyper-masculine since you have unnatural levels of male hormone in your system). At a certain point, non-bodybuilders form a stereotype about bodybuilders and AAS users, who have to act a certain way to live up to the expectation that they are hyper-masculine. This social conformity may also be driven by the FCE; if you assume everyone is taking AAS even if they aren’t, you will likely follow suit.

      I will definitely have to ask participants about this during a follow-up.



  6. Since 150 of 441 (34%) bodybuilders were steroid users, didn’t they actually under estimate the prevalence of their behavior in the population? How “false” is their consensus?


    1. Hi, we selectively sampled subreddits with a lot of steroid users, so the 34% is due to us purposefully finding users. The 34% is not a prevalence estimate of steroid users in the general population or even the fitness community, but rather a direct result of us recruiting from steroid using communities to find users.

      (Also, the actual root of the FCE is just that users and nonusers differ in their estimates. It doesn’t actually matter if either group is right, if this had been a representative prevalence estimate)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s